RF_COM_H_47_SUDHA_PART_2.pdf
Media
- extracted text
-
RF_COM_H_48_SUDHA_PART_2.
NOT FOR DISSEMINATION
DRAFT 1
WORLD REPORT ON VIOLENCE AND HEALTH
WORLD HEALTH ORGANIZATION
GENEVA
2001
Dear Reader:
Please note that this draft document is a work in progress, and many of its components are still
under development. As such, this document should not be cited, quoted or distributed to anyone.
Also note that this is a first attempt to assemble the chapters and other parts of the document.
Although most components of the Report are included herein, some of them are still in
preparation. The Table of Contents of this document indicates which parts of the document are
still being prepared. Please also note that the chapters are at different stages of development;
while some have undergone several revisions and have been peer reviewed, others are still at
earlier stages of development. The status of each chapter is indicated on the title sheet for each
chapter.
It is our plan to include testimonies of victims and perpetrators of violence in separate boxes in
the text of the chapters. For now, a table describing various samples of testimonies may be found
in Appendix D. We would like to draw your attention to these because some violence-related
issues will be addressed only through these testimonies. The testimonies represent a broad
geographic scope and as such will add national and regional victim and perpetrator perspectives
to the Report.
With regard to the tables, figures and boxes, please note that all are not included in this first draft
of the Report. Those which are included are placed either within the narrative text of the chapter
or at the end of the chapter before the list of references. Their placement will be standardized in
future drafts ofthe Report.
The expected date ofpublication of the Report is September 2001.
Violence and Injury Prevention Department
World Health Organization
October 2000
ii
NMH/VIP/WVRDRAFT.00.1
Distribution: Limited
Original: English
WORLD REPORT ON VIOLENCE AND HEALTH
Violence and Injury Prevention Department
Non-communicable Diseases and Mental Health Cluster
World Health Organization
Geneva
iii
NMH/VIP/WVRDRAFT.00.1
Distribution: Limited
Original: English
Copies of this document are available from:
Violence and Injury Prevention Department
World Health Organization
20 Avenue Appia
1211 Geneva 27
Switzerland
Fax: 0041 22 791 4332
Email: pvi@who.int
© World Health Organization, 2001
This document is not issued to the general public, and all rights are reserved by the World Health
Organization (WHO). The document may not be reviewed, abstracted, quoted, reproduced or
translated, in part or in whole, without the prior written permission of WHO. No part of this
document may be stored in a retrieval system or transmitted in any form or by any means electronic, mechanical or other - without the prior written permissions of WHO.
The views expressed in documents by named authors are solely the responsibility of those authors.
iv
Preface
To be drafted by a prominent world leader who has contributed to the prevention of
violence world-wide.
(N. Mandela?)
V
Foreword
To be drafted by Dr. Gro Harlem Brundtland, Director General, WHO
vi
Contributors
Editorial Committee
Etienne Krug. World Health Organization
Linda Dahlberg, US Centers for Disease Control and Prevention
James Mercy, US Centers for Disease Control and Prevention
Rafael Lozano, World Health Organization
Anthony Zwi, London School of Hygiene and Tropical Medicine
Advisory Committee
Nana Apt, University of Ghana
Philippe Biberson, Medecins sans Frontieres
Jacquelyn Campbell, Johns Hopkins University, USA
Radhika Coomaraswamy, International Center for Ethnic Studies, Sri Lanka
William Foege, Emory University, USA
Adam Graycar, Australian Institute of Criminology, Australia
Rodrigo Guerrero, Fundacion Carvajal, Colombia
Marianne Kastrup, World Psychiatric Association, Denmark
Reginald Moreels, former Belgian Minister of Development Cooperation, Belgium
Paulo Sergio Pinheiro, University of Sao Paolo, Brazil
Mark Rosenberg, Task Force for Child Survival, Carter Center. USA
Terezinha da Silva, University Eduardo Mondlane, Mozambique
Datuk Mohd Sham Kasim, International Society for Prevention of Child Abuse and Neglect
Malaysia
WHO Secretariat
Tomris Turmen, Special Advisor to the Director General
Derek Yach, Executive Director, Non communicable Diseases and Mental Health
Yasuhiro Suzuki, Executive Director, Health Technology and Pharaceuticals
Julio Frenk, Executive Director, Evidence and Information for Policy
Etienne Krug, Violence and Injur}7 Prevention Department
Claude Romer, Violence and Injury Prevention Department
vii
Chapters
Introduction
Lead Authors:
Authors Boxes:
Youth Violence
Lead Author:
Other Authors:
Authors Boxes:
Peer Reviewers:
Linda Dahlberg, US Centers for Disease Control and Prevention
Etienne Krug, World Health Organization
Box 1. Violence and Inequality, Lucie Laflamme/Karin Engstrom, Sweden
Box 2. Case study: Cali, Colombia, Rodrigo Guerrero/Alberto Concha
Eastman, Colombia
Box 3. Violence and the Media (To be completed)
Box 4. Police violence (To be completed)
James Mercy, US Centers for Disease Control and Prevention
Alexander Butchart, Karolinska Institutet (at time of writing) and
Unisa Institute for Social and Health Sciences, South Africa
David Farrington, Institute of Criminology, University of Cambridge, UK
Magdalena Cerda, Uruguay/World Health Organization
Linda Dahlberg, US Centers for Disease Control and Prevention
Box 1. Parenting Training in South Africa (To be completed)
Box 2. Trends in Youth Violence in Eastern Europe (To be completed)
Box 3. Firearms (To be completed)
Nancy Cardia, University of Sao Paolo, Brazil
Alberto Concha Eastman, Colombia/Pan American Health Organization
Adam Graycar, Australian Institute of Criminology, Australia
Kenneth Powell, Georgia Division of Public Health, USA
Garth Stevens, Unisa Institute for Social and Health Sciences, South Africa
Mohamed Seedat, Unisa Institute for Social and Health Sciences, South Africa
Child Maltreatment
Lead Authors:
Corrine Wattam, University of Central Lancashire, UK
Desmond Runyan, University of North Carolina, USA
Other Authors:
Fatma Hassan, Suez Canal University, Egypt
Dipty Jain, Government Medical College, Nagpur, India
Laurie Ramiro, College of Medicine, Manila, Philippines
Beatriz Viscarra, Universidad de la Frontera, Chile
Authors Boxes:
Box 1. Implementation of the UN Convention on the Rights of the Child in
Relation to Child Maltreatment, Marcus Stahlhofer, World Health
Organization
Box 2. Learning from People with Direct Experience of Child Maltreatment,
Corrine Wattam, UK
Box 3. Infanticide in Hungary, Maria Herczog, Hungary
Box 4. Zimbabwe: Insights into a Comprehensive and Effective Response,
Naira Khan, Zimbabwe
viii
Peer Reviewers:
Box 5. Training for Health Professionals in Germany, Reiner Frank/Maru
Kopecky-Wenzel, Germany
Box 6. Early Intervention to Prevent Child Abuse in the United States, Anne
Cohn Donnelly/Deborah Daro, USA
Box 7. Corporal Punishment: State-authorized Violence to Children, Peter
Newell, UK/Akila Belembaogo, Burkina Faso/UNICEF
Tilman Fumiss, University Hospital of Munster, Germany
Fu Yong Jiao, Shaanxi Provincial People's Hospital, China
Philista Onyango, African Network for the Prevention and Protection Against
Child Abuse and Neglect, Kenya
Zelidad Alma de Ruiz, Family Institute of Santo Domingo, Dominican
Republic
Violence against Women by Intimate Partners
Lori Heise, Center for Health and Gender Equity, USA
Lead Author:
Claudia Garcia-Moreno, Mexico/World Health Organization
Other Authors:
Box 1. Enhancing Comparability of Data on Domestic Violence, Mary
Authors Boxes:
Ellsberg, USA/Nicaragua
Box 2. "La Ruta Critica:" An Evaluation of Institutional Responses to
Domestic Violence, Pan American Health Organization
Box 3. Examples of Innovative Prevention Programs
Box 4. How Women Experience Violence: Women's Voices from North and
South
Peer Reviewers: Jill Astbury, University of Melbourne, Australia
Jacquelyn Campbell, Johns Hopkins University School of Nursing, USA
Radhika Coomaraswamy, International Center for Ethnic Studies, Sri Lanka
Terezinha da Silva, University of Eduardo Mondlane, Mozambique
ix
Elder Abuse
Lead Author:
Other Authors:
Authors Boxes:
Peer Reviewers:
Lia Daichman, International Network for the Prevention of Elder Abuse,
Argentina
Rosalie Woolf, International Network for the Prevention of Elder Abuse, USA
Gerry Bennett, International Network for the Prevention of Elder Abuse, UK
Box 1. Permanent Forum Against Abuse of the Older Person, N. Gilliland/ LE
Picado, Costa Rica
Box 2. Emotional Abuse, Deprivation of Freedom and Financial Exploitation
(Argentina)
Box 3. Elder Abuse Prevention Centre Help Line Service, Y. Yamada, Japan
Box 4. Influencing the Future: An Intergenerational Curriculum on Elder
Abuse, Elizabeth Podnieks, Canada
Box 5. Witchcraft: A Violent Threat, HelpAge International
Nana Apt, University of Ghana
Robert Agyarko, World Health Organization
Malgorzata Halicka, University of Bialystok, Poland
Jordan Kosberg, The University of Alabama, USA
Yui-huen Kwan, City University of Hong Kong, China
Siobhan Laird, University of Ghana
Ariela Lowenstein, Haifa University, Israel
Sexual Violence
Puma Sen, CHANGE, UK
Lead Author:
Peer Reviewers: Chapter not yet reviewed
Self-directed Violence
Diego DeLeo, Australian Institute for Suicide Research and Prevention
Lead Author:
Jose Bertolote, Brazil/World Health Organization
Other Authors:
David Lester, USA
Box 1. Suicide in China, Michael R. Phillips, China
Authors Boxes:
Box 2. Suicide Among the Inuit, Antoon Leenaars, Netherlands
Box 3. Aboriginal Suicide in Australia, Ernest Hunter, Australia
Box 4. Depression and Suicide, Danuta Wasserman, Sweden
Box 5. Alcohol and Suicidal Behaviour, Danuta Wasserman, Sweden
Box 6. Suicidal Behaviour in Children and Young People, Annette Beautrais,
New Zealand
Box 7. Access to Means of Suicide, Rachel Jenkins
Peer Reviewers: Chapter not yet reviewed
x
Collective Violence
Anthony Zwi, London School of Hygiene and Tropical Medicine
Lead Author:
Richard Garfield, Columbia University, and
Other Authors:
Alessandro Loretti, Italy/World Health Organization
Authors Boxes: Box 1. Proposed Codes of Conduct for Humanitarian Activity
Box 2. Small Arms and Light Weapons, David Meddings, Canada and Etienne
Krug, World Health Organization
Box 3. Violence against Humanitarian Aid Workers (To be completed)
Box 4. Child Soldiers, Rachel Brett, Switzerland
Box 5. Interethnic Violence in India (To be completed)
Box 6. Genocide (To be completed)
Peer Reviewers: Suliman Baldo, Human Rights Watch/Africa
Robin Coupland, International Committee of the Red Cross
Marianne Kastrup, World Psychiatric Association, Denmark
Arthur Kleinman, Harvard University, USA
Paulo Sergio Pinheiro, University of Sao Paolo, Brazil
Jean Rigal, Medecins sans Frontieres
Michael Toole, Macfarlane Burnet Centre for Medical Research, Australia
Other Contributors
The Editorial Committee would like to acknowledge the following other contributors: Karin
Engstrom, Nynke Poortinga, Gabriella Rosen and Laura Sminkey.
xi
Participants in Regional Consultations
AFRO
Nana Apt, University of Ghana
Colete Delhot, World Health Organization, AFRO
Naira Khan, Training and Research Support Center's Child and Law Project, Zimbabwe
Philista Onyango, African Network for the Prevention and Protection Against Child
Abuse and Neglect, Kenya
Welile Shasha, World Health Organization, South Africa
Safia Singhateh, World Health Organization, AFRO
Derek Yach, World Health Organization, Geneva
Greer van Zyl, World Health Organization, South Africa
AMRO
Nancy Cardia, University of Sao Paolo, Brazil
Arturo Cervantes, Instituto de Investigaciones, Mexico
Alberto Concha-Eastman, World Health Organization, AMRO
Steven Corber, World Health Organization, AMRO
Rodrigo Guerrero, Inter American Coalition for the Prevention of Violence, Colombia
Soledad Larrain, Universidad Catolica de Chile
Daniel Lederman, World Bank
Rodolfo Pena
Mark Rosenberg, Carter Foundation
EMRO
Abdul Rahman Al-Awadi, former Minister of Health and Minister of Planning, Egypt
Sarnia Galal, Ministry of Environment, Egypt
Hani Jahshan, National Institute of Forensic Medicine, Jordan
Ahmed Abdul Latif, World Health Organization, EMRO
Ahmed Mohit, Task Force on Violence and Health, EMRO
Lamis Nasser, Human Forum for Women's Rights, Jordan
Asma Fozia Qureshi, Aga Khan University, Pakistan
Seham Abdel Salam, Documentation Centre for Violence, Egypt
Sima Samar, SHOHADA
Mervat Abu Shabana, Task Force on Violence and Health, EMRO
Fady Shammas, Ministry of Public Health, Lebanon
xii
EURO
Assia Brandrup-Lukanow, World Health Organization
Kevin Browne, University of Birmingham, UK
Gani Demolli, World Health Organization, Kosovo
Maria Herczog, National Institute of Family and Social Policy, Hungary
Kari Killen, Institute of Child Welfare Research, Norway
Viviana Mangiaterra, World Health Orgamzation
Znhal Orhon, CAN Institute, Turkey
Annemiek Rigters, Consultant, Netherlands
Vlad Romano, National Agency for Protection of Child’s Rights, Romania
Berit Shei, Danish Rape Crisis Center
SEARO
Srikala Bharath, NUMHANS, India
Gopalkrishna Gururaj, National Institute of Mental Health and Neuro-Sciences, India
Chumrutai Kanchanachitra, Institute of Population and Social Research, Thailand
Mintasih Latief, Jakarta State Mental Hospital, Indonesia
Panpimol Lotrakul, Ministry of Public Health, Thailand
Sawat Ramaboot, World Health Organization, SEARO
Sanjeeva Ranawera, Presidential Secretariat, Sri Lanka
Prawate Tantipiwatanaskul, Child Mental Health Center, Thailand
WPRO
Liz Eckerman, Deakin University, Australia
An Lin, Beijing Medical University, China
Bernadette Madrid, University of Philippines
Jose Maria Ochave, Center for Science and Technology Law, Philippines
Linda Milan, World Health Organization, WPRO
Mohd Sham Kasim, University of Putra Malaysia
Michael Toole, Macfarlane Burnet Centre for Medical Research, Australia
Financial support for this Report was provided by:
Government of Belgium
Government of Finland
Rockefeller Foundation
US Centers for Disease Control and Prevention
xiii
CONTENTS
I.
Preface (To be completed)
II.
Foreword (To be completed)
III.
Executive Summary (To be completed)
IV.
Why this Report?
A.
Boxes
1. WHA 49.25 Prevention of Violence: A Public Health Priority
V.
Introduction: Violence as a Public Health Problem
A.
Introduction
B.
Definition of Violence
Typology of Violence
C.
D.
How do we Measure Violence, Its Impact and Prevention?
How Big is the Problem of Violence?
E.
How and Why Does Violence Occur?
F.
What Can Be Done to Prevent Violence?
G.
H.
Conclusion
I.
Boxes
1. Violence and Inequality
2. Case study: Cali, Colombia
3. Violence and the Media (To be completed)
4. Police violence (To be completed)
VI.
Youth Violence
Introduction
A.
B.
Definitional Issues
Patterns and Trends in Mortality
C.
D.
Predictors, Causes, and Correlates of Youth Violence
E.
A Typology of Youth Violence Prevention Strategies
F.
The Effectiveness of Youth Violence Prevention Strategies
G.
Summary and Recommendation
H.
Boxes
1. Parenting Training in South Africa (To be completed)
2. Trends in Youth Violence in Eastern Europe (To be completed)
3. Firearms (To be completed)
xiv
VII.
Child Maltreatment
A.
Introduction
B.
International Issues in Defining Maltreatment
C.
Epidemiology
D.
Dynamics of Maltreatment
E.
Etiology
F.
Consequences of Maltreatment
G.
Recognition / Training of Health Professionals
H.
Intervention
Prevention
I.
J.
Policy Responses
Summary
K.
L.
Boxes
1. Implementation of the UN Convention on the Rights of the Child in
Relation to Child Maltreatment
2. Learning from People with Direct Experience of Child Maltreatment
3. Infanticide in Hungary
4. Zimbabwe: Insights into a Comprehensive and Effective Response
5. Training for Health Professionals in Germany
6. Early Intervention to Prevent Child Abuse in the United States
7. Corporal Punishment to Children: State-authorized Violence to Children
VIII.
Violence Against Women by Intimate Partners
Introduction
A.
Nature and Magnitude of the Problem
B.
The Dynamics of Abuse
C.
Women's Response to Abuse
D.
Explaining Intimate Partner Violence
E.
The Consequences of Abuse for Health and Well-Being
F.
Prevention and Policy Response
G.
Principles of Good Practice
H.
Conclusions and Recommendations
I.
Boxes
J.
1. Enhancing Comparability of Data on Domestic Violence
2. "La Ruta Critica:" An Evaluation of Institutional Responses to Domestic
Violence
3. Examples of Innovative Prevention Programs
4. How Women Experience Violence: Women's Voices from North and
South
xv
IX.
Elder Abuse
A.
Introduction
B.
Definition of Elder Abuse
C.
Scope of the Problem
D.
Elder Abuse in Domestic Settings
E.
Elder Abuse in Institutional Settings
F.
Responses to the Problem
G.
Summary and Recommendations
H.
Boxes
1. Permanent Forum Against Abuse of the Older Person (Costa Rica)
2. Emotional Abuse, Deprivation of Freedom and Financial Exploitation
(Argentina)
3. Elder Abuse Prevention Centre Help Line Service (Japan)
4. Influencing the Future: An Intergenerational Curriculum on Elder
Abuse (Canada)
5. Witchcraft: A Violent Threat
X.
Sexual Violence
A.
Introduction
B.
Concepts and Definitions
C.
Forms and Contexts of Sexual Violence
D.
Magnitude of the Problem
E.
Risk and Protective Factors
F.
Responses to Sexual Violence
G.
Recommendations
H.
Conclusion
I.
Boxes
1. Working with Prostituted Women - Maiti Nepal (To be completed)
2. Providing Support to Raped Women - One of Fiji/Ireland/Malaysia
(To be completed)
3. Testimony of a Raped Wife - East Africa (To be completed)
4. Child Sexual Abuse - UK/Europe (To be completed)
5. Sexual Abuse of Women in Prison - Amnesty International / HRW
USA (To be completed)
6. Domestic Workers - Kalayaan/Philippino/South Asian Women in the
Middle East (To be completed)
7. FGM/Child Sexual Abuse/VVF - Forward Project in Nigeria (To be
completed)
xvi
XI.
Self-Directed Violence
A.
Introduction
B.
Definitions
C.
Epidemiology of Suicide Behaviour
D.
Risk Factors for Suicidal Behaviour
E.
Response to the Problem: Prevention, Policy Responses
F.
Recommendations
G.
Boxes
1. Suicide in China
2. Suicide Among the Inuit
3. Aboriginal Suicide in Australia
4. Depression and Suicide
5. Alcohol and Suicidal Behaviour
6. Suicidal Behaviour in Children and Young People
7. Access to Means of Suicide
XII.
Collective Violence
A.
Introduction
B.
Definitions
C.
Data Sources, Trends and Emerging Patterns
D.
Direct and Indirect Health Impact
E.
Contributory Causes to Violent Political Conflict
F.
Responding to Collective Violence
G.
Recommendations
H.
Boxes
1. Proposed Codes of Conduct for Humanitarian Activity
2. Small Arms and Light Weapons
3. Violence against Humanitarian Aid Workers (To be completed)
4. Child Soldiers
5. Interethnic Violence in India (To be completed)
6. Genocide (To be completed)
XIII.
Summary Chapter: Public Health Challenges - What Lies Ahead
XIV. Appendix A: Statistical Annex
XV.
Appendix B: Contact Information for Study Contributors
XVI.
Appendix C: Useful Resources
XVII. Appendix D: Testimonies of Victims and Perpetrators of Violence
xvii
Why this Report?
Status:
Draft: 1st
Date of Current Draft: 15 October 2000
xviii
Violence is a leading public health problem. Each year, more than 2 million people die as the
result of injuries due to violence. Many more survive their injuries but remain permanently
disabled. Among persons 15-44 years of age, interpersonal violence is the 3rd leading cause of
death, war the 6th and suicide the 4th. In addition to injuries and death, violence can result in a
wide variety of other health problems. These include profound mental health consequences,
sexually transmitted diseases, unwanted pregnancies, as well as behavioural problems such as
eating and sleeping disorders.
The World Health Organization as well as other public health agencies has been concerned by the
impact of violence on individuals, communities and their health systems for several years. In
1996, in view of what it described as a dramatic increase in the incidence of intentional injuries,
the Forty-Ninth World Health Assembly adopted resolution WHA 49,25 declaring violence a
leading worldwide public health problem. In the resolution, the WHA recognized the serious
immediate and future long-term implications that violence represents for individuals, families,
communities and countries. The WHA also recognized the growing consequences of violence for
health care services and its detrimental effect on scarce health care resources. The WHA urged
Member States to assess the problem of violence in their own territory. It also requested the
Director-General, within available resources, to initiate public health activities to address the
problem of violence (See box). The WHO Plan of Action to prevent violence followed the WHA
resolution.
Purpose of the Report
This Report is an important step to draw attention to the public health aspects of violence and to
follow up on the WHA Resolution and Plan of Action. The goals of the report are to raise world
wide awareness about the public health aspects of violence and to highlight the contributions of
the public health approach to understanding and responding to violence. More specific objectives
of the document are: 1) to describe the magnitude and impact of violence cross-nationally; 2) to
elucidate cross-national patterns of violence; 3) to provide a baseline for measuring change and
progress; 4) to summarize existing information on risk factors, prevention approaches, and policy
responses; 5) to provide directions for future research; and 6) to make recommendations for
future action in public health. This report, however, is only a beginning. We hope that it will
stimulate discussion at local, national, regional and global levels and be a platform for increased
action towards violence prevention.
The report focuses on aspects of violence that relate to public health. The authors recognise the
importance of other fields, such as the judicial sector, and will to some extent discuss the
relationship and important links with other sectors, but will keep the main focus of the report on
violence as a public health problem.
Not all types of violence are addressed in the Report. Although all types of violence are
important and can have dramatic consequences for victims and their families, the main focus of
this first World Report on Violence and Health is on the types of violence that are present world
wide, in the everyday lives of people, and that constitute the bulk of the burden of violence. The
report covers, for example, violence that occurs in the family and in the community, as well as
violence due to political conflict.
xix
How is the report organized?
The report is organized in two main parts: 1) topic specific chapters, and 2) a data annex. The
introductory chapter of the report provides a broad overview of violence as a public health
problem. Subsequent chapters focus on specific types of violence, beginning with youth
interpersonal violence. The next three chapters focus on violence in the family or close
environment: child abuse, intimate partner violence, and elderly abuse, followed by a chapter on
sexual violence. The next two chapters focus on self-inflicted violence, with a focus on suicidal
behaviour, and collective violence, respectively. The final chapter of the report summarizes
crosscutting patterns, issues, and gaps, and discusses the remaining public health challenges in
the area of violence and ways to address those challenges in different regions of the world. The
chapters are organized using a similar structure, first providing a discussion of definitions and
typology, then a summary of data on fatal and non fatal outcomes of violence, followed by a
discussion on risk and protective factors as well as prevention strategies and policies, and finally
recommendations for future research and public health action.
The chapters contain boxes, in which specific issues are highlighted or case studies presented.
The chapters also contain testimonies from victims or perpetrators illustrating the impact of
violence on their lives and that of their relatives. Because it is beyond the scope of this report to
cover all types of violence fully and adequately, each chapter has a specific focus. For example,
the focus of the chapter on child maltreatment is on abuse against children in the family. Some of
the other forms of child maltreatment, such as female genital mutilation or child prostitution are
addressed in other parts of the Report, while other forms such as child labour or using children as
soldiers are not addressed in the report. The chapter on youth violence is focused on interpersonal
violence among adolescents and young adults in the community. Other forms of violence
involving youth are discussed in other parts of the report. The chapter on elder abuse is focused
on abuse by care givers in domestic and institutional settings and not on all forms of violence
against older persons.
The second part of the Report, the Statistical Annex, contains tables of data. These tables provide
estimates for fatal outcomes of violence for the world and regions for 1999. They also contain
the most recent country specific data as reported to WHO by Member States.
XX
How was the report developed?
This report is the result of work conducted by a broad network of experts from around the globe.
A small Editorial Committee has co-ordinated the process. An Advisory Committee with 13
prominent members from all regions of the world as well as several WHO representatives
provided guidance to the Editorial Committee at several stages during the writing of the Report.
Experts on specific violence-related topics were invited to form multi-cultural groups to write the
chapters and boxes. Each of the chapters was peer reviewed by at least 5 scientists, each from a
different region, who were asked to provide input not only on the scientific content but also on
the relevance of the chapter in their own culture. Consultations were held with members of the
WHO regional offices and diverse groups of experts within each region to add regional and
cultural perspectives, knowledge, and insight for moving the field forward. Stories of victims of
violence illustrating the diverse impact of violence on their own and their family’s life were
collected from grass roots organizations and in some cases directly from victims themselves. A
large amount of data is presented in the chapters and in the statistical annex. References are
provided for the data presented in the chapter. The data presented in the tables is from the WHO
Mortality and Morbidity database which contains data reported annually by WHO member states.
xxi
Box 1
WHA49.25 Prevention of Violence: A Public Health Priority
The Forty-ninth World Health Assembly,
Noting with great concern the dramatic worldwide increase in the incidence of intentional injuries affecting people
of all ages and both sexes, but especially women and children;
Endorsing the call made in the Declaration of the World Summit for Social Development for the introduction and
implementation of specific policies and programmes of public health and social services to prevent violence in
society and mitigate its effect;
Endorsing the recommendations made at the International Conference on Population and Development (Cairo,
1994) and the Fourth World Conference on Women (Beijing, 1995) urgently to tackle the problem of violence
against women and girls and to understand its health consequences;
Recalling the United Nations Declaration on the elimination of violence against women;
Noting the call made by the scientific community in the Melbourne Declaration adopted at the third international
conference on injury prevention and control (1996) for increased international cooperation in ensuring the safety of
the citizens of the world;
Recognising the serious immediate and future long-term implications for health and psychological and social
development that violence represents for individuals, families, communities and countries;
Recognising the growing consequences of violence for health care services everywhere and its detrimental effect on
scarce health care resources for countries and communities;
Recognising that health workers are frequently among the first to see victims of violence, having a unique technical
capacity and benefiting from a special position in the community to help those at risk;
Recognising that WHO, the major agency for coordination of international work in public health, has the
responsibility to provide leadership and guidance to Member States in developing public health programmes to
prevent self-inflicted violence and violence against others,
1. DECLARES that violence is a leading worldwide public health problem;
2. URGES Member States to assess the problem of violence on their own territory and to communicate to WHO
their information about this problem and their approach to it;
3. REQUESTS the Director-General, within available resources, to initiate public health activities to address the
problem of violence that will:
(1) characterize different types of violence, define their magnitude and assess the causes and the public health
consequences of violence using also a "gender perspective" in the analysis;
(2) assess the types and effectiveness of measures and programmes to prevent violence and mitigate its effects,
with particular attention to community-based initiatives;
(3) promote activities to tackle this problem at both international and country level including steps to:
(a) improve the recognition, reporting and management of the consequences of violence;
(b) promote greater intersectoral involvement in the prevention and management of violence;
(c) promote research on violence as a priority for public health research;
(d) prepare and disseminate recommendations for violence prevention programmes in nations, States
and communities all over the world;
(4) ensure the coordinated and active participation of appropriate WHO technical programmes;
(5) strengthen the Organization's collaboration with governments, local authorities and other organizations of
the United Nations system in the planning, implementation and monitoring of programmes of violence
prevention and mitigation;
4. FURTHER REQUESTS the Director-General to present a report to the ninety-ninth session of the Executive
xxii
Board describing the progress made so far and to present a plan of action for progress towards a science-based
public health approach to violence prevention.
Hbk
Res.,
Vol.
Ill
(3rd
ed.),
1.11
(Sixth plenary meeting, 25 May 1996 - Committee B, fourth report)
xxiii
Chapter 1
Violence as a Public Health Problem
Status:
Draft: 1st
Date of Current Draft: 17 October 2000
1
Table of Contents
I.
Introduction
Public Health Approach to Violence
IL
Definition of Violence
m.
Typology of Violence
IV.
How do we Measure Violence, Its Impact and Prevention?
A. Types of Data
B. Sources of Data
C. Availability and Quality of Data
D. Other Considerations
v.
How Big is the Problem of Violence?
A. Estimates of Mortality
B. Estimates of Disability Adjusted Life Years
C. Estimates of Non-fatal Violence
D. Consequences of Violence
VI.
How and Why Does Violence Occur?
A. Multifaceted Nature of Violence
B. Ecological Framework for Understanding Violence
C. Links between Different Types of Violence
D. Research Gaps
VII.
What Can Be Done to Prevent Violence?
A. Cross Cultural Examples of Successful Approaches
B. Lessons Learned from Research on Violence
C. Primary, Secondary and Tertiary Prevention Efforts
D. Need for Multi-sector Responses Across and Within Types of Violence
E. Putting Knowledge into Practice
VIII.
Conclusion
2
Introduction
Over the past few years, we have been exposed almost daily to the terrible images of human
misery caused by deadly conflicts in East Timor, Sierra Leone, Kosovo, Rwanda, and the
Democratic Republic of Congo. The mass graves, mass rapes, and exodus of people are the
most visible part of the iceberg of violence. More discrete, but widespread, is the daily
suffering of children who are abused by their care givers, women victimized by partners,
elderly persons maltreated by care givers, persons who attempt or take their own lives, and
youths who cannot attend school without risk of being threatened, beaten or shot. Public
health is increasingly taking a stand against accepting violence as an inevitable part of the
modem world and is taking actions to prevent it.
The public health approach to violence is interdisciplinary, science-based, and focused on
prevention. Public health draws upon and applies the expertise and body of knowledge from
many fields, including medicine, epidemiology, sociology, psychology, criminology,
education, economics, and other fields. Bringing together the strengths and approaches of
each of these fields allows public health to be innovative and responsive to the wide range
and far-reaching problems of disease, illness, and injury around the globe. Public health
emphasizes collective action and believes that cooperative efforts from such diverse sectors
as health, education, social services, justice, and policy are necessary to solve the problem of
violence. Each sector has an important role to play in addressing the problem of violence,
and, collectively, the approaches taken by each have the potential to produce important
reductions in violence.
The public health approach to violence is based on science. In moving from problem to
solution, the public health approach to violence has four key steps (Figure 1). The prevention
of violence begins with the systematic and ongoing collection of data to describe the
magnitude, scope, and characteristics of violence at local, national, and international levels.
The first step essentially uncovers the “who,” “what,” “when,” “where,” and “how” of
violence. The second step of the public health approach addresses the question of “why”
violence occurs and involves conducting research to determine the causes and correlates of
violence, which factors increase or decrease the risk for violence, and which factors are
potentially modifiable through interventions. The third step is to find out what can be done to
prevent violence by using the information from the previous steps to design, implement, and
evaluate interventions. The fourth step is to implement the most promising interventions in
different settings, to disseminate information broadly, and to determine the cost-effectiveness
of programs.
Public health is also well known for its emphasis on prevention. Rather than simply
accepting or reacting to violence, public health is based on the strong conviction that violent
behaviour and its associated consequences can be prevented. The wide variation in rates of
homicide among nations or within nations over time suggests that violence is the product of
complex, yet modifiable social and environmental factors. Public health both challenges and
seeks to empower people, communities, and nation states to see violence as a problem that
can be understood and solved (1).
In this introductory chapter we use the public health approach to provide an overview of the
problem of violence and how it can be prevented. We begin by describing how violence is
3
defined by public health officials and propose a typology for delineating the form and context
of various types of violence. We then discuss how violence is measured and describe the
global burden of violence. Finally, we provide a framework for understanding how and why
violence occurs and what can be done globally to prevent it. Subsequent chapters in this
report provide more in-depth discussion of definitions of violence, the magnitude and impact
of various types of violence, as well as research, prevention strategies, and policy responses
for interpersonal violence, sexual violence, self-directed and collective violence.
Definition of Violence
Any comprehensive analysis of violence must begin with a definition of violence. For our
purposes, the definition should circumscribe the types of violence of interest in such a way as to
facilitate their scientific measurement. The World Health Organization defines violence as:
The intentional use ofphysical force or power, threatened or actual, against oneself,
another person, or against a group or community that either results in or has a high
likelihood of resulting in injury, death, psychological harm, maldevelopment, or
deprivation. (2)
There are several aspects to the definition that merit further discussion. The definition used by
the World Health Organization associates intentionality with the commission of the act itself,
unconnected to the results it brings. Excluded from the definition are unintentional incidents
(e.g., most traffic injuries and bums). The inclusion of the word power, in addition to the use of
physicalforce, broadens the nature of the violent act and expands the conventional understanding
of violence to include those acts that result from a power relationship, including threats and
intimidation. The use ofpower also serves to include neglect or acts of omission in addition to
more obviously violent acts of commission. Thus, the use ofphysicalforce or power should be
understood to include neglect and all types of physical, sexual, and psychological abuse, as well
as suicide and other self-abusive acts (2).
The definition also includes a broad range of outcomes, including psychological harm,
deprivation, maldevelopment, injury, and death. The broad range of outcomes reflects a growing
interest in public health to capture violence that does not necessarily result in injury or death, but
poses a substantial burden to individuals, families, communities, and health care systems
worldwide. For example, many forms of violence against women, children, or the elderly can
result in a range of physical, psychological, and social problems that do not necessarily result in
injury, disability, or death. These consequences can be immediate, as well as latent, and can last
years beyond the initial abuse. Defining outcomes solely in terms of injury or death thus limits
our understanding of the full impact of violence on individuals, communities, and society at
large.
One of the more complex aspects of the definition pertains to intentionality and there are a few
points to keep in mind regarding this aspect. First, even though violence is distinguished from
unintentional injuries, the intent to use force does not necessarily indicate intent to cause damage.
Indeed, there may be a great disparity between intended behavior and intended consequence. A
perpetrator may intentionally commit an act which, by objective standards, may be judged as
4
dangerous and highly likely to result in adverse health effects, but may not perceive it as such.
This distinction is particularly salient in the case of a youth involved in a physical fight with
another youth. The use of a fist against the head or the use of a weapon in the dispute increases
the likelihood of injury or death, though neither outcome may be intended. Shaken baby
syndrome, in which a parent intentionally shakes a crying child with the intent to quiet it and
instead causes brain damage, is another example of the intentional use of force without
necessarily intending to cause an injury.
A second distinction related to “intentionality” lies between the intent to injure and the intent to
“do violence.” Violence, according to Walters and Park (3) is culturally determined. Intention,
antecedent conditions, injury and other outcomes may all be part of the cultural definition. Some
perpetrators mean to do harm, but based upon their cultural backgrounds, do not perceive their
acts as violent. Female genital mutilation and corporal punishment of children are both examples
of where the act is regarded as a rite of passage or a disciplinary practice, respectively, and not
regarded as violent. The definition used by WHO, however, defines violence as it relates to the
health of individuals regardless of perceptions and cultural definitions of violence.
A final facet of intentionality lies in its graded nature. Some perpetrators mean both to injure
their victims and to do violence, but not to the extent that they actually do. A suicidal person may
not want to succeed, but rather to attract help. According to the WHO definition, however, intent
is simply present or absent, regardless of such distinctions. In defining violence as a whole, we
are interested in the outcomes of an act committed with the intent to harm, and not in whether
that intent exactly matches the resulting health outcomes.
Other definitional distinctions are addressed more directly in the chapters throughout this report.
For example, the WHO definition does not make a distinction between public and private acts
of violence, between reactive aggression (i.e., in response to antecedent conditions such as
provocation) or proactive aggression (i.e., instrumental or in anticipation of more self-serving
outcomes) (4), or between criminal and non-criminal acts. Each of these aspects, however, are
implicitly included in the definition and are important for understanding the etiology of violence
and for designing prevention programs.
Typology of Violence
One of the goals set forth in the WHO report Violence: A Public Health Priority (1995) was the
development of a typology of violence (2). Such a typology can graphically represent the
different types of violence and the links between them. Few typologies exist already and none
are comprehensive enough for our purposes (5). The typology proposed below first divides
violence into three broad categories: 1) self-directed, 2) interpersonal, and 3) collective. The
initial categorisation organizes violence according to characteristics of those who commit the
violent act. This initial categorisation differentiates violence which one inflicts upon oneself,
violence inflicted by another individual or small group of individuals, and violence inflicted by
larger groups such as states and nations.
5
A Typology of Violence
violence
I
I ben-inniaea
_____ I
I succsj oenawx j |
Nature of
Violence:
Deprivation ■>
]
I interpersonal
soctai
I
seri-aouse
11
|
I
cnaa
|
j ! ramiy/parmer
—i -partner
--I-
1 J
eiaer
comniunay
1 jaoquanianca 1 I
uoiiecave.
| | poJicai
j
~| |
economic
|
| T
stranger
111
12
^21.2
The three broad categories of violence are each divided further to reflect more specific categories
of violence. Violence that is directed towards one’s self is subdivided into two categories: a)
suicidal behaviour, and b) self-abuse. The former includes suicidal thoughts or ideation, attempts
(also called parasuicide or deliberate self-injury in some countries), and completed suicides.
Self-abuse, on the other hand, includes acts such as self-mutilation and other self-destructive
behaviours whose immediate intent is abusive, but not necessarily fatal.
Interpersonal violence is also divided into two main categories: a) family/partner (i.e., violence
that largely occurs between family members and intimates in the home, though not exclusively
in the home), and b) community violence (i.e., violence between individuals outside of the home
among persons known, but not related to one another, and among persons unknown to each
other). While the former includes child abuse, intimate partner violence, and elder abuse, the
latter would include, for example, gang violence, random acts of violence, stranger rape or sexual
assault, and violence in institutional settings such as schools, workplaces, and nursing homes.
Collective violence is subdivided into three categories: a) social, b) political, and c) economic
violence. Unlike the other two broad categories of violence, the subcategories of collective
violence draw attention to the possible motives of violence committed by larger groups of
individuals and states. Collective violence that is committed to advance a particular social
agenda would include, for example, hate crimes committed by well- or loosely organized groups,
terrorist acts, and mob violence. Political violence would include war, conflict, state violence
and other such acts committed by larger groups. Economic violence would include attacks
motivated by larger groups for economic gain including attacks to disrupt economic activity, to
deny access to essential services, or to create division and economic fragmentation. Clearly, acts
committed by larger groups can be committed to advance more than one agenda or have multiple
motives.
Each of the broader categories of violence is linked to a vertical axis revealing the nature of the
violent act it encompasses. While the horizontal axis answers the question of whom, the vertical
6
axis answers the question of how. Physical, sexual, psychological violence, and deprivation or
neglect represent the four major categories. With the exception of self-directed violence, these
four categories are included under each of the types of violence previously described. For
example, violence against children committed within domestic settings would encompass
physical, sexual, and psychological abuse, as well as neglect. Community violence committed
by persons known to one another would include, for example, physical assaults between youths,
dating violence, sexual harassment in the workplace, and neglect of older persons in long term
care facilities. Political violence would include such acts as rape during conflicts and physical
and psychological warfare.
The typology, while far from being universally accepted or perfect, does provide a useful
framework for understanding the broad, far-reaching, and complex patterns of violence taking
place around the world and in the everyday lives of individuals, families, and communities. It
also overcomes many of the limitations of other typologies by delineating the nature of violent
acts, the relevance of setting, the relationship between offender and victim, and in the case of
collective violence, possible motivations for violent acts.
How Do We Measure Violence, Its Impact and Prevention?
Different types of data are needed to describe the magnitude and impact of violence, to
understand which factors increase the risk for violent victimization and perpetration, and to
understand how well violence prevention programs are working. Some of these types of data
and sources are described in Table 1. Data on fatalities, specifically homicide and suicide,
can provide infomation on the extent of lethal violence in a particular community or country
and, when compared to other deaths, are useful indicators of the burden posed by violencerelated injuries. These data can also be used for monitoring changes in lethal violence over
time, identifying high risk groups and communities, and for making within and between
country comparisons.
Mortality data, however, represent only one possible type of data for describing the
magnitude of the problem. Since non-fatal outcomes are much more common than fatal
outcomes and because certain types of violence are not fully represented by mortality data
(e.g., child abuse, elder abuse, and violence against women), other types of data are necessary
for capturing the “who,” “what,” “when,” “where,” and “how” of violence and for describing
its full impact on the health of individuals and communities. Morbidity and other health data,
self-report, community, crime, cost, and policy or legislative data are examples of other types
of data that are useful for describing and understanding violence. For example, morbidity,
community, crime, and self-report data can be used to describe the characteristics of the
persons involved in violence, the circumstances surrounding violent events, the temporal and
geographic characteristics of violence, and some of the physical, mental, reproductive, and
other consequences of violence for individuals and communities. Cost data can be used to
describe the economic burden on health care systems, the years of potential lost life
associated with violence, and the potential cost savings associated with prevention programs.
There are a number of potential sources for the various types of data, including individuals,
agency or institutional records, local programs, community and government records,
7
population-based and other surveys, as well as special studies.- Though not listed in Table 1,
almost all sources include basic demographic information (e.g., a person’s age and sex).
Some sources include information specific to the violent event or injury (e.g., medical
records, police records, death certificates or mortuary reports). For example, emergency
department data may have information on the nature of the injury, how the injury was
sustained, and the place and time of occurrence. Police data may include information on the
relationship between victim and perpetrator, whether a weapon was involved, and other
circumstances related to the offense. Other data sources have much more detailed
information about the person, his or her background, attitudes and behaviours, or involvement
in violence (e.g., surveys, special studies) and are better for capturing violence that is not
reported to hospitals, police, or other agencies.
The availability, quality, and usefulness of the various data sources for measuring different
types of violence within and between countries varies considerably. Countries around the
world are at varying stages in the development of their data collection capacity.
Mortality data, among all sources of data, are the most widely collected and available. Many
countries maintain birth and death registries and keep basic counts of homicides and suicides.
Calculating rates from these basic counts, however, is not always possible because population
data are either not available or are unreliable. This is especially true in areas where
populations are in flux (e.g., areas experiencing conflict or continuous movements among
population groups) or where populations are more difficult to count (e.g., densely populated
or very remote areas). Systematic data on non-fatal outcomes, however, is not available in
most countries of the world, though efforts for developing such systems or collecting such
data are currently underway. A few documents providing guiding principles or data elements
for measuring different types of violence in different settings have also been published in
recent years (6-9).
Even when data are available, the quality of the information may be poor or less than
adequate for research and prevention purposes. Given that agencies and institutions keep
records for their own purposes and follow their own record-keeping procedures, the data from
these sources may be incomplete or lacking the kind of information necessary for describing
and understanding the problem of violence. For example, data from health services are
collected to allow optimal treatment of the patient. The medical record may contain
diagnostic information about the injury and course of treatment, but not the circumstances
surrounding the injury. These data may also be considered confidential and not available for
research purposes. Surveys, on the other hand, contain more detailed information about the
person, his or her background and involvement in violence, but are limited to how well a
person recalls events, admits to engaging in certain behaviors, which questions are asked,
how and by whom they are asked, as well as when, where, and how well the instrument is
administered.
Although beyond the scope of the discussion here, it should be noted that there are a number
of other challenges associated with the collection of violence-related data. These challenges
include, but are not limited to, developing measures that are relevant and specific to sub
population groups and different cultural contexts (6,7,9,10); developing tools to better link
data across sources to increase its usefulness; developing protocols to protect the
confidentiality of victims and ensure their safety (11); as well as attending to the many ethical
8
considerations associated with violence research.
How Big is the Problem of Violence?
The prevention of violence, according to the public health approach, begins with a description
of the magnitude and impact of the problem. A basic understanding of the patterns of
violence in a community, the groups at greatest risk for violence, and other characteristics of
the problem, is useful for identifying intervention strategies and targeting prevention
resources. All of the chapters in this report provide a description of fatal and non-fatal
outcomes as well as consequences for specific subtypes of violence. Here we describe some
of the global patterns of violence using data compiled specifically for this report from the
World Health Organization’s mortality database on diseases and injuries, the World Health
Organization’s Burden of Disease and Injury Report, as well as data from surveys and special
studies of violence.
Estimates of Mortality
In 1998, an estimated 2.3 million people died from violence, for an overall age-adjusted rate
of 38.4 per 100,000 (Annex Table 3). The vast majority of violence-related deaths in the
world occurred in low to middle income countries. Less than 10% of all violence-related
deaths occurred in high-income countries. Approximately 42% of the 2.3 million violencerelated deaths were suicides, 32% were homicides, and 26% were war-related. In 1998,
suicide was the 12th leading cause of death in the world, homicide the 15th and war the 19th.
Similar to many other health problems in the world, violence is not distributed evenly among
sex or age groups. In 1998, there were 736,000 homicides (overall age-adjusted rate of
12.2/100,000) (Annex Table 4). Males accounted for nearly 80% of all homicides and had
rates that were more than three times the rates of homicides among females (19.0 vs.
5.4/100,000). The highest rates of homicide among males were for those between the ages of
15 and 44 (31.1/100,000), while for females, the highest rates of homicide occurred among
those between the ages of 0 and 4 (9.3/100,000). With the exception of the youngest age
groups (i.e., 0-14 years of age), male homicide rates were approximately 3 to 6 times higher
than female homicide rates across each of the various age groups (i.e., 15-44 years, 45-59
years, 60+ years).
The different age and sex patterns for homicide among males and females, reflect in part,
different types of violence. The high rates of male homicide in the 15-44 year age group are
driven largely by high rates of youth interpersonal violence among males between the ages of
15 and 24. These patterns are described in more detail in Chapter 2. The high rates of child
homicide among females reflect the most severe form of child maltreatment. The highest
rates of homicide among female children 0-4 years of age are found in China, India, and some
of the low and middle income Eastern Mediterranean countries. In China, for example, the
rate of female child homicide is double the rate of male child homicide in the 0-4 age group
(15.7 vs. 7.9/100,000).
9
Worldwide, suicide claimed the lives of nearly 1,000,000 people in 1998 (overall ageadjusted rate of 16.4/100,000) (Annex Table 5). Approximately 60% of all suicides occurred
among males, and over half (53%) occurred among those between the ages of 15 and 44. For
both males and females, suicide rates increase with age and are highest among those over 59
years of age. Suicide rates, however, are generally higher among males than females (19.9 vs.
12.9/100,000). This is especially true among the oldest age groups, where worldwide, male
suicide rates among those 60+ years of age are almost twice as high as female suicide rates in
this age group (50.9 vs. 30.0/100,000).
Rates of violent death also vary according to country income levels. In 1998, the rate of
violent death in low/middle income countries (42.2 per 100,000) was more than double the
same rate in high-income countries (17.3 per 100,000) (Annex Table 3). The proportion of
violent deaths due to homicide or suicide also differed by income group (Annex Tables 4 and
5). In high-income countries in the Americas, there were three times more suicides than
homicides, while in low/middle income countries in the Americas there were more than four
times more homicides than suicides. There are also considerable regional differences in the
relative importance of these types of deaths. For example, in Sub Saharan Africa and in Latin
America and the Caribbean, it is estimated that there are 13 and 5 homicides, respectively, for
each suicide. However, in high-income countries in Europe and the Western Pacific there are
approximately 4 suicides for each homicide.
Within regions there are also large differences between countries. For example, in 1994
Colombia reported a male homicide rate of 146.5/100,000 population, while Mexico reported
a rate of 32.3 and Cuba 12.6 per 100,000 population (12). Large differences within countries
also exists between, for example, urban and rural populations, rich and poor, and between
different racial and ethnic groups. In the United States, for example, African-American
youths aged 15-24 had rates of homicide in 1998 (56.5/100,000) that were more than twice
the rate of their Hispanic counterparts (23.3/100,000), and over 13 times the rate of their
white, non-Hispanic counterparts (4.2/100,000) (13).
Estimates of Disability Adjusted Life Years
Mortality is an important indicator of the burden of violence and is probably the only one for
which relatively accurate data has been collected in some parts of the world. However, as
mentioned earlier, mortality data provide only a partial picture of the magnitude of the
problem. For every person that dies from a violence-related injury, many more survive, often
with permanent disabling sequelae. In recent years, a new indicator has been developed that
combines the numbers of years of life lost from premature death with the loss of health from
disability. This indicator is known as the Disability Adjusted Life Years or DALY (14).
In 1998, war, self-directed, and interpersonal violence were ranked respectively xx, xx and
XXth among the leading causes of the world’s disability adjusted life years (DALYs) lost
(Annex Table 15). It is estimated that in 2020, war will rank 8th, interpersonal violence 12th
and self-directed violence 14th. In low and middle income countries, war is among the 15
causes of DALYs lost for persons aged 0 to 44, while homicide is the 4th leading cause of
DALYs lost in this age group, considered in many parts of the world to be the most
10
economically productive age group. In high-income countries, violence also ranks among the
15 leading causes of DALYs for persons aged 0 to 44. Similar to mortality patterns, males
and those persons under the age of 45 are disproportionately represented among victims of
premature death and illness in both their absolute numbers and rates of disability adjusted life
years lost.
Estimates of Non-Fatal Violence
It is important to realise that the above rankings for mortality and DALYs most likely
underestimate the true burden of violence. In all parts of the world, deaths and the proportion
of lost years of life from premature death and loss of health from disability represent the “tip
of the iceberg.” Physical and sexual assaults occur on a daily basis, though precise national
and international estimates of each are clearly lacking. Not all assaults result in injuries
severe enough to require medical attention and, even among those that do result in serious
injuries, surveillance systems for reporting and compiling these injuries are either not
available or are presently under development in some countries.
Much of what is known about non-fatal violence comes from surveys and special studies of
different population groups. For example, lifetime estimates of physical assaults by intimate
partners, from national surveys of women, range from 5.1 in the Philippines and 9.5 in
Paraguay, to 22.1 in the USA, 29.0 in Canada, and 34.4 in Egypt (15-19). Lifetime estimates
of sexual assault (attempted or completed) among women living in cities or provinces around
the world range from 15.3 in Toronto, Canada to 21.7 in Leon, Nicaragua, 23.0 in North
London, and 25.0 in one province in Zimbabwe (20-23). Rates of physical fighting in the
past year among adolescent males range from 22% among boys (grade 7) in Sweden, 44% of
boys (grades 9-12) in the USA, to 76% among boys (grades 8-10) in Jerusalem (24-26).
It is important to keep in mind that since these data are largely based on self-reports, it is
difficult to know whether they over- or underestimate the true extent of physical and sexual
assaults among these population groups. Certainly in those countries with strong cultural
pressures to keep violence “behind closed doors,” the nature and extent of non-fatal violence
is likely to be underreported. Victims of violence may be reluctant to discuss violent
experiences not only out of shame and taboos, but also out of fear and because the admission
of experiencing certain violent events (e.g., rape) in some countries may result in death (e.g.,
the killing of women who have been raped to preserve family honour, also known as “honour
killings”).
There are also a number of other health consequences associated with violence, and similar to
non-fatal violence, it is difficult to know the true extent of these consequences or the precise
burden of these consequences on health care systems, as well as the economic productivity of
nations around the world. The direct and indirect health care costs associated with violence
related injuries are estimated to be in the billions of dollars in some developed countries such
as the United States. National and international estimates of violence-related health
consequences, such as depression, smoking, alcohol and drug use, unwanted pregnancy,
sexually transmitted diseases, HIV/AIDS, and other infections (all of which have been linked
11
to violence in small scale studies)(27-32) are also lacking. The economic burden associated
with these consequences is also unknown.
How and Why Does Violence Occur?
Violence is a multifaceted and complex problem. No single factor explains why some
individuals behave violently toward others or why violence is more prevalent in some
communities and not others. Violence is the result of the complex interplay of individual,
relationship, sociocultural, and environmental factors. Understanding how these factors are
related to violence is one of the important steps in the public health approach to preventing
violence.
Some of the chapters in this report use an ecological model to understand the multifaceted nature
of violence. The ecological model explores the relationship between individual and contextual
factors and considers violence as the product of multiple levels of influence on behaviour (Figure
2). Beginning with the individual, the first level of the ecological model seeks to identify the
biological and personal history factors that an individual brings to his or her behaviour. Not only
are demographic factors considered, but also factors such as impulsivity, low educational
attainment, alcohol abuse, and prior history of aggression and abuse. In other words, at this level
the ecological model focuses on the characteristics ofthe individual that increase the likelihood
of violent victimisation. This level of the ecological model also focuses on the characteristics
of the individual that increase the likelihood of perpetration of violence.
The second level of the ecological model explores how proximal social relationships (e.g., with
peers, intimate partners, or family members) increase the risk for violent victimisation and
perpetration. In the case of partner violence or child maltreatment, for example, interacting
almost daily and sharing a common domicile with an abuser may increase the opportunity for
violent encounters. Because individuals are bound together in a continuing relationship, it is
likely that the victim will be repeatedly violated by the offender (33). In the case of youth
interpersonal violence, previous research shows that youths are much more likely to engage in
negative activities when those behaviours are encouraged and approved by their friends (34,35).
Peers, intimate partners, and family members all have the potential to shape an individual’s
behaviour and range of experience.
The third level of the ecological model examines the community contexts where social
relationships are embedded (e.g., school, workplace, neighbourhoods) and seeks to identify the
characteristics of these settings that are associated with violent victimization and/or perpetration
(e.g., high residential mobility, high population density, heterogeneity, institutional practices),
or factors related to such settings (e.g., social isolation, drug trafficking). Previous research
indicates that opportunities for violence are greater in some community contexts than others (e.g.,
in areas of poverty, physical deterioration, or where there are few institutional supports).
The last level of the ecological model examines the larger societal factors that influence
differential rates of violence, including those factors that create an acceptable climate for
violence; those factors that reduce inhibitions against violence; and those factors that create and
sustain gaps between different segments of society or that create tensions between groups or even
12
countries. Larger societal factors include, for example, cultural and subcultural norms that
support violence as an acceptable way to resolve conflict, attitudes that support suicide as an
individual choice instead of a preventable act of violence, norms that support parental rights over
child welfare, norms that encourage male dominance over women and children, norms that
support police use of excessive force over citizens, or norms that support political conflict.
Larger societal factors also include the health, education, economic, and social policies that
sustain gaps between the rich and poor or that keep one group at a disadvantage over another
group (see Box 1).
The ecological framework alerts us to the multifaceted nature of violence and the interaction of
risk factors operating within a broader social, cultural, and economic context. It is important to
note that while some risk factors may be unique to a particular type of violence, more often the
various types of violence share a number of risk factors. Prevailing cultural norms, poverty,
social isolation, and such factors as alcohol consumption, substance abuse, and access to firearms
are risk factors for more than one type of violence. As a result, it is not unusual for some
individuals at risk for violence to experience more than one type of violence. Women at risk for
physical violence by intimate partners, for example, are also at risk for sexual violence (36).
It is also not unusual to see links between different types of violence. For example, previous
research shows that exposure to violence in the home is associated with violent victimisation and
perpetration in adolescence and adulthood (37). The experience of rejection, neglect, or
indifference from parents leaves children at greater risk for aggressive and antisocial behaviour,
including abusive behaviour as adults (38-40). Associations have been found between suicide
and several types of violence, including intimate partner violence (41), sexual assault (27,28),
and elder abuse (42,43). Suicide rates have been shown to decrease during wartime, in Sri Lanka
for example, and to increase after hostilities cease (44,45). In many countries that have suffered
violent conflict, the rates of interpersonal violence remain high even after cessation of hostilities
because of the social acceptability of violence and the availability of weapons among other
factors.
Identifying the factors that increase or reduce the risk for violent victimization and perpetration
at each level of the ecological model is an important step toward preventing violence. Once these
factors are identified, researchers and prevention specialists can then begin to develop
interventions or prevention programs aimed at reducing or modifying the risk for violence. The
ecological model in this regard serves a dual purpose — each level in the model represents a
level of risk and each level can also be thought of as a key point for intervention. Given the links
between violence and the interaction between individual factors and the broader social, cultural,
and economic context suggests that addressing risk factors across the various levels of the
ecological model may contribute to decreases in more than one type of violence.
13
What Can Be Done to Prevent Violence?
This section of the introduction is still being written. Listed below is a brief outline of the
discussion for this section.
Outline:
I. Violence is preventable.
□ Cross-cultural examples of successful approaches
II. Lessons learned from research on violence
□ Start early
□ Disrupt the developmental pathways of violence
□ Break the cycle of violence
□ Intervene at multiple levels (individual, family, community, society)
□ Address victims and perpetrators
III. Prevention and Policy Responses
□ Importance of primary, secondary, and tertiary prevention
□ Need for multi-sector responses across and within types of violence
□ What can the health sector do?
□ What can policy-makers do?
IV. Putting Knowledge into Practice
□ Importance of small scale efforts
□ Promising vs. proven approaches: balancing public health action
□ Avenues for sharing knowledge and information
14
Conclusion
Violence is a major public health problem worldwide, resulting in over two million deaths
each year. It comes in many forms, is present in a variety of contexts, and affects men,
women, and children of all ages and socioeconomic backgrounds. Not only is violence
responsible for numerous deaths and injuries worldwide, it is also an important risk factor for
many other immediate and long-term health problems — problems that can affect individuals,
families and communities for years. The personal, psychological, and social costs of violence
have serious implications for the future human, economic, and social development of nations
around the world.
While human atrocities around the globe capture widespread attention, the everyday violence
of life in families and communities continues unabated. Violence, however, is preventable.
Public health is concerned with the well being of populations as a whole and believes that
collective action is necessary to achieve important reductions in violence. Creating safe and
healthy communities around the globe requires commitment on the part of multiple sectors at
the international, national, and community levels to document the problem, build the
knowledge base, promote the design and testing of prevention programs, and promote the
dissemination of lessons learned.
Public health officials have a very important role to play in this process. Through their vision
and leadership, much can be done to establish national plans and policies for violence
prevention, to help facilitate the collection, availability and quality of data to effectively
document and respond to the problem, to build important partnerships with other sectors, and
to ensure an adequate commitment of resources to support prevention efforts. Violence is
not simply a social ill or a social justice problem, it is an important health problem that
deserves urgent attention.
15
References:
1. Mercy JA, Rosenberg ML, Powell KE, Broome CV, Roper WL. Public health policy for
preventing violence. Health Affairs 1993; 12(4):7-29.
2. World Health Organization. Violence: A Public Health Priority, 1995.
3. Walters RH, Parke RD. Social motivation, dependency, and susceptibility to social influence.
In: Berkowitz L (ed.) Advances in experimental social psychology, Vol 1. New York, NY:
Academic Press 1964; 231-276.
4. Dodge KA, Coie JD. Social information processing factors in reactive and proactive
aggression in children’s peer groups. J Pers Soc Psychol 1987; 53:1146-58.
5. Foege WH, Rosenberg ML, Mercy JA. Public health and violence prevention. Current
Issues in Public Health 1995; 1:2-9.
6. World Health Organization. WHO multi-country study of women’s health and domestic
violence. Core Protocol, Geneva, Switzerland: WHO/EIP/GPE/99.3, 1999.
7. WHO Working Group for Injury Surveillance Methodology Development and its
Technical Group. ICECI - Guidelines for counting and classifying external causes of
injuries for prevention and control, Report No. 208. Geneva, Switzerland: World Health
Organization, April 1998.
8. Sethi D, Krug E. Guidelines for surveillance of injuries due to landmines and unexploded
ordnance. Geneva, Switzerland: Injuries and Violence Prevention Department, Noncommunicable Diseases and Mental Health Cluster, World Health Organization, 2000.
9. Saltzman LE, Fanslow JL, McMahon PM, Shelley GA. Intimate partner surveillance:
uniform definitions and recommended data elements, Version 1.0. Atlanta, GA, USA:
National Center for Injury Prevention and Control, Centers for Disease Control and
Prevention, 1999.
10. Dahlberg LL, Toal SB, Behrens CB. Measuring violence-related attitudes, beliefs, and
behaviors among youths: a compendium of assessment tools. Atlanta, GA, USA: Centers
for Disease Control and Prevention, National Center for Injury Prevention and Control,
1998.
11. World Health Organization. WHO putting women first: ethical and safety
recommendations for research on domestic violence against women. Geneva,
Switzerland: WHO/EIP/GPE/99.2, 1999.
12. World Health Organization. 1996 World Health Statistics Annual. Geneva, Switzerland:
World Health Organization, 1998.
16
13. Murphy SL. Deaths: final data for 1998. National vital statistics reports; 48, 11.
Hyattsville, MD: National Center for Health Statistics, 2000.
14. Murray CJL, Lopez AD (eds.). The global burden of disease: a comprehensive
assessment of mortality and disability from diseases, injuries, and risk factors in 1990 and
projected to 2020. Global Burden of Disease and Injury Series, Vol. 1. World Health
Organization. Cambridge, MA: Harvard University Press, 1996.
15. Demographic and Health Surveys (DHS). Domestic violence and rape. In: National safe
motherhood survey 1993. Calverton, USA: Macro International Inc., 1994.
16. Encuestra nacional de demografia y salud reproductiva, 1995-1996. Centro Paraguaya de
estudios de poblacion, 1996.
17. Tjaden P, Thoennes N. Prevalence, incidence, and consequences of violence against
women: findings from the National Violence Against Women Survey. Washington, DC.:
US Department of Justice, National Institute of Justice, NCJ 172837, November 1998.
18. Rodgers K. Wife assault: the findings of a national survey. Juristat service bulletin of the
Canadian Centre for Justice Statistics 1994; 14(9): 1-22.
19. Demographic and Health Surveys (DHS). Egypt demographic and health survey. Cairo,
Egypt: National Population Council and Macro International Inc., 1995.
20. Randall M et al. Sexual violence in women’s lives: findings from the women’s safety
project, a community-based survey. Violence Against Women 1995; 1(1):6-31.
21. Ellsberg MC. Candies in hell: domestic violence against women in Nicaragua. Umea,
Sweden: Department of epidemiology and public health. Umea University, 1997.
22. Mooney J. The hidden figure: domestic violence in North London. London, UK:
Middlesex University, 1993.
23. Watts C et al. Women, violence and HIV/AIDS in Zimbabwe. SafeAIDS 1997; 5(2):2-6.
24. Grufman M, Berg-Kelly K. Physical fighting and associated health behaviours among
Swedish Adolescents. ActaPaediatr 1997; 86(1):77-81.
25. Kann L, Kinchen SA, Williams BI, Ross JG, Lowry R, Grunbaum J, Kolbe L. Youth risk
behavior surveillance - United States, 1999. CDC Surveillance Summaries, June 9, 2000.
MMWR 2000; 49(No. SS-5):7,43.
26. Gofin R, Palti H, Mandel M. Fighting among Jerusalem adolescents: personal and
school-related factors. J Adolesc Health 2000; 27(3):218-23.
27. Kaplan SJ et al. Adolescent physical abuse: risk for adolescent psychiatric disorders. Am
J Psychiatry 1998; 155(7):954-59.
17
28. Kaslow NK et al. Factors that mediate and moderate the link between partner abuse and
suicidal behavior in African-American women. J Consult Clin Psychol 1998; 66(3):53340.
29. Pederson W et al. Alcohol and sexual victimization: a longitudinal study of Norwegian
girls. Addiction 1998; 91(4):565-81.
30. Holmes MM et al. Rape-related pregnancy: estimates and descriptive characteristics from
a national sample of women. Am J Obstetrics and Gynecology 1996; 175(2):320-25.
31. Kakar F, Bassani F, Romer CJ, Gunn SWA. The consequences of landmines on public
health. Prehospital Disaster Medicine 1996; 11:41-45.
32. Toole MJ. Complex emergencies: refugee and other populations. In: Noji E (ed.) The
Public Health Consequences of Disasters. New York, NY: Oxford University Press
1997; 419.42.
33. Reiss AJ, Roth JA (eds.). Violence in Families. Understanding and preventing violence:
panel on the understanding and control of violent behavior. Vol. 1. Washington, DC:
National Academy Press, 1993:221-245.
34. Thomberry TP, Huizina D, Loeber R. The prevention of serious delinquency and
violence: implications from the program of research on the causes and correlates of
delinquency. In: Howell JC, Krisberg B, Hawkins JD, Wilson JJ (eds.) Sourcebook on
serious, violent and chronic juvenile offenders. Thousand Oaks, CA: Sage 1995:213-37.
35. Lipsey MW, Derzon JH. Predictors of serious delinquency in adolescence and early
adulthood: a synthesis of longitudinal research. In: Loeber R, Farrington DP (eds.)
Serious and violent juvenile offenders: risk factors and successful interventions.
Thousand Oaks, CA: Sage 1998:86-105.
36. Heise LL, Ellsberg M, Gottemoeller M. Ending violence against women. Baltimore,
MD: Johns Hopkins University School of Public Health, Population Information Program,
Population Report Series L, No. 11, 1999.
37. Maxfield MG, Widom CS. The cycle of violence. Revisted 6 years later. Archives of
Pediatric Medicine 1996; 150(4):390-5.
38. Farrington DP. The family backgrounds of aggressive youths. In: Hersov LA, Berger M,
Shaffer D (eds.) Aggression and antisocial behavior in childhood and adolescence.
Oxford, England: Pergamon Press 1978;73-93.
39. McCord J. A forty year perspective on the effects of child abuse and neglect. Child
Abuse and Neglect 1983; 7:265-70.
18
40. Widom CS. Child abuse, neglect, and violent criminal behavior. Criminology 1989;
27:251-72.
41. Stark E et al. Killing the beast within: woman battering and female suicidality.
International Journal of Health Services 1995; 25(l):43-64.
42. Bristowe E, Collins JB. Family mediated abuse of non-institutionalised elder men and
women living in British Columbia. Journal of Elder Abuse and Neglect 1989; l(l):45-54.
43. Pillemer KA, Prescott D. Psychological effects of elder abuse: a research note. Journal of
Elder Abuse and Neglect 1989; l(l):65-74.
44. Somasundaram DJ, Rajadurai S. War and suicide in Northern Sri Lanka. Acta Psychiatr
Scandi 1995; 91(l).T-4.
45. Lester D. The effect of war on suicide rates: a study of France from 1826 to 1913. Eur
Arch Psychiatry ClinNeurosci 1993; 242(4):248-9.
19
Figure 1: The Public Health Approach to Violence
Implement and
Disseminate
Widely
Design & Test
Interventions
Identify Risk
and Protective
Factors
Describe the
Problem
Problem
Response
Adapted from: Mercy JA, Rosenberg ML, Powell KE, Broome CV, Roper WL. Public Health Policy for
Preventing Violence. Health Affairs 1993; 12(4):7-29.
20
Table 1: Type and sources of data for describing the magnitude and impact of violence
and for understanding the etiology of violence
Type of Data
Mortality
Morbidity and other health
data
Self-report
Community
Crime
Cost
Policy or legislative
Examples of the types of
information collected
Characteristics of the
Death certificates, vital
statistics registries, coroner decedent, cause of death,
location, time, manner of
or mortuary reports
death_____ ___________
Diseases, injuries, as well
Hospital, clinic, or other
as
information on physical,
medical records
mental, or reproductive
health_________________
Attitudes, beliefs,
Surveys, special studies,
behaviours, cultural
focus groups
practices; exposure to
violence in the home or
community____________
Population counts and
Population records, local
density; levels of income,
government, or other
education, unemployment;
institutional records
divorce rates___________
Type
of offence,
Police records, judiciary
characteristics of offender,
records, surveys
relationship between
victim and offender,
circumstances of event
Expenditures on health,
Program, institutional or
housing, social services;
agency records; special
costs of treating violencestudies
related injuries; utilisation
patterns_______________
Government or legislative Laws, institutional policies
and practices
records
Data Sources
21
Figure 2: Ecological Model for Understanding Violence
: •
■.
Structural
Community
Relationship
I
.•
•
■
J
Individual
1
_____
/
Adapted from: Tolan P, Guerra N. What works in reducing adolescent violence: an empirical review of the
field. Boulder, CO: Center for the Study and Prevention of Violence, 1994, and Heise LL.
Violence against women: an integrated ecological framework. Violence Against Women 1998;
4:262-290.
22
Box 1: Violence and Social Inequality
The issue
Over the years, a lot of scientific evidence has been produced in many different countries, showing that people
from lower socioeconomic groups have mortality rates significantly higher than those from higher groups, at
least with regard to each major cause of death. In fact, social inequality in health has survived major
improvements in medical science, several stages of technological development, considerable demographic
changes, and substantial efforts to set up more equitable public-health systems, governments and states.
What have changed considerably, however, are the health outcomes contributing to social-health differentials. Of
particular concern nowadays are deaths from accidents and violence, since they are — or are becoming — some of
the causes of death with the steepest social-class gradients. The largest differentials are found in childhood and
youth, periods of life where injuries account for many premature deaths in most parts of the world.
State of current knowledge
As is the case for any health outcome, socioeconomic differences in violence-inflicted injuries have both
individual and contextual explanations. Nevertheless, research has been concerned mainly with the former rather
than the latter.
Studies dealing with differences between individuals reveal that people with low socioeconomic status, measured
in terms of income, profession or education, are at greater risk of being injured by violence both inside the home
(especially boys in childhood and women in adulthood) and outside the home (especially men, except with
regard to sexual assault). This has been observed on continents throughout the world: Africa (Egypt, South
Africa), America (Brazil, the US), Asia (India, Taiwan), Europe (Denmark, Great Britain, Greece, Spain) and
Australia. A majority of the studies conducted have concerned the industrialized world.
For their part, comparisons between living areas indicate that, in general, the lower the material standard of
living, the worse is the quality of life as indicated, for instance, by rates of mortality and morbidity for violenceinflicted injuries, including sexual violence. In addition, there is growing evidence that large income differences
within individual countries (as is the case in for example Brazil, Colombia and South Africa as opposed to for
example Austria, Canada, Poland and Scandinavia), states (for instance in some states in the US) or areas are
associated with higher incidence rates of various forms of violence and violence-inflicted injuries.
In sum, there is a rather consistent pattern in the social distribution of death by violence-inflicted injury, which is
to the detriment of less privileged individuals and people in poorer living areas. Nevertheless, the mechanisms
underlying the differences are not well understood, and many crucial questions — such as how places and people
interact in the social etiology of injuries due to violence — remain largely unanswered. It is also unclear whether
social differentials in morbidity patterns are in all instances similar to those with regard to mortality. Are people
of low socioeconomic status more likely to die from any injuries they sustain?
Why are social differences in violence-inflicted injuries a public-health issue?
The social patterning of violence-related injuries is a crucial public-health issue for several reasons. One is the
current increase in the concentration of wealth between and within countries. This is expected to have spillover
effects on the social fabric and, consequently, on the incidence of crime and violence and the social distribution
of victims.
Another is that social differences in violence-related injuries are neither unavoidable nor irreversible. In fact,
abatement strategies may even help combat social inequalities in individual risks without addressing the
upstream mechanisms of wealth distribution. One can think here of collective measures, such as weapon/alcohol
control or street lightning, which impact on the mediating factors that increase the risk of violence and affect
their social distribution.
A third reason is that the existence of social differences in wealth may hinder the penetration of injury-control
and safety-promotion strategies or impede their long-lasting effects. Accordingly, it is essential to reflect on the
23
strategies for prevention that are most likely to work, where - and for whom - they are needed most. This is of
key importance when there is an expectation for prevention to be mediated by safety-promotion work at
community level.
Public-health actions
Actors in the public-health arena cannot do everything, but they certainly can contribute to diminishing and
combating social differences in violence-inflicted injuries, and thereby reduce the overall risk. The production
and distribution of educational material targeted at the whole population, and also at particular groups (e.g.
actual or potential victims or offenders, health-care personal), might enable public-health agents to improve the
capacity of individuals and communities to act to protect and improve their safety. Improved efficacy and better
localization of health and social services are other possibilities.
As local partners, public-health agents can act as facilitators in mobilizing the social and material resources
needed locally for various services (e.g. shelters, telephone help lines, street lighting, school curriculum). They
may also act as advocates on behalf of individuals or communities to enable them to overcome structural barriers
(legal, economic, or journalistic).
A definite moral stance should be adopted so as to help reduce the cultural and social acceptance of use of
violence in various situations. Complementary, targets could be set that are specifically aimed at reducing the
size of the gap between the mortality and morbidity rates of the most and the least advantaged groups and areas.
Measures of these differences could be employed as indicators of the potential for improvement in a nation’s or
living area’s health-and-safety status.
Lucie Laflamme and Karin Engstrom
Karolinska Institute, Department of Public Health Sciences, Division of Social Medicine, Norrbacka, SE-171 76
Stockholm, Sweden
National Institute of Public Health, Injury Prevention Program, SE-103 52 Stockholm, Sweden
24
Box 2: DESEPAZ
In 1992 the City of Cali, Colombia, led by its elected public health specialist Mayor, set up a comprehensive
program aimed to reduce the high levels of crime in the city. In Cali, a city with approximately 2 million
inhabitants, rates of homicide had increased from 23 in 1983 to 85 per 100,000 inhabitants in 1991, while those
from traffic accident deaths were around 21. Violence was defined as a major priority for the city and a broad
spectrum of political forces within the city were convened to design a strategy to reduce it. The resulting
program was named DESEPAZ, an acronym for Desarrollo. Seguridad y Paz.
DESEPAZ main guidance principles were: 1) Reliable and timely information about crime is crucial and
consequently epidemiological surveillance and research had to be organized. 2) Violence is multicausal and thus
no simple strategies would solve the problem. 3) Strengthening of the Police and the Judiciary System are
prerequisites for orderly city life. 4) Prevention should be prioritized. 5) Community commitment from
different stakeholders such as entrepreneurs, church leaders, academicians, journalists, and the citizens
themselves should be accomplished. 6) Education for tolerance and respect to others rights is necessary. 7)
Diminishing social and economic inequities should be actively pursued.
Following the public health strategy, descriptive epidemiological studies were carried out in order to identify
risk factors for violence. Actions were taken based on the information provided by the surveillance system.
Restriction on the sale of alcohol and banning the permits to carry handguns were put on weekends and other
special occasions. Investments for institutional strengthening of the police, the judiciary systems and the human
rights advocate’s office were approved. Education in civil rights was prepared for the police and the civil
society. Special projects towards prevention of violence and promoting tolerance among the citizens were
implemented. Television spots showing the importance of tolerance, self-control and respect for others’ rights
were displayed at peak rating hours. Cultural and educational projects were organized to promote dialogue,
conflict’s resolution, and education at school and family levels, in collaboration with NGO’s.
Special projects to improve income and safe leisure and recreation for young people were organized. The city
administration contracted specially trained youth groups, most of them coming from gangs, to build roads, clean
parks and take care of open spaces. Community participation and commitment was obtained through weekly
“Community Councils” where leaders and plain people presented and discussed with the mayor and members of
his cabinet proposals to deal with crime situations.
The great majority of funds for DESEPAZ came from the municipal budget, although contributions from the
national government and other sources were obtained. The continuation of DESEPAZ by the next two
administrations is by itself an important recognition to the degree of acceptance of the projects. It must be said
though that support has fluctuated and in some projects has not existed. In 1998 the Inter American Bank
approved a 10 million dollar 4-year loan for the implementation of new preventive projects and the maintenance
of those initially defined and there exists new enthusiasm and support. Public opinion around violence has
moved from a passive attitude to an active demand for more prevention. Different members of the civil society
are pushing to improve and to incorporate new strategies to prevent any form of violence. Reduction of violence
has become a city goal.
The homicide rate declined from an all time high 124 in 1994 to 86 in 1997, a 30% reduction in a period of three
years. In 1998, mostly due to a weakening of the prevention strategies, the rate increased to 87. In absolute
numbers there has been a reduction of about 600 killings, which is permitting the police and the judiciary to
utilize their scarce resources on more organized forms of crime.
25
Chapter 2
Youth Violence as a Global Public Health Problem
Status:
Draft: 3rd
Peer Reviewed: Yes
Date of Current Draft: 10 September 2000
Table of Contents
l.
IL
m.
IV.
V.
VI.
vn.
Introduction
Definitional Issues
Patterns and Trends in Mortality
A.
Materials and Method
B.
Youth Homicide Data Availability by Region
C.
Youth Homicide Rates by WHO Regional and Economic Groupings
D.
Youth Homicide Rates by Age and Sex
E.
Methods of Youth Homicide
F.
Trends in Youth Homicide Rates and Methods
G.
Other Studies of Youth Homicide
H.
Conclusions
Predictors, Causes, and Correlates of Youth Violence
A.
Individual, Family, Peer, and Neighborhood Influences
1. Measurement
2. Specialization Versus Versatility
3. Continuity
4. Biological Risk Factors
5. Psychological/Personality Factors
6. Family Factors
7. Peer, Socio-economic, Neighborhood Factors
8. Situational Factors
Macro-level Influences on Youth Violence
B.
1. Economic Conditions
2. Political Structure
3. Social Integration
4. The Cultural Context
5. Social Change
C. Conclusions
A Typology of Youth Violence Prevention Strategies
The Effectiveness of Youth Violence Prevention Strategies
A.
Infancy and Early Childhood
B.
Adolescence
Early Adulthood
C.
Summary and Recommendation
2
Introduction
Youth violence is a global tragedy. Adolescents and young adults are the primary victims and
perpetrators of violence in almost every region of the world.1 There is substantial evidence that
homicide and nonfatal assaultive violence involving youth are important contributors to the global
burden of premature death, injury, and disability.1’2 Communities in many parts of the world are
searching for successful ways to prevent youth violence. The current challenge is to identify and
successfully implement policies and programs that are effective in reducing these destructive
behaviors and their detrimental health and social consequences.
It is important to acknowledge at the outset that youth violence is a complex problem caused by the
interaction of numerous biological, psychosocial, and environmental factors. Patterns and trends in
youth violence are closely linked to stages of human development and the social context in which
such development occurs.3 Youth violence is also closely linked with other forms of violence.
Witnessing violence in the home or being physically abused as a child, for example, may engender
adolescent violence by socializing children to believe that violence is an acceptable strategy for
resolving interpersonal problems.4’5 In some nations, prolonged exposure to war and other forms of
armed conflict may contribute to a culture of violence that, in turn, increases rates of youth violence. ‘
8 Understanding the factors that place young people at risk for or protect them from violent
victimization and perpetration is critical towards developing effective violence prevention policies
and programs.
We must also acknowledge, given the complexity of this issue, that there is no single solution to the
problem of youth violence. Multi-faceted approaches that combine effective and complementary
strategies are needed. Moreover, it is also clear that the same approach may not be equally effective
in different cultural contexts. Consequently we cannot assume that what works to prevent youth
violence in the United States will necessarily be effective in South Africa or Colombia and visa versa.
Over the past two decades, however, we have learned a great deal about the etiology and prevention
of youth violence. This knowledge, although based primarily on research conducted in western
cultures and wealthier countries, provides a foundation upon which to guide potentially productive
programs for youth violence prevention. There is much more to be learned about prevention,
however, by comparing and contrasting knowledge about youth violence across different cultural and
political contexts.
The purpose of this chapter is fourfold: 1) to describe global patterns in violent victimization and
perpetration among youth, 2) to review what is known about the predictors, causes, and correlates
of youth violence, 3) to highlight promising and effective prevention strategies, and 4) to provide
recommendations for future action.
3
Definitional Issues
In this chapter we are focusing on the violent victimization of youth by others and the perpetration
of interpersonal violence by youth. Youth are defined as including adolescents and young adults
between 10 and 24 years of age. For our purposes, youth violence is defined as the intentional,
threatened, or actual use of physical force or power against another person that either results in or has
a high likelihood of resulting in injury, death, psychological harm, maldevelopment, or deprivation.9
Types of violence that impact youth, but are excluded from the above definition, include suicidal
behavior, war-related violence, and political violence. Although within the scope of the above
definition, we will not emphasize youth violence that occurs in the context of dating relationships
or in the context of unequal relationships of responsibility, trust, or power (e.g., child abuse and
maltreatment).
Various approaches to categorizing youth violence have been used, although there is no widely
accepted standard. The primary types of interpersonal violence are homicide, rape, robbery, and
assault. These types can be further subdivided by characteristics of the offender (e.g. stranger,
acquaintance, intimate), the motivation of the offender (i.e. emotional violence where the main
objective is to hurt the victim versus instrumental violence where the main objective is to obtain
something), characteristics of the victim (e.g. homicides of children, parents, siblings), or
characteristics of the offense (e.g. school violence, gang violence).10
In this chapter we address youth violence from the perspective of both the victim or victimization and
the offender or perpetration. In describing the health consequences and burden of youth violence we
focus on epidemiologic patterns in deaths, injuries, and violent victimizations suffered by youth. In
reviewing what’s known about risk factors for youth violence we focus on what is known to cause
youth to behave violently towards others. The primary, but not exclusive thrust of youth violence
prevention programs is to prevent the expression of violence by youth. Nevertheless, there are
measures that can be taken to protect youth from being victimized. Youth victimization and
perpetration are closely related in that they tend to occur under similar circumstances, share a
common etiology, and may be prevented through similar policies and prevention strategies.
Moreover, perpetrators are disproportionately likely to be victimized, and vica versa.11,12
Patterns and Trends in Mortality
Fundamental to understanding youth violence is the descriptive epidemiological information
provided by routine mortality surveillance at the country level. Although fatalities represent only the
tip of the youth violence iceberg, death rate differences between countries and regions are perhaps
the only reliable way of measuring youth violence and identifying cross-national and historical
changes in the size and shape of the problem.
The WHO’s burden of disease projections for homicide, shown in Table xxx of the main appendix,
provide indirect sources of epidemiological information about fatal youth violence. For 1998, these
4
projections estimated that 736,000 homicides occurred in the world, of which 69% involved victims
aged 15 to 44 years. Nineteen percent of victims were 45 years and over, and 11% were aged 0 to 14
years. There were an estimated six males per female victim aged 15 to 44, and homicide rates in this
age range were between two and 10 times higher in low- to middle-income regions of the world than
in high-income regions.
The availability of these projections for all continents creates the impression that youth homicide
death rates are equally well monitored in all world regions. This perception is far from the truth, and
masks major differences between regions in the availability of actual data on youth homicide.
Furthermore, the age categories used in the projections obscure key changes in the patterns of
victimization occurring from 10 to 24 years of age. The regional differences in data availability show
where resource development in violence and injury surveillance is most required, while age changes
provide clues about possible underlying causes. In this section we therefore use actual data on youth
homicides instead of projections, and divide the period from 10 to 24 years old into three
developmentally important age bands: pre-adolescence from 10 to 14 years; adolescence from 15 to
19 years, and young adulthood from 20 to 24 years.
Materials and Methods
To describe the epidemiology of youth homicides we used mortality data from WHO, PAHO and the
Brazilian Health Ministry (www.datasus.gov.br), and for countries and regions inadequately covered
by these sources we carried out a literature review of studies from 1990 onwards. Homicides were
defined as cases coded E960 to E969 in ICD9 and X85 to¥09 in ICD 10.Because socio-economic
status is a well-established macro-level predictor for overall homicide rates, youth homicide rates
were calculated for regions defined according to country income levels. These regions are the same
as used in the main appendix data tables, and were defined by dividing WHO geographical areas into
clusters of low- to middle-income and high-income countries according to World Bank criteria.
The most recent year for which data were available across the largest number of countries (N = 68)
was 1994. To minimise the effect of differences in country population size and annual fluctuations
in reporting completeness, one year weighted averages using data for two or more years between
1992 and 1996 were computed. For each age and sex category, the number of homicides and
population totals for all country years in each region were summed seperately, and the homicide
totals divided by the population totals to give the weighted regional homicide rates. The ICD 9
external cause data were also pooled by region and year, and for each region the percentages of
homicides due to firearms (E965), sharp instruments (E966) hanging and strangulation (E963) and
other causes (E960-962, E964 and E967-969) were calculated. As well as the cross-sectional
analyses, the data were sufficient to prepare regional trends in homicide rates and methods from 1985
to 1995 for Europe (both low- to middle- and high-income), the low- to middle and high-income
Americas, and the high-income Western Pacific.
5
Youth Homicide Data Availability by Region
The WHO regional and economic groupings used in this section are shown in footnote 1, which also
indicates the countries for which mortality data on youth homicides were available. There were no
data for Africa, and none for the low- to middle-income Eastern Mediterranean, India and the lowto middle-income Western Pacific. Under a third of the regional populations for the high-income
Eastern Mediterranean and low- to middle-income South East Asia were represented. These five
regions were therefore excluded from the subsequent comparisons.
Except for China where the mortality data were drawn from a population sample, over 70% of the
regional populations were represented by the mortality data for the remaining six regions included
in the analysis. These regions were the high income Americas (AMRO-H), the low- to middle
income Americas (AMRO-LM), high income Europe (EURO-H), low- to middle-income Europe
(EURO-LM), China, and the high income Western Pacific (WPRO-H).
Youth Homicide Rates by WHO Regional and Economic Groupings
Of the six regions with adequate mortality data, the highest youth homicide rates were in the lowto middle-income Americas, and the lowest in the high-income Western Pacific and European
regions. Table 1 shows the regional youth homicide rates for 10 to 24 year olds, indicators of intra
1 Countries in the five regions where data represented over 70% of the regional population are listed below, with
those returning data indicated by italic type.
Low- to midlle income Americas
Antigua and Barbuda, Argentina, Barbados, Belize, Bolivia, Brazil, Chile, Colombia, Costa Rica, Cuba, Dominica,
Dominican Republic, Ecuador El Salvador, Granada, Guatemala, Guyana, Haiti, Honduras, Jamaica, Mexico,
Nicaragua, Panama, Paraguay, Peru, Saint Kitts and Nevis, Saint Lucia, Saint Vincent and the Grenadines,
Suriname, Trinidad and Tobago, Uruguay, Venezuela.
High-income Americas
Bahamas, Canada, United States
Low- to middle-income Europe
Albania,Armenia, Azerbaijan, Belarus, Bosnia and Herzegovina, Bulgaria, Croatia, Czech Republic, Estonia,
Georgia, Hungary, Kazakhstan, Kyrgyzstan, Latvia, Lithuania, Malta, Poland, Republic ofMoldova, Romania,
Prussian Federation, Slovakia, Slovenia, Tajikistan, TFYR Macedonia, Turkey, Turkmenistan, Ukraine, Uzbekistan,
Yugoslavia.
High-income Europe
Andorra, Austria, Belgium, Denmark, Finland, France, Germany, Greece, Iceland, Ireland, Israel, Italy,
Luxembourg, Monaco, Netherlands, Norway, Portugal, San Marino, Spain, Sweden, Switzerland, United Kingdom.
High-income Western Pacific
Australia, Brunei Darussalam, Japan, New Zealand, Republic ofKorea, Singapore.
6
regional variations in rates, and the projected overall homicide rates for 1998, based on burden of
disease projections.
Table 1 shows that the youth homicide rate was over 10 per 100 000 population in three regions.
These high violence regions were the low- to middle- and high-income Americas, and low- to
middle-income Europe. Among these high-violence regions the rate in the low- to middle-income
Americas (24.9/100,000) was twice that in low- to middle-income Europe (10.1/100,000) and the
high-income Americas (12.8/100,000). In China the youth homicide rate was 3.1 per 100 000
population, while the high-income European and Western Pacific regions had rates of under two per
100,000.
Comparison of the youth homicide rates with the projected overall homicide rates indicates the extent
to which youth homicides drive the overall rate in a region. For the low- to middle-income Americas,
the high income Americas, low- to middle-income Europe, and China the ratio of youth homicides
to the overall rate was about equal, meaning that in these regions homicides among 10 to 24 year olds
contributed disproportionately to the overall homicide rate. In contrast, the ratio of youth homicide
to overall homicide rates in the high-income European and Western Pacific regions was around 0.2,
meaning that overall rates in these regions were not as greatly influenced by homicides in younger
age groups.
Youth Homicide Rates by Age and Sex
In all regions and age groups male youth homicide rates were higher than female rates, suggesting
that maleness is a universal risk factor for homicide in ages 10 to 24. Across all regions and both
sexes homicide rates increased sharply between each of the age ranges (i.e. 10 to 14, 15 to 19 and 20
to 24 years), suggesting that developmental changes (involving biological maturation and
socialisation) occurring at these ages are also a comon risk factor across diverse cultural settings.
While common to all regions, the size of these age and sex differences varied considerably between
them, as shown in Figure 1. These variations suggest that the impact of developmental changes on
youth homicide rates is amplified or reduced by region-specific social and environmental factors.
Homicide rates among 10 to 14 year olds were the lowest in all regions for both sexes, and varied
least between regions. This contrasted with the age ranges 15 to 19 and 20 to 24 years, where the
highest rates and the largest regional variations occurred. While male rates exceeded female rates in
all age groups and all regions, it is notable that male youth in the low violence, high-income
European and Western Pacific regions suffer homicide rates lower than the rates for female youth in
the high violence regions of the Americas and low- to middle-income Europe. The number of male
youth homicides for every female death was greatest in the three high violence regions (the two
Americas and low- to middle-income Europe) and smallest in the low violence regions (high income
Europe and the Western Pacific).
7
Methods of Youth Homicide
The methods of perpetrating youth homicide include firearms, knives and blunt instruments, hanging
and strangulation, and an array of ‘other’ methods for inflicting physical injury such as poisoning
and pushing from heights. For all regions, over 60% of all youth homicides in males and females
were accounted for by firearms; sharp instruments; and hanging/strangulation. As shown in Figure
2, there were substantial differences by region and sex in the distribution of these methods.
Figure 2 shows that firearms were the most frequent method used in male and female homicides for
the high-income Americas and both European regions, although they were only marginally more
common than sharp instrument deaths among females in high-income Europe. By contrast, firearms
accounted for the smallest percentage of homicides among females in the low violence, high-income
Western Pacific, and the second lowest percentage among their male counterparts in the region.
While the high violence, high-income Americas were distinguished by over 80% of male and 60%
of female youth homicides being inflicted by firearms, deaths in high violence, low- to middle
income Europe were more evenly spread between the different causes. The causal profile for males
in low- to middle-income Europe was nearly identical to that for high-income Europe, with firearms
accounting for around 46% of youth homicides, sharp objects 31%, and other causes 20% of deaths
in both regions. The profiles for females in both regions were also very similar, with the exception
of an elevated proportion of firearm deaths among females in low- to middle-income Europe.
Trends in Youth Homicide Rates and External Causes
Trend lines showing youth homicide rates were plotted for the low- to middle-income Americas,
high-income Americas, high-income Western Pacific, and both European regions for the years
froml985 to 1995.
High-income Americas. Trends for the high income Americas suggest that youth homicide victims
are becoming younger, with rates increasing more rapidly in 15 to 19 year olds than in 20 to 24 year
olds (Figure 3). Homicide rates among 15 to 19 year old males nearly doubled between 1987 (9.2)
and 1991 (17.9), while among 20 to 24 year old males there was a less pronounced increase from
16.4 in 1987 to 23.0 in 1991. Thereafter rates in both these high-incidence male groups stabilised
until 1993, and from 1994 began to decline.
Firearm-related homicides in 10 to 24 year olds increased from around 65% of all youth homicides
in 1985 to over 85% in 1993, and then decreased slightly to around 80% in 1995. This trend for the
percentage of firearm deaths closely matched that for homicide rate changes in 15 to 19 and 20 to 24
year olds, and was mirrored by the curve for homicides due to sharp instruments, which decreased
in proportion to the increase in firearm inflicted deaths.
Low- to Middle Income Americas. Rates among 15 to 19 and 20 to 24 year old males and females
increased from already high levels in 1985 to extremely high levels in the early 1990s (Figure 4). In
8
20 to 24 year old males, homicide rates increased from just over 60 in 1987 to around 90 per 100,000
for the years 1991 to 1994. Rates in 15 to 19 year old males followed a similar pattern of increase,
climbing from around 30 in 1985 to over 50 per 100,000 in the period 1991 to 1994. In females, the
homicide rate increase among 20 to 24 year olds (from under five in 1985 to over six between 1991
and 1994) was nearly overtaken by the more rapid increase in 15 to 19 year olds, among whom the
homicide rate nearly doubled between 1985 and 1991. Among 10 to 14 year olds, homicide rates in
females doubled from 0.7 in 1984 to 1.4 in 1994, while for males in this age group the increase was
more gradual, from 2.5 in 1985 to 3.8 in 1994.
High-income Europe. Male and female rates in high-income Eourope were distinguished by their
consistency over time (Figure 5). Rates for males aged 20 to 24 varied between a 1986 low of 2.0 and
a 1992 high of 2.8, and apart from the period 1991 to 1993 showed no sustained upward trend. Rates
in males aged 15 to 19 remained between one and 1.5, and in 10 to 14 year olds hovered around 0.3
per 100 000. Female youth homicide rates were similarly consistent, varying for 20 to 24 year olds
between 0.8 and 1.2, for 15 to 19 year olds between 0.5 and 0.7, and for 10 to 14 year olds remaining
at approximately 0.3 per 100 000.
The trend for external causes showed that although firearm-related cases accounted for the largest
proportion of deaths in all years, they exceeded 50% of all cases only in 1989 and 1991, and in most
years accounted for just over 40% of youth homicides. Sharp instruments were the next most
prominent cause (accounting for between 20% and 30% of cases), and the proportion of deaths
involving sharp instruments increased as firearm-related deaths decreased, and decreased as firearmrelated deaths increased.
Low- to middle-income Europe. Low- to middle-income Europe is composed mainly of the ex
soviet transition societies, and its youth homicide trends were dominated by a rapid increase in rates
that coincided with the late 1980s collapse of communism (Figure 6). This increase began with a
gradual rise between 1988 and 1990, followed by a very steep increase from 1991 to 1994. For males
aged 20 to 24 there was a four-fold increase in rates between 1988 and 1994 (11.6 to 48.7), and for
males aged 15 to 19 a five-fold increase (4.9 to 23.7). Thereafter male homicide rates in these two
oldest age groups began declining. In 10 to 14 year old males there was a gradual homicide rate
increase from 0.7 in 1985 to 1.84 in 1994, followed by a decline to 1.4 in 1995. For females aged 20
to 24 homicide rates nearly tripled from 2.6 in 1988 to 7.1 in 1995 - the post-1994 decline seen in
males of this age range was therefore not evident for females. For females aged 15 to 19 rates
increased from 2.0 in 1988 to 4.4 in 1994, and then declined slightly. As for males, female homicide
rates in 10-14 year olds increased steadily, from 0.4 in 1995 to 0.9 in 1995.
The trend for homicide methods in low- to middle-income Europe shows that the steep escalation
in homicide rates from 1991 onwards coincided with a rapid decrease in the percentage of sharp
instrument related deaths and a corresponding rise in the proportion of homicides involving firearms.
From 1985 to 1995, the percentage of youth homicides inflicted by sharp instruments decreased from
60% to under 30%, while the proportion involving firearms more than doubled from 20% to 50%.
9
High-income Western Pacific. As for high-income Europe, the trends for both males and females
were distinguished by consistently low rates (Figure 7). In the 20 to 24 year age range, male rates
fluctuated between 1.3 and 1.8, and female rates between 0.8 and 0.9 per 100 000. In 15 to 19 year
olds, male rates ranged from 0.7 to 1.3, and female rates vary between 0.3 and 0.6. The lowest rates
for males and females occurred in 10 to 14 year olds, and for both sexes hovered around 0.4 per 100
000.
The trend for external causes shows that the proportion of firearm-related deaths decreased from
around 16% between 1985 and 1987, to under 10% between 1991 and 1993, before again moving
slightly upward to just under 15% in 1995. This was complemented by an increase in the percentage
of youth homicides inflicted by sharp instruments, which rose from 25% in 1985 to nearly 40% in
1995, and decrease in the proportion of hanging and strangulation deaths.
Other Studies of Youth Homicide
Epidemiological findings on youth homicide in countries and regions inadequately covered by actual
mortality returns were scant. Almost all the studies identified were for Africa, and because few
studies were explicitly concerned with documenting youth homicide, the information was often
sketchy.12’22 There were some common epidemiologic patterns evident across these studies (Table
2) First, there was a steep increase in homicide rates from adolescence to young adulthood. Second,
studies that mentioned the sex of victims showed a marked preponderance of male over female
victims. Third, the studies where rates were estimated suggested substantial variation in youth
homicide rates between countries and regions. For example, the rates for South Africa and Tanzania
placed these countries among the high violence regions, while the rates for Fiji and Taiwan placed
those countries among the low violence regions.
Conclusions
The major barrier to describing the global epidemiology of youth homicide is insufficient
surveillance data for some of the low- to middle-income regions (especially Africa and the Eastern
Mediterranean) where the highest projected rates occur. Our knowledge of age- and sex-specific
patterns and trends is thus biased towards the high-income regions that tend to have much lower'
rates. Concerted effort is therefore needed to eliminate these knowledge gaps by developing violence
surveillance systems in Africa, parts of the low- to middle-income Americas, the eastern
Mediterranean, south east Asia, India, and the low-to middle-income Western Pacific.
Within the knowledge limits imposed by these biases in data availability, the cross-sectional and
trend analyses showed dramatic variations in rates and the distribution of homicide methods within
and between regions, confirming that high youth homicide rates are by no means inevitable. Two rate
patterns were consistent across all regions and years and can therefore be considered as reflecting
10
universal risk factors. First the preponderance of male over female victims, and, second, the
occurrence in males and females of steady increases in homicide rates from 10 to 24 years of age.
Crosscutting these universal patterns were region and time specific variations in three areas. First,
the ratio of male incidence to female incidence was highest in the high-violence regions and lowest
in the low-violence regions. Second, the magnitude of the increase from mid-adolescent to early
adulthood was different across regions. While the relative increase in homicide rates at early
adulthood (between 15 to 19 and 20 to 24 years) was very similar in all regions, the mid-adolescent
increase between 10 to 14 and 15 to 19 years was substantially greater in the high violence regions
than in the low violence regions. This means that the main contributor to regional differences in
youth homicide levels is the rate among 15 to 19 year olds, which, as shown by the trend analyses
for the high violence regions, also showed a more rapid increase in homicide rates from 1985 to 1995
than the other age groups. Thirdly, the trend analyses highlighted the acute effect of massive and
sudden political change on homicide rates in low-to middle-income Europe, as compared to the large
but less abrupt increases in youth homicide rates for the high and low- to middle-income Americas,
both of which indicate a substantial change in risk factors.
The trends in homicide methods showed that high violence countries and years were associated with
a high proportion of firearm related homicides. This was most clearly apparent in low- to middle
income Europe, where the percentage of youth homicides inflicted by firearms increased from 20%
in 1985 (when the homicide rate was 12) to 50% in 1995 when the homicide rate was 35. In both
high violence regions of the Americas firearms accounted for over 70% of youth homicides in all
years, as against averages of 45% and 15% in the low violence regions of high-income Europe and
the high-income Western Pacific respectively.
Predictors, Causes, and Correlates of Youth Violence
We now review the scientific literature on the possible causes of youth violence. This review is
focused on those factors that can change over time in the interest of shedding light on possible
prevention strategies. We review this literature using a conceptual framework sensitive to the factors
that support or impede healthy human development and the environmental context in which such
development occurs.3
This literature review is divided into two sections. Because we are interested in the strongest
evidence for causes of youth violence the first section focuses on longitudinal studies of large
community samples. Longitudinal studies allow for the measurement of risk factors prior to the
occurrence of violence, which is an essential form of evidence for establishing whether a risk factor
is a predictor or possible cause. Unfortunately, virtually all of the longitudinal research studies of
youth violence are based on samples from Western societies (Canada, Denmark, Finland, Great
Britain, New Zealand, Sweden, United States).24 Consequently, it is difficult to determine, based on
extant research, whether what we currently know about the causes of youth violence can be
generalized beyond Western societies. Also most of this research is based on males. This part of the
11
literature review focuses primarily on factors that influence violent behavior at the individual, family,
peer, and neighborhood levels.
The second section of the literature review focuses on the influence of macro-level factors such as
a nation’s socioeconomic status, the nature and stability of it’s political system, and the influence of
persistent exposure to civil disturbances and war on youth homicide rates. Some of the studies
reviewed in this section may not focus on violence as a juvenile phenomenon, but they do address
factors that may affect youth as a key risk group. The influence of macro-level factors on youth
violence is an important complement to the literature that addresses individual, family, peer, and
neighborhood level risk and protective factors.
Individual, Family, Peer, and Neighborhood Influences on Youth Violence
Violent offenses, like other crimes, arise from interactions between offenders and victims in
situations. Some violent acts are probably committed by youth with relatively stable and enduring
violent tendencies, while others are committed by more ’’normal" youth who find themselves in
situations that are conducive to violence. This section summarizes knowledge about the development
of violent persons (i.e. persons with a relatively high probability of committing violent acts in any
situation) and the occurrence of violent acts. Risk factors for violence are defined as factors that
predict a high probability of violence. In order to determine whether a risk factor (e.g. school failure,
poor parental supervision, delinquent peers) is a predictor or possible cause of violence, the risk
factor needs to be measured before the violence. Hence, longitudinal follow-up studies are needed.
In the interests of throwing light on possible causes of violence and prevention methods, the
emphasis is on risk factors that can change over time. Thus, genetic factors that are fixed at birth,
such as the XYY chromosome abnormality, are not discussed, but biological factors that can change,
such as the resting heart rate, are included. The main focus in this section is on individual level
studies as opposed to aggregate level ones (e.g. of rates of violence in different areas), and on violent
offenders rather than victims of violence. However it should be noted that victims of violence
overlap significantly with violent offenders.11 In Pittsburgh, boys who were killed or injured by guns
were particularly likely to have sold drugs, carried a gun themselves, and been involved in a gang
fight.12
Specialization or Versatility. Generally, young violent offenders tend to be versatile rather than
specialized. They tend to commit many different types of crimes and also show other problems such
as truancy and school dropout, substance use, persistent lying, and sexual promiscuity. However,
there is a small degree of specialization in violence superimposed on this versatility.25
As an indication of their versatility, violent youth typically commit more nonviolent offenses than
violent offenses. In the Cambridge Study, the convicted violent delinquents up to age 21 had nearly
three times as many convictions for nonviolent offenses as for violent offenses.26 In the Oregon
Youth Study, which is a longitudinal survey of over 200 boys from age 10, the boys arrested for
12
violence had an average of 6.6 arrests of all kinds.27 Generally, violent males have an early age of
onset of offending of all types.28 Both in official records29 and self-reports,30 an early age of onset
of violent offending predicts a relatively large number of violent offenses.
Hamparian and her colleagues29 identified all (811) youth bom in 1956-58 and arrested for violence
as juveniles in Columbus, Ohio. These youth were arrested for 2282 nonviolent offenses and 1091
violent offenses, including 12 murders, 40 rapes, 255 robberies, and 466 assaults. The murderers had
6.3 arrests of all kinds, compared with 5.6 for the rapists, 5.4 for the robbers, and 4.5 for the
assaulters. There was a considerable amount of versatility in violent offending. For example,
comparing the first and second violent arrests: 48% of the murder/assault first arrest cases committed
robbery/rape on the second arrest; 42% of the robbery first arrest cases committed
murder/assault/rape on the second arrest; and 57% of the rape first arrest cases committed
murder/assault/robbery on the second arrest.
Continuity. In general, there is continuity from juvenile to adult violence and from childhood
aggression to youth violence. In the Columbus study, 59% of violent juveniles were arrested as adults
in the next 5 to 9 years, and 42% of these adult offenders were charged with at least one Index
(serious) violent offense.31 More of those arrested for Index violence as juveniles were rearrested
as adults than of those arrested for minor violence (simple assault or molesting) as juveniles: 63%
compared with 53%. In the Cambridge Study, one-third of the boys convicted of violence between
ages 10 and 20 were reconvicted of violence between ages 21 and 40, compared with only 8% of
those not convicted of youth violence/2
Childhood aggression predicts youth violence. In the Orebro (Sweden) follow-up of about 1000
youth/3 two-thirds of boys who were officially recorded for violence up to age 26 had high
aggressiveness scores at ages 10 and 13 (rated by teachers), compared with 30% of all boys.
Similarly, in the Jyvaskyla (Finland) follow-up of nearly 400 youth,34 peer ratings of aggression at
ages 8 and 14 significantly predicted officially recorded violence up to age 20.
One possible explanation of the continuity over time is that there are persisting individual differences
in an underlying potential to commit aggressive or violent behaviour. In any cohort, the people who
are relatively more aggressive at one age also tend to be relatively more aggressive at later ages, even
though absolute levels of aggressive behaviour and behavioral manifestations of violence are
different at different ages. There may also be developmental sequences or pathways over time from
one type of aggression to another. For example, in the Pittsburgh Youth Study, which is a follow-up
of over 1,500 Pittsburgh boys originally studied at ages 7, 10 and 13, Loeber and his colleagues
reported that childhood aggression (e.g. bullying) escalated into gang fighting and later into youth
violence/5
Biological Risk Factors. According to Raine/6 one of the most replicable findings in the literature
is that antisocial and violent youth tend to have low resting heart rates. This can be easily
demonstrated by taking pulse rates. The main theory underlying this finding is that a low heart rate
indicates low autonomic arousal and/or fearlessness. Low autonomic arousal, like boredom, leads
to sensation-seeking and risk-taking in an attempt to increase stimulation and arousal levels.
13
Conversely, high heart rates, especially in infants and young children, are associated with anxiety,
behavioral inhibition, and a fearful temperament.37
In the British National Survey of Health and Development,38 which is a prospective longitudinal
survey of over 5300 children bom in England, Scotland, or Wales in 1946, heart rate was measured
at age 11. A low heart rate predicted convictions for violence and sexual offenses up to age 21; 81 %
of violent offenders and 67% of sexual offenders had below-average heart rates. There was an
interaction between heart rate and family background. A low heart rate was especially characteristic
of boys who had experienced a broken home before age 5, but among these boys it was not related
to violence or sexual offenses. A low heart rate was significantly related to violence and sexual
offenses among boys who came from unbroken homes.
In the Cambridge Study, resting heart rate was measured at age 18 and was significantly related to
convictions for violence and to self-reported violence at age 18, independently of all other variables.39
More than twice as many of the boys with low heart rates (65 beats per minute or less) were
convicted for violence as of the remainder.
Perinatal (pregnancy and delivery) complications have been studied, because of the hypothesis that
they might lead to neurological damage, which in turn might lead to violence. In a Danish perinatal
study, Kandel and Medmck40 followed up over 200 children bom in Copenhagen in 1959-61. They
found that delivery complications predicted arrests for violence up to age 22; 80% of violent
offenders scored in the high range of delivery complications, compared with 30% of property
offenders and 47% of nonoffenders. However, pregnancy complications did not significantly predict
violence.
Interestingly, delivery complications especially predicted violence when a parent had a history of
psychiatric illness; in this case, 32% of males with high delivery complications were arrested for
violence, compared with only 5% of those with low delivery complications.41 Unfortunately, these
results were not replicated by Denno42 in the Philadelphia Biosocial Project, which is a follow-up of
nearly 1000 African American births in Philadelphia in 1959-62. It may be that pregnancy and
delivery complications predict violence only or mainly when they occur in combination with other
family adversities. Interactions between biological and psychosocial factors are quite common.
Psychological/Personality Factors. Among the most important personality dimensions that predict
youth violence are hyperactivity, impulsiveness, poor behavioral control, and attention problems.
Conversely, nervousness and anxiety are negatively related to violence. In the Dunedin (New
Zealand) follow-up of over 1,000 children, ratings of poor behavioral control (e.g. impulsiveness,
lack of persistence) at age 3-5 significantly predicted boys convicted of violence up to age 18,
compared to those with no convictions or with nonviolent convictions.43 In the same study, the
personality dimensions of constraint (e.g. cautiousness, avoiding excitement) and negative
emotionality (e.g. nervousness, alienation) at age 18 were significantly correlated with convictions
for violence.44
Many other studies show linkages between these personality dimensions and youth violence. In the
14
Copenhagen perinatal project, hyperactivity (restlessness and poor concentration) at age 11-13
significantly predicted arrests for violence up to age 22, especially among boys experiencing delivery
complications.41 More than half of those with both hyperactivity and high delivery complications
were arrested for violence, compared to less than 10% of the remainder. Similarly, in the Orebro
longitudinal study in Sweden, hyperactivity at age 13 predicted police-recorded violence up to age
26. The highest rate of violence was among males with both motor restlessness and concentration
difficulties (15%), compared to 3% of the remainder.45
Similar results were obtained in the Cambridge and Pittsburgh studies.10 High daring or risk-taking
at age 8-10 predicted both convictions for violence and self-reported violence in the Cambridge
Study. Poor concentration and attention difficulties predicted convictions for violence in the
Cambridge Study and reported violence (by boys, mothers, and teachers) in Pittsburgh. High
anxiety/nervousness was negatively related to violence in both studies, and low guilt significantly
predicted court referrals for violence in the Pittsburgh study.
The other main group of psychological factors that predict youth violence include low intelligence
and low school attainment. In the Philadelphia Biosocial Project,42 low verbal and performance IQ
at ages 4 and 7, and low scores on the California Achievement Test at age 13-14 (vocabulary,
comprehension, maths, language, spelling) all predicted arrests for violence up to age 22. In Project
Metropolitan in Copenhagen,46 which is a follow-up study of over 12,000 boys bom in 1953, low
IQ at age 12 significantly predicted police-recorded violence between ages 15 and 22. The link
between low IQ and violence was strongest among lower class boys.
Similar results were obtained in the Cambridge and Pittsburgh studies.10 Low nonverbal IQ at age
8-10 predicted both official and self-reported violence in the Cambridge Study, and low school
achievement at age 10 predicted official violence in both studies. The extensive meta-analysis by
Lipsey and Derzon47 also showed that low IQ, low school attainment, and psychological factors such
as hyperactivity, attention deficit, impulsivity, and risk-taking were quite important predictors of later
serious and violent offending.
Impulsiveness, attention problems, low intelligence, and low attainment could all be linked to deficits
in the executive functions of the brain, located in the frontal lobes. These executive functions
include sustaining attention and concentration, abstract reasoning and concept formation, goal
formulation, anticipation and planning, programming and initiation of purposive sequences of motor
behaviour, effective self-monitoring and self-awareness of behavior, and inhibition of inappropriate
or impulsive behaviors.48 Interestingly, in the Montreal longitudinal-experimental study, which is
a follow-up of over 1,100 children from age 6, a measure of executive functions based on cognitiveneuropsychological tests at age 14 was the strongest neuropsychological discriminator of violent and
nonviolent boys.49 This relationship held independently of a measure of family adversity (based on
parental age at first birth, parental education level, broken family, and low socioeconomic status).
Family Factors. Numerous family factors predict violence. In her follow-up of 250 treated Boston
boys in the Cambridge-Somerville Study, McCord'0 found that the strongest predictors at age 10 of
later convictions for violence (up to age 45) were poor parental supervision, parental aggression,
15
including harsh, punitive discipline, and parental conflict. An absent father was almost significant
as a predictor, but tlie mother’s lack of affection was not significant. She also demonstrated that
fathers convicted for violence tended to have sons convicted for violence?1 In later analyses, she
showed that violent offenders were less likely than nonviolent offenders to have experienced parental
affection and good discipline and supervision, but equally likely to have experienced parental
conflict.52
Similar results have been obtained in other studies. In the Chicago Youth Development Study,53
which is a longitudinal follow-up of nearly 400 inner-city boys initially studied at age 11-13. poor
parental monitoring and low family cohesion predicted self-reported violent offending. Also, poor
parental monitoring and low attachment to parents predicted self-reported violence in the Rochester
Youth Development Study,54 which is a longitudinal study of nearly 1,000 children originally studied
at age 13-14. Broken families between birth and age 10 predicted convictions for violence up to age
21 in the British National Survey,55 and single parent status at age 13 predicted convictions for
violence up to age 18 in the Dunedin study.40 Parental conflict and a broken family predicted official
violence in the Cambridge and Pittsburgh studies, and coming from a single-parent female-headed
household predicted official and reported violence in Pittsburgh.10
Harsh physical pumshment by parents, and child physical abuse, typically predict violent offending
by sons.' Harsh parental discipline predicted official and self-reported violence in the Cambridge
Study. 0 In a follow-up study of nearly 900 children in New York State, Eron and his colleagues57
reported that parental punishment at age 8 predicted not only arrests for violence up to age 30, but
also the severity of the man's pumshment of his child at age 30 and also his history of spouse assault.
In a longitudinal study of over 900 abused children and nearly 700 controls, Widom discovered that
recorded child physical abuse and neglect predicted later arrests for violence, independently of other
predictors such as gender, ethnicity, and age.58 In the Rochester Youth Development Study, Smith
and Thomberry showed that recorded childhood maltreatment under age 12 predicted self-reported
violence between ages 14 and 18, independently of gender, ethnicity, socioeconomic status, and
family structure?9
Large family size (number of children) predicted youth violence in both the Cambridge and
Pittsburgh studies.10 In the Oregon Youth Study,27 large family size at age 10 predicted self-reported
violence at age 13-17. Young mothers (mothers who had their first child at an early age, typically
as a teenager) also tend to have violent sons, as Morash and Rucker60 demonstrated in the Cambridge
Study for the prediction of self-reported violence at age 16. Interestingly, the relationship between
a young mother and a convicted son in this study disappeared after controlling for other variables,
notably large family size, a convicted parent, and a broken family.61 A young mother also predicted
official and reported violence in the Pittsburgh Youth Study.10
Peer, Socioeconomic, and Neighborhood Factors. Having delinquent friends is an important
predictor of youth violence; peer delinquency predicted self-reported violence in the Rochester
Youth Development Study.54 A mental health survey administered to 221, 12 to 17 year old
residents of a marginal neighborhood in Lima, Peru, found that having a friend that consumed drugs
was associated with violent behavior.62 What is less clear is how far the link between delinquent
16
friends and delinquency is a consequence of co-offending,63 which is particularly common under age
21. Elliott and Menard concluded both that delinquency caused delinquent peer bonding and that
delinquent peer bonding caused delinquency.64 However, there seems to be no information
specifically about the link between peer violence and youth violence.
In general, coming from a low socioeconomic status (SES) family predicts youth violence. For
example, in the National Youth Survey, the prevalences of self-reported felony assault and robbery
were about twice as high for lower class youth as for middle class ones.65 In the Lima study, low
maternal education and housing density were found to associated with youth violence while under
the influence of alcohol.62 Gianini, Litvoc and Neto66 found, in an emergency-room based study of
young adults in Sao Paulo, Brazil, that after adjusting for sex and age, the risk of victimization was
significantly higher for the subproletariat, with an Odds Ratio of 4.2 and 95% CI of 1.99-8.84.
Similar results have been obtained for official violence in Project Metropolitan in Stockholm,67 in
Project Metropolitan in Copenhagen,46 and in the Dunedin Study in New Zealand.43 Interestingly,
all three of these studies compared the SES of the family at the boy's birth, based on the father's
occupation, with the boy's later violent crimes. The strongest predictor of official violence in both
the Cambridge and Pittsburgh studies was family dependence on welfare benefits.10
Generally, boys living in urban areas are more violent than those living in rural ones. In the US
National Youth Survey, the prevalence of self-reported felony assault and robbery was considerably
higher among urban youth.65 Within urban areas, boys living in high crime neighborhoods are more
violent than those living in low crime neighbourhoods. In the Rochester Youth Development Study,
living in a high crime neighborhood significantly predicted self-reported violence.54 Similarly, in the
Pittsburgh Youth Study, living in a bad neighborhood (either as rated by the mother or based on
census measures of poverty, unemployment, and female-headed households) significantly predicted
official and reported violence.10
In the United States the availability of guns, gangs and drugs in a neighborhood are important risk
factors for youth violence.67,68 The number of arrests ofjuveniles for homicide in the United States
more than doubled between 1984 and 1993 (from 5.4 to 14.5 per 100,000 population).69 Blumstein70
suggested that this increase was linked to concurrent increases in gun carrying, gangs, and battles
over crack cocaine selling. In the Rochester Youth Development Study, about 30% of the sample
were gang members, but they accounted for about 70% of violent crimes and 70% of drug selling.
The incidence of violence increased after joining a gang and decreased after leaving it. More than
half of those in gangs said that they owned guns for protection.72 In the Pittsburgh Youth Study,
initiation into drug selling coincided with a significant increase in weapon carrying, and 80% of those
selling hard drugs at age 19 were carrying a gun.73 In Rio de Janerio, Brazil, where the majority of
homicide victims and aggressors are 25 years of age and younger, drug trafficking is responsible for
a high proportion of homicides, conflicts and injuries.74
Situational Factors. It might be argued that all the risk factors reviewed so far in this section biological, psychological/personality, family, peer, socioeconomic, and neighborhood — essentially
influence the development of a long-term individual potential for violence. In other words, they
contribute to between-individual differences: why some people are more likely than others, given the
17
same situational opportunity, to commit violence. Another set of influences -- situational factors —
explain how the potential for violence becomes the actuality in any given situation. Essentially, they
explain short-term within-individual differences: why a person is more likely to commit violence in
some situations than in others. Situational factors may be specific to particular types of crimes:
robberies as opposed to rapes, or even street robberies as opposed to bank robberies. One of the most
influential situational theories of offending is routine activities theory.75 This suggests that, for a
predatory crime to occur, the minimum requirement is the convergence in time and place of a
motivated offender and a suitable target, in the absence of a capable guardian.
Much work on describing situations leading to violence has been carried out in Great Britain under
the heading of crime analysis.76 This begins with a detailed analysis of patterns and circumstances
of crimes and then proceeds to devising, implementing, and evaluating crime reduction strategies.
For example, Barker and her colleagues77 analyzed the nature of street robbery in London. Most of
these crimes occurred in predominantly ethnic minority areas, and most offenders were 16-19 year
old Afro-Caribbean males. The victims were mostly Caucasian females, alone, and on foot. Most
offenses occurred at night, near the victim’s home. The main motive for robbery was to get money,
and the main factor in choosing victims was whether they had a wealthy appearance.
In their Montreal longitudinal study of delinquents, LeBlanc and Frechette provided detailed
information about motives and methods used in different offenses at different ages. For example,
for violence at age 17, the main motivation was utilitarian or rational. For all crimes, however, the
primary motivation changed from hedonistic (searching for excitement, with co-offenders) in the
teenage years to utilitarian (with planning, psychological intimidation, and use of instruments such
as weapons) in the twenties.79 In the National Survey of Youth, which was a cross-sectional survey
of nearly 1400 American youth aged 11-18, assaults were usually committed for retaliation or
revenge or because of provocation or anger.80
In the Cambridge Study, motives for physical fights depended on whether the boy fought alone or
with others. In individual fights, the boy was usually provoked, became angry, and hit out to hurt
his opponent and to discharge his own internal feelings of tension. In group fights, the boy often said
that he became involved to help a friend or because he was attacked, and rarely said that he was
angry. The group fights were more serious, occurring in bars or streets, and they were more likely
to involve weapons, produce injuries, and lead to police intervention. Fights often occurred when
minor incidents escalated, because both sides wanted to demonstrate their toughness and masculinity
and were unwilling to react in a conciliatory way.
Many of the boys in the Cambridge Study fought after drinking alcohol, and it is clear that alcohol
intoxication is an immediate situational factor that precipitates violence. In a Swedish study, about
three-quarters of violent offenders and about half of the victims of violence were intoxicated at the
time.82 Conventional wisdom suggests that alcohol consumption has a disinhibiting effect on
behavior that encourages both offending
and victimization. However, the biological links between
on
alcohol and violence are complex.
Behaviors leading up to violence have been studied. Wolfgang classified actions leading to homicide
18
in Philadelphia based on police records.84 Most commonly, homicides arose from trivial altercations
(insults or jostling), domestic quarrels, jealousy, or altercations over money. Similarly, violent
offenses in London usually arose from family disputes or quarrels between neighbors or persons
working together.85 In Sweden, most violent crimes were preceded by arguments, either arising out
of the situation or based on existing social relationships.82 However, in all these studies, a minority
of violent acts were basically unprovoked attacks or robberies. Pallone and Hennessy86 referred to
"tinderbox criminal violence”, defined as violence occurring between similar types of people, known
to each other, ostensibly to settle long-lasting or emerging disputes.
Much is known about the situations in which violence occurs.87 For example, in the Swedish study,
violence preceded by situational arguments typically occurred in streets or restaurants, while violence
preceded by relationship arguments typically occurred in homes.82 In England, stranger assaults
typically occurred in streets, bars, or discotheques, nonstranger assaults typically occurred at home
or work, and robberies typically occurred in the street or on public transport.88 Violence in public
places could be investigated using systematic observation, for example recording incidents from
closed-circuit television cameras mounted on buildings. More research on situational influences on
violent acts needs to be incorporated in prospective longitudinal studies, in order to link up the
developmental and situational perspectives.
Macro-level Influences on Youth Violence
A number of factors that operate at the societal level may increase the vulnerability of nations or
regions to violence, a vulnerability which may have a particularly important effect on adolescent and
young adult populations. These macro-level conditions include the ways in which our political,
economic, and social systems are structured as well as the changes that occur in these systems.89 This
section reviews the literature on the ways in which these macro-level factors may contribute to the
development of social contexts conducive to violence and shape the development of violent groups
and persons.
Studies of the macro-level determinants of violence use a variety of methodologies ranging from
case studies to ecological analyses using econometric modeling or cross-national comparisons of
secondary data. These types of study designs are not as useful as the longitudinal studies of
individuals reviewed earlier for establishing causal relationships. Nevertheless, they are very useful
for identifying potentially important associations and explanations for how the social context and
changes in it may influence violent behavior.
Economic Conditions. A key set of macro-level risk factors for violence includes those factors that
reflect the economic structure and condition of a society. Several studies have demonstrated a link
between economic growth, income inequality, and violence. Fajnzylber, Lederman and Loayza90
found, in their empirical, cross-country study of 45 developed and developing countries (1965-95),
that income inequality, as measured by the Gini coefficient, had a significant and positive effect on
the homicide rate. The gross domestic product (GDP) growth rate also had a significantly negative
19
effect on the homicide rate, but it was mediated by the level of income inequality, and its impact is
lessened when income inequality was higher.
The presence of an illicit drug trafficking industry as significant component of the economy of a
nation appears to play an important role in contributing to violence. Fajnzylber, Lederman and
Loayza found a positive and significant correlation between countries considered significant
producers of illicit drugs and homicide rates.
Political Structure. The quality and level of state governance, both in terms of its legal framework
and its social protection policies, is a potentially important determinant of violence. In particular,
the extent to which a society enforces it’s laws through arrest and prosecution of violent offences may
act as a deterrent against violence or serve to incapacitate violent offenders from committing further
offences. Fajnzylber, Lederman and Loayza90 found that the homicide arrest rate had a significant
negative effect on homicide rate. In their study objective measures of governance, in particular, arrest
rates, had significant negative effects on crime rates, while more subjective measures (Rule of Law
and Voice and Accountability indices) had weak effects.
The quality of State governance however, can have an impact on violence, particularly as it affects
young people. Noronha et al.91 concluded in their study of the types of violence that affected different
ethnic groups in Bahia, Brazil, that dissatisfaction with police, justice and prisons influenced the
generation and reproduction of authoritarism and the use of private modes of justice, de Souza
Minayo74 found that one of the key forms of violence against youth in Rio de Janeiro was police use
of force, including violent searches, physical violence, sexual abuse and rape, and bribery, especially
targeted against boys from lower socioeconomic strata. Sanjuan92 included the sense of class
dependent justice as a key factor in the creation of a “culture of violence” among the marginalized
youth population in Caracas. Lack of access to regular forms of administrative justice may contribute
to the formation of youth groups according to normative and protective codes. The expression of
certain forms of violence may become routine, according to the public perception of justice and to
the sense of absence of legal limits. For example, Aitchinson93 concluded that post-apartheid South
Africa impunity for former human rights abusers and the inability of the police as an institution to
shift towards a democratic society, have contributed to the perception of societal insecurity and to
the risk of extrajudicial actions involving violence.
Another aspect of governance concerns social protection by the State. Pampel and Gartner94 used a
scale of collectivism (level of development of national institutions for collective social protection)
to explain why countries with equal percentage increases in the proportion of the population aged 15
to 29 showed unequal increases in homicide rates. They concluded that national institutions for
collective social protection reduce the likelihood that an increase in the 15-29 year-old population
will be accompanied by an increase in homicide rates. Messner and Rosenfeld95 examined the effects
of political efforts (which they refer to as "decommodification”, measure developed based on levels
and patterns of welfare expenditures) to insulate personal well being from market forces such as
economic recessions. Higher welfare expenditures were negatively related to homicide rates,
suggesting that societies where individuals are protected from market forces have less homicide.
20
Social integration. Social integration refers to the extent to which social networks and support
systems exist which interweave family, community, organisational, and institutional structures. From
a structural perspective, Gartner6 operationalised social integration as the divorce rate per 1 000
marriages and Cutright and Briggs7 as the divorce rate per 10 000 men aged 15 to 64. Societal
integration was also measured by the percent of the population belonging to distinct ethnic or
linguistic groups (an index of ethnic-linguistic fractionalisation). In both studies increased ethniclinguistic heterogeneity was associated with increased homicide rates, but more clearly so for adults
than children. The divorce rate was positively associated with homicide rates in Gartner6 and Cutright
and Briggs,7 although only in the latter with homicide death rates among 1 to 4 year olds. Fajnzylber,
Lederman and Loayza90 found ethnic-linguistic fractionalisation to be positively associated with
homicide rates, independently of income inequality.
Social capital is also a measure of the level of social integration in a society or community. Social
capital is defined as “the rules, norms, obligations, reciprocity and trust embedded in social relations,
social structures, and society’s institutional arrangements which enables its members to achieve their
individual and community objectives.”96 Lederman, Loayza and Menendez96 found, in a study of the
effect of social capital on incidence of crime in 39 developed and underdeveloped countries (198094), that the prevalence of trust among community members had a strong and significant effect on
reducing the incidence of violent crimes. Wilkinson, Kawachi and Kennnedy97 showed that social
capital scores indicating low social cohesion and high levels of interpersonal distrust were associated
with elevated homicide rates and increased economic inequality.
The lack of social capital may hinder young people from reaping the benefits of human capital
investments: youth who lack social capital in neighborhoods tend to do poorly in school and have an
increased probability of dropping out.98 Moser and Holland99 conducted a Participatory Urban
Appraisal of five communities representative from Jamaica's poor urban areas. They found a cyclical
relationship between violence and the destruction of social and human capital. Community violence
created an "area stigma", restrained physical mobility and employment and educational opportunities
due to fear, made businesses reluctant to invest in the community, and made neighbors less likely to
invest in local infrastructure. The absence of social capital and increased stress and mistrust resulting
from the erosion of appropriate infrastructure, social space and opportunities was perceived to
increase the likelihood of violent behavior, especially among young people.
The Cultural Context. Culture is reflected in the system of norms and values that are passed across
generations in any society. These norms and values play an important role in how members of a
society respond to their environment and changes in it. The cultural context can influence violence,
for example, by prescribing it as a method of conflict resolution and socializing youth to adopt
norms and values that support violent behavior. Cultural norms and the process of socialization are
influenced by conditions at the social and community levels.
Indicators of officially approved violence, such as involvement in wars and the death penalty, are
often associated with high rates of illegitimate, interpersonal violence. Gartner6 operationalised the
cultural context through the number of civil and international wars per nation from 1900 to 1980, the
total battle deaths per one million population over this period, and whether or not a country had the
21
death penalty. Cutright and Briggs7 used total battle deaths per million population as an indicator of
offically approved violence. These measures were significantly associated with homicide rates at all
ages and in both sexes. In a study of gangs in El Salvador, Smutt and Miranda8 found that 6 out of
10 gang members had directly experienced the armed conflict, suggesting that prolonged exposure
to armed conflict contributed to the creation of a culture of violence.
Cultural contexts which fail to provide nonviolent alternatives for conflict resolution may have higher
rates of youth violence. The ACTIVA study (Multisite Study on Cultural Attitudes and Norms related
to Violence in Selected Cities of Latin America and Spain), evaluated the strength of the association
between aggressive behaviours and attitudes and self-efficacy for alternatives to violence across
different settings.100 In all cities, the lack of self-efficacy for alternatives to violence was strongly
related with all forms of violence. Some attitudes towards specific behaviours were also associated
with violence. For example, the acceptance of corporal punishment was associated with aggression
toward the child and attitudes that supported slapping were associated with partner violence. Bedoya
Marin and Jaramillo Martinez101 describe, in their study of gang typologies in Medellin, Colombia,
how low-income youth are influenced by the culture of violence at the societal level and certain
subcultural elements at the community level. They argue that a “culture of violence*’ is fostered at
the community level through the tacit institutional support of funds generated through drug
trafficking, the growing value of “easy money” and means required to obtain it, and the corruption
of the police, judicial, military, and administrative forces. At the same time, behaviors such as
fighting by machete use have been transmitted through generations and adapted to the present
settings.
Cross-national cultural influences have also been linked with the rise in juvenile violence. Rodgers102
indicates, in a literature survey of youth gangs in Latin America and the Caribbean, that violent Los
Angeles-style gangs have emerged in northern and south-western Mexican towns, where immigration
from the United States is highest. These gangs emulate the behavioural patterns of their North
American counterparts, and have gang members who once emigrated to the United States, and thus
assimilated U.S. gang culture. A similar process has been found in El Salvador, with the high rate
of deportation of Salvadorian nationals from the United States since the 1992 peace accords, many
of whom were members of gangs in the United States.
Social Change. Co-occurring social changes at the structural level, such as demographic explosion
of the youth population, emigration, urbanisation, and changing social policies, have been linked with
an increase in youth violence. In countries affected by economic crises and ensuing structural
adjustment policies, notably in Latin America and Africa, there has been a shrinkage of real wages,
informalization of labor, and a substantial decline on basic urban infrastructure and social
services.103,104 Rapid urbanization has resulted in the concentration of poverty in cities where the
growth rate of the young population is greatest.105 These changes have resulted in higher number of
youths living in female-headed households, increased vulnerability to poverty, need to supplement
family incomes, and reduced socialization capacity within the family. A state of tension exists
between the decreasing supportive infrastructure and the demographic pressure created by a growing
population group of adolescents and young adults.
22
Lauras-Locoh and Lopez-Escartin104 suggest, in their analysis of youth demography in Africa, that
this tension has been expressed, for example, in school-based and student revolts. Diallo CoTrung106 describes the phenomenon in Senegal where the population younger than 20 doubled
between 1970-88 in the midst of a recession and the implementation of structural adjustment policies.
Students responded to the reduction in state public investment and the ensuing degradation of the
quality of education and increasing social inequality in access to schooling with strikes and student
revolts. Rarrbo107 found, in his study of Algerian youth, that the convergence of the demographic
explosion and the accelerated urbanization created tension-prone conditions, such as unemployment,
underemployment, and housing problems. Expectations of a stable job disappeared and frustrations
resulting from assimilation into urban culture and its temptations (designer goods, gadgets, luxury
goods) made young people more vulnerable to delinquent peer influence. Dinnen108 describes, in an
ethnographic account of a criminal group surrender in Papua New Guinea, how the evolution of
“raskolism” (criminal gangs) occurred within the broader context of decolonization and ensuing
social and political change, notably population expansion unmatched by economic growth. Such a
phenomena has also been cited as a concern in former communist economies.109 As the markets open
and the state machinery disappears unemployment soars and the social welfare system is reduced.
Young people lack legitimate alternatives and the necessary social support at the juncture between
school and work. They are exposed to the increased stress felt within the family and the general
disintegration of public order and social controls.
Conclusions
Although we have much more to learn it is clear that youth violence results from a complex
interaction among factors associated with our individual characteristics (biological and psychological
factors), the immediate social contexts in which we interact with others (family and friendship
networks), the nature of the commumties where we live (socio-economic status, presence of gangs
and illicit drug dealing), and the characteristics of the broader society and culture in which we exist
(economic conditions and norms and values). These different levels of influence on our behavior are
referred to as ecological contexts. Figure 9 displays the factors we discussed in the previous
literature review in an ecological model. The point of this model is to illustrate that not only is youth
violence influenced by many factors, but that one cannot understand youth violence without
considering how these different contexts operate together to cause or mitigate the expression of
violence by youth. This ecological perspective on youth violence has important implications for
preventing this global problem.
A Typology of Youth Violence Prevention Strategies
Youth violence is caused by a complex interaction among multiple factors at the individual, family,
community, and societal levels. Ultimately, therefore, public policies to reduce the health burden of
youth violence will need to be multifaceted to have measurable effects on the problem. One of the
23
more difficult challenges in violence prevention is how to design programs to address multiple risk
factors. 1 As the preceding discussion has shown, there are a number of factors — some residing
in the individual and others in the family and social environment — that increase the probability of
aggression and violence during childhood, adolescence, and early adulthood. Designing programs
to address multiple risks involves designing programs that influence not only individual cognitive,
social, or behavioral factors, but also the social systems that shape cognition, beliefs, development,
and behavior.
A typology of youth violence prevention strategies is useful when considering the range of prevention
strategies that might make-up a comprehensive effort to prevent youth violence. We propose a
typology based on two key dimensions: (1) the stages of human development and (2) an ecological
model of the multiple and interconnected social contexts in which human development occurs.
Within each stage of the human development, there are basic needs related to physical, emotional,
and social development, and developmental tasks related to the acquisition and mastery of
knowledge, skills, and relationships.3 Prevention strategies should clearly specify the developmental
stage of the people targeted for services and those services must be tailored for the respective
developmental needs of those people. Risk factors within such a framework serve as barriers to
successful development? For example, child abuse may be a barrier to the development of strong
emotional bonds between an infant and parent. Programs that address developmental needs, remove
barriers, and foster support for healthy development across a variety of contexts are key to preventing
violence.3
Developmental needs and tasks take place within a social context. The ecological model used in our
typology conceptualizes the multiple factors that can influence violence in the context of different
systems of influence (e.g., individual factors, close interpersonal relationships, proximal social
contexts, and societal macro-systems).111 Prevention measures that influence individual factors
would attempt to modify risk or protective factors associated with individual skills, attitudes, or
beliefs (e.g., poor peer relation skills, low academic achievement, or inappropriate beliefs about the
use of violence against others). Strategies addressing close interpersonal relationships would attempt
to influence the nature and quality of the interactions youth have with the people they interact with
on a regular basis (e.g., poor emotional bonding between parents and children, intense peer pressure
to engage in violence, or the absence of a strong relationship with a caring adult). Intervention
addressing close proximal social contexts would be designed to modify the day to day environments
and settings in which youth interact to increase their safety and support healthy development (e.g.,
modify elements of the physical environment that heighten the likelihood of assault, e.g. poor
lighting), or the lack of opportunities to engage in pro-social activities in neighborhood institutions
such as schools or churches). Strategies that address the societal macro-system would reduce
economic or social barriers to healthy development or enhance opportunities to achieve key
developmental needs (e.g., modification of norms or values imbedded in the culture that promote
violence, improve economic conditions).
Tables 3 and 4 provide examples of youth violence prevention strategies organized along the two key
dimensions in our proposed typology. The developmental stages from infancy to early adulthood are
24
one axis of the typology and the ecological systems through which violence can be prevented are
organized along the other axis. The prevention strategies presented in this table are not intended to
be exhaustive nor do they represent strategies that have all proven to be effective; rather, they are
presented to illustrate the breadth of potential solutions and to emphasize the need to consider
addressing the problem simultaneously at different stages of development and through different social
systems of influence. There are at least two noteworthy limitations of this typology. First, the
typology does not differentiate between strategies directed at preventing violent behavior and those
designed to prevent violent victimization. Although this important distinction is not immediately
obvious, both are useful approaches to prevention and are intermixed among the array of strategies.
Second, the typology does not differentiate among specific types of violence. One could imagine
developing a separate array of prevention strategies for different types of violence (e.g., school
violence, gang violence, violence between intimates). Tables 3 and 4 include strategies that address
a number of different types of violence.
The Effectiveness of Youth Violence Prevention Strategies
As is evident from the above typology there is a broad range of strategies for preventing youth
violence. Unfortunately the evidence for the effectiveness of many of these strategies has never been
demonstrated in a scientifically rigorous way. There are several types of evidence that should be
considered before concluding that a program is effective.112 Programs should be evaluated, wherever
possible, using experimental designs that provide evidence of statistically significant effects on the
reduction of violent behavior or associated injuries. These results should be replicated in multiple
sites and across different cultural contexts. Finally there should be evidence that the impact is
sustained over time.
We review what’s known about the effectiveness of a range of violence prevention strategies. This
review is organized by the developmental stage of the persons that are being targeted for reducing
violent behavior, the future potential for violent behavior, or the risk of victimization. The social
context through which the intervention operates to influence violent behavior is considered within
each developmental stage. We have focused this review on the results of experimental and quasiexperimental studies of the effectiveness of interventions to prevent youth violence. These types of
evaluation studies allow for the measurement of the independent effects of the intervention on violent
behavior and/or injury by using control groups and collecting data longitudinally (i.e. prior to and
after the implementation of the intervention) in the research methodology. As in the case of the
literature on risk factors these studies have been conducted primarily in Western societies.
Infancy and Early Childhood
The biological and environmental factors that can ultimately influence the likelihood of youth
violence are first evident during infancy and early childhood. Children are bom with a basic
25
temperament and constellation of genetic characteristics and predispositions that interact with a
child's environment to influence their likelihood of expressing aggression and violence later in life.
Neurological impairments, the nature of the patterns of interactions between a child and its
caregivers, exposure to neglect and/or physical abuse, and environmental stresses associated with
poverty may all contribute to the propensity for aggressive and violent behavior.5,26,113-123 There are
a number of prevention strategies targeted at infants and/or their caregivers that appear effective in
preventing the subsequent expression of aggressive and violent behavior during adolescence and
young adulthood.
Home Visitation. Home visitation is an intervention in infancy and early childhood that typically
involves weekly to monthly visits to provide parenting and health information, emotional support,
counselling, and referrals to outside agencies to families at high-risk for abuse or other health
problems.124 Although findings are mixed some studies have found evidence that over the long-term,
visited children have lower rates of antisocial behavior than those who did not receive home visits.125’
129
Parenting Programs. Parenting skills can be learned and improved through training. These
programs are designed to improve the emotional bond between parents and their children, to
encourage parents to use consistent and contingent child-rearing methods and to help develop
parental problem-solving skills and self-control in raising children. Several programs have found
this type of training to be successful and there is some evidence of a long-term impact in reducing
antisocial behavior.130’133 In a study of the cost effectiveness of early interventions for preventing
serious forms of crime in California, training for parents of school-age children exhibiting aggressive
behavior was estimated to prevent 157 serious crimes (i.e., homicide, rape, arson, robbery, aggravated
assault, and residential burglary) per million dollars spent.134 Parent training was estimated to be
about three times as cost effective as the California “three-strikes” law (a law that put in place harsh
sentences for repeat offenders).
Early Childhood Education. Pre-school enrichment programs can be helpful by giving young
children a head start in the skills needed to be successful in school as they grow older. These
programs can strengthen a child's bonds to school, improve school achievement, and self-esteem.124
Long-term follow-ups of prototypical pre-school enrichment programs have found many positive
benefits for children, including less involvement in violent and other delinquent behaviors than for
similar children not enrolled in such programs.126,135’136
Other interventions during these early developmental stages that might prove to be effective include
programs to: prevent unintended pregnancy in order to reduce the constellation of risk factors
associated with early childbirth that could contribute to youth violence,124 increase access to pre and
post-natal care in order to reduce the potential for birth trauma and enhance well baby care thus
minimizing health problems that could contribute to youth violence,3 and monitor lead levels and
remove toxins in order to prevent damage to a child's brain that could indirectly contribute to youth
violence by limiting cognitive abilities and school performance.3 Further evidence of the long-term
effects of such programs on violent and aggressive behavior is needed.
26
Adolescence
Aggressive behavior patterns, poor problem-solving skills, and other experiences in early childhood
contribute to the potential for adolescent violent behavior, by interfering with the development of
positive peer relationships, academic achievement, and adjustment in school and other social contexts
during adolescence.26,54’115’137-143 There are a number of prevention strategies targeted at adolescents,
their peers, parents, and social environments that appear very promising for preventing the expression
of aggressive and violent behavior.
Social Development Programs. Social development programs seek to improve children's social
skills with peers and others and to promote behavior that is positive, friendly, and cooperative. ' 5
These programs typically focus on one or more of the following dimensions: anger management,
perspective taking, moral development, social skills, social problem solving, or conflict resolution.
There is evidence that these types of program can be effective in reducing behavior problems and
improving social skills.146-147 Available results suggest that they are more promising when part of
more comprehensive efforts that include the direct involvement of teachers, parents, and peers.
Mentoring. Mentoring programs match youth with a non-familial caring adult, particularly those
youth growing up in a single parent family or adverse situation.148 Mentors can be older classmates,
teachers, counselors, police officers, or members of the community. The goals of mentoring are to
assist youth in developing skills and to provide a sustained relationship with a more experienced
person who serves as a positive role model and guide.145 An evaluation of the Big Brothers/Big
Sisters Program, the oldest and best known mentoring program in the United States, found that a
positive mentoring relationship led to reductions in self-reported anti-social behaviors, such as hitting
and drug use.149
Family Therapy. There are many forms of family therapy, but, in general, they share the common
goals of improving communication, interaction, and problem solving between parents and their
children.145 Some programs also focus on improving a family's ability to deal with factors in the
youth’s and family's ecology that may contribute to antisocial behavior and help the family make
better use of resources that exist within their community. These programs are most appropriate for
families experiencing a high level of conflict and behavioral problems.145 Family therapy programs
can be costly, but there is substantial evidence that they can be effective in improving family
functioning and reducing child behavior problems.150-152 Functional Family Therapy153 and
Multisystemic Therapy154 are two approaches used in the U.S. that have been demonstrated to have
strong and long-term effects on reducing the violent and delinquent behavior ofjuvenile offenders
1 1o
at lower costs than other treatment programs.
Other strategies that may be effective in addressing violence during this developmental stage, but
have not been adequately evaluated, include programs to: provide incentives for graduating from high
school and attending college for underprivileged youth, prevent adolescents from gaining access to
firearms without the supervision of an adult, situational crime prevention strategies,155 supervised
after-school recreation,124 and public information campaigns to promote pro-social norms.3 Programs
that do not appear to be effective in reducing adolescent violent behavior include: peer mediation
27
(i.e., programs where students help other students resolve disputes), 124 peer counseling,124 and
replacing school work with vocational training and employment.124
Early Adulthood
A weak family environment, involvement in delinquent peer groups, and school failure makes the
transition to adulthood more difficult and the likelihood of violent behavior greater.110,156 Factors
such as unemployment, relationship difficulties, skills deficits, drug and/or alcohol abuse may
contribute to feelings of estrangement and poor social adjustment in early adulthood.110 The net
effect of these factors is diminished opportunity to successfully meet the challenges of adult life and
a greater potential to engage in violent behavior. Failure to adapt to these challenges also has
potential implications for violent behavior that is expressed towards the children and partners of these
young adults.
There is a dearth of prevention programs that target young adults. Given the high-risk faced by
persons in this developmental stage for both violent perpetration and victimization in almost every
part of the world, this represents a huge gap in the existing knowledge base. Programs are needed
which assist adolescents in making the transition to adulthood and help young adults learn to adapt
to changing roles and new environments. Attenuating the negative effects of low opportunities for
meaningful employment would appear to be a particularly high priority for young adults living in
poor neighborhoods. The prevention of child and partner abuse should be a particularly high priority
in this developmental stage.
Multiple Component Prevention Programs
As is evident from our review of risk factors and prevention strategies, youth violence is caused by
a complex interaction among multiple factors and efforts to reduce this problem in a substantial way
will need to be multifaceted. Ideally programs should approach youth through multiple systems of
influence (individual, family, community, and society) and provide a continuum of interventions and
activities that span the stages development. Such programs can address co-occurring risk factors,
such as school attainment, early pregnancy, unsafe sex, and drug use, and can address the needs of
1 09
adolescents in all spheres of their lives.
The DESEPAZ program in Cali, Colombia, illustrates the role of a municipal government in
organizing a comprehensive repsonse to the problems of crima and violence. In Cali the Municipal
Security Council gathered government officials on a weekly basis to study the epidemiology of
violence in select neighborhoods of the city and to develop plans of action. Actions in the areas of
epidemiological analysis, social communication, institutional strengthening of the legal sector,
community mobilisation and infrastructure development, were taken based on the information
provided by the surveillance system and the community-based consultation process. Restriction on
28
the sale of alcohol and banning the permits to carry hand guns were put on weekends and other
special occasions. Public opinion around violence moved from a passive attitude to an active demand
for more prevention. The homicide rate declined from 124/100,000 in 1994 to 86/100,000 in 1997,
a 30% reduction in a period of three years.
ESSOR in Mozambique presents a case of a comprehensive adolescent delinquency prevention
program implemented in two low-income neighborhoods of Maputo.158 The program targeted
adolescents between 13-18 years of age and functioned within a larger community-based
development initiative. Socio-educational activities on identified adolescent risk behaviors prepared
participants to disseminate acquired information into the community. Sports and leisure activities
were used to promote self-expression, team-building, and a development of trust between social
action agents and the youth. Professional training was offered to youth older than 15 years of age.
The program included regular home visitation by a social action agent for all participating youths.
While at the beginning of the program only 5% of the youth showed "constructive” behaviors, 18
months after the number had increased to 23%. The pre-intervention evaluation indicated most
youths were involved in gangs, while 18 months after, 5% of youths reported belonging to gangs or
prostituting themselves. At the same time, parents reported that parent-child relations improved: 86%
of those parents who had reported difficult interactions before the program have reported an
improvement. Community members and institutions included in the evaluation reported that 84%
believed significant changes had favored the decrease of delinquency.
The Children At Risk Program in the United States presents another example of a community-based,
comprehensive initiative for at-risk youth aged 11-15.159 Based in six high-risk neighborhoods, the
program aimed to prevent drug abuse and selling and promote healthy development for youth at risk
of delinquency. In a similar fashion as ESSOR; case managers worked with the youth and their
families to develop and follow-up a service plan, and facilitate the use of existing community social
services. The project site also offered family services, educational support, after-school and summer
activities, mentoring, enhanced community policing, and collaboration with the juvenile justice
system. A preliminary evaluation 12 months after the initiation of the project indicated that
participants had a lower number of contacts with the police than did youth in the control group (41
versus 69). The treatment group also had less contacts with juvenile court: 34 contacts compared to
71 in the control group.
The Project for the Promotion of Community Self-Management (PROFAC), implemented in the
marginalized urban community of Rincon Grande de Pavas in Costa Rica, has adopted a multi
faceted approach to work with gangs.100 A negotiation process with the youth achieved a cease-fire.
Strategies include involvement in community reconstruction projects, the creation of a youth
orgamzation, and the establishment of a community business. Preliminary results indicate continued
project involvement of former gang members, maintenance of the cease-fire, and sustained
reconstruction of destroyed property.
29
Summary and Recommendations
In sum, deaths and injuries from youth violence constitute a substantial public health problem in
many regions of the world. Substantial variations in the magnitude of this problem exist within and
between nations and regions of the world. Longitudinal research has revealed a great deal about the
patterns and causes of youth violence. We know that violent youth are versatile in committing other
types of crime and antisocial acts, but there is some specialization in violence. There is significant
continuity from childhood aggression to youth violence and from youth violence to adult violence.
The major risk factors for youth violence are biological factors (low heart rate),
psychological/personality factors (impulsivity, low intelligence), family factors (poor supervision,
harsh discipline, violent parents, large families, young mothers, broken families), peer delinquency,
low socioeconomic status, and disintegrated neighborhoods. Studies of macro-level influences
suggest that economic inequality, the extent to which legal and social protections are supported and
enforced, low levels of social integration, the presence of cultural norms supportive of violence, and
rapid social changes may all influence the level and nature of youth violence in a society. Given that
youth violence is caused by a complex interaction among these biological, psychological, and social
factors, public policies and prevention programs to reduce the health burden of youth violence will
need to be multifaceted to have measurable effects on the problem. The evaluation research on youth
violence suggests that interventions applied during infancy and early childhood are more effective
than interventions applied during adolescence and early adulthood. We have also learned that the
effectiveness of prevention programs depends on the quality of the implementation process.
Problems such as poor staffing, departures from intended procedures, and the lack of administrative
support may contribute to program ineffectiveness. Based on our findings we offer the following
recommendations for how we can improve our understanding of youth violence and its prevention.
The foundation of efforts to prevent youth violence should be the development of data systems to
routinely monitor patterns and trends in associated violent behaviors, injuries, and deaths. The data
from these systems provides information useful for targeting and formulating public policies and
prevention programs and evaluating their effectiveness. The following aspects of the public health
surveillance of youth violence require greater attention:
•
There is a need for the development of simple approaches to youth violence surveillance that can
be applied in a wide range of cultural settings.
•
Uniform standards for defining and measuring youth violence should be developed and
incorporated into injury and violence surveillance systems. These standards should include age
categories that are sensitive to the developmental differences in the patterning of risk and
victimisation.
•
Priority must be given to developing and implementing systems for the surveillance of deaths due
to violence in regions where homicide data are currently inadequate or lacking. These regions are
Africa, the Eastern Mediterranean, India, the low- to middle-income Western Pacific and parts
of the low- to middle-income Americas.
30
•
Surveillance activities should be complemented by special studies dedicated to establishing th
ratio of fatal to non-fatal cases by method of attack, age and gender. Such data can then be use
to estimate the dimensions of the youth violence problem where only one type of data (i.t
mortality or morbidity) is available.
•
All countries and regions should be encouraged to establish centres where routinely availabl
information from health services (such as energency departments), the police and othc
authorities in regular contact with victims and perpetrators of violence can be collated an
compared in order to inform prevention programmes.
Scientific evidence regarding the patterns and causes of youth violence, both qualitative an
quantitative, is fundamental to the development of rational responses to this problem. While grez
strides have been made in improving our understanding significant gaps remain. The followin
areas of research require greater attention:
•
Studies that focus specifically on youth violence as opposed to childhood aggression c
delinquency.
•
Research into the validity and relative advantages of using official records, hospital records, an
self-reports to measure youth violence.
•
There is significant continuity from childhood aggression to youth violence, but reasons fo
discontinuity or desistance and the role of protective factors, should be investigated.
•
Estimates of the total cost of violent youth to society are needed (including the cost of co
occurring problems such as substance abuse and conduct problems) for better assessing the cos
effectiveness of prevention and treatment programs.
•
New longitudinal studies are needed that measure a broad range of risk and protective factors t«
further investigate developmental pathways to youth violence.
•
Research is needed that contrasts violent offenders with nonviolent offenders and non-offender
and on which risk factors have differential effects on persistence, escalation, de-escalation
desistance of violent offending at different ages.
•
Research is needed to identify protective factors for youth violence.
•
Research is needed on female involvement in youth violence.
•
Cross-cultural research on the causes, development, and prevention of youth violence is needec
to help explain the remarkable variation in levels of youth violence that exists across nations ani
regions of the world.
•
Cross-cultural research is needed on the societal and cultural influences on youth violence.
31
•
Research is needed that can provide clearer guidance as to how we might modify macro social
and economic factors in ways that are effective in reducing youth violence.
•
Institutions are needed to organize, coordinate, and fund cross-national research on youth
violence.
The demand for knowledge about how to effectively prevent youth violence has never been greater.
To date, however, most of the resources committed to prevention have been invested in untested
programs, many of which are based on questionable assumptions and delivered with little consistency
or quality control. Progress in our ability to effectively prevent and control youth violence requires
evaluation. The following areas of evaluation research require greater attention:
•
We need to adopt consistent standards in the application of evaluation methods for the assessment
of the effectiveness of youth violence prevention programs and policies. These standards should
include the application of an experimental design, evidence of statistically significant effects on
the reduction of violent behavior or injuries associated with such behavior, replication at multiple
sites and across different cultural contexts, and evidence that the impact is sustained over time.
•
Greater investment should be made in assessing the cost effectiveness of youth violence
prevention programs and policies. A better understanding of the relative costs and benefits of
alternative approaches to prevention will be helpful in selecting particular approaches given the
limited availability of resources for prevention.
•
Longitudinal evaluation designs are needed to assess the long-term impact of interventions
applied in infancy or childhood on the occurrence of youth violence.
•
Evaluations of the impact of social experiments that seek to modify social conditions associated
with youth violence such as income inequality or the concentration of poverty should be
conducted whenever the opportunity arises.
•
Systematic reviews of youth violence interventions are needed, such as the Cochrane
Collaborative on health care interventions.
Much greater attention needs to be given to the application of what we have learned about the causes
and prevention of youth violence. We currently lack the appropriate infrastructure to translate and
communicate what is known to practitioners and policy-makers. This is a problem we face in every
part of the world. Effective implementation of youth violence prevention programs and policies
requires a greater commitment to building an adequate capacity to provide the appropriate training
and technical assistance. The following areas of program implementation issues require greater
attention:
•
Global coordination is needed to facilitate the development of organizational networks focused
on information sharing, training, and technical assistance.
•
Simultaneous investment should be made in the application of internet technology and more
generally available mechanisms for sharing information in settings lacking access to computers.
•
Information on youth violence and its prevention is generally available from information sources
in western cultures, but much harder to find for other regions of the world. The development of
international clearinghouses to identify and translate relevant information from every region of
the world and from fugitive information sources is needed.
•
Research is needed on how best to implement youth violence prevention strategies and policies.
The identification of effective prevention strategies is insufficient to assure their successful
implementation. Therefore, we need to learn much more about how to successfully implement
promising strategies particularly across different cultural settings.
•
The prevention of youth violence should be integrated, wherever feasible, with programs to
prevent child abuse and other forms of intimate violence in the family.
Both the demand for and supply of information about the causes and prevention of youth violence
is rapidly increasing. Continued progress in this area will require substantial investments to improve
the capacity to conduct public health surveillance, to carry out the needed etiologic and evaluation
research (including longitudinal studies), and the investment in a global infrastructure to disseminate
and apply what is learned. If these investments can be realized youth violence will come to be
generally viewed as a preventable public health problem within the foreseeable future.
33
Table 1. Youth homicide rates (10-24 years, both sexes) and projected overall homicide
rates per 100 000 by region, 1992-1996
Regional rates and between-country variations
Projected homicide
Actual homicide rate,
10-24 year oldsb
Region"
Highest
Lowest
Median
rate, all agesc
481
Africa
No data
America, high income
12.81
16.35
1.71
14.02
4.2 ???
America, low- to middle
24.88
95.3
0.68
8.76
26.3
income
11.6
Eastern Mediterranean
Insufficient data
Europe, high-income
1.02
1.69
0
0.97
4.3
Europe, low- to middle
10.10
49.6
0.40
4.77
9.8
income
No data
8.2
Insufficient data
8.8
China
3.14
4.8
Western Pacific, high-
0.77
India
South East Asia, low- to
middle-income
2.65
0.32
1.43
4.2
income
Western Pacific, low- to
No data
8.8
middle-income
a
All countries within each region and countries with youth homicide data are shown in endnote one.
b Calculated from country mortality data provided by WHO, PAHO and the Brazilian Ministry of Health.
c Projections prepared by WHO Global Programme for Evidence and Information for Policy.
Table 2. Summary of Special Studies of Youth Homicide in Selected Low- and Middle-Income Countries
Country, Year, Reference
Sample size and method
Findings and comments
South Africa, 199813
Descriptive analysis of all homicides reported to the police and
A total of 25 039 homicides were recorded for the year,
South African Police Services, Crime Information Analysis Centre,
recorded in the national crime information analysis centre
equivalent to an incidence rate of 59.6 per 100 000 population.
unpublished data
Of the homicides, 3.7% were aged 0-11 years, 3.4 percent
12-17 years, and 84.7% 18-49 years.
South Africa, 1992-199514
Verbal autopsies of 932 deaths identified during annual health and
Homicide incidence was 2 per 100000 in 5-14 years and 46
Kalin, Tollman, Garenne & Gear, 1999
demographic surveys in a rural area
per 100 000 in ages 15-49. Males were the victim for 83% of
the homicides, and the main external causes were stabbings
and gunshots.
South Africa, January-December 199915
Sample of 14754 known and suspected non-natural deaths
Homicide accounted for 6826 (46%) of all eases. Of the
Violence and Injury Surveillance Consortium, 1999 Annual Report
registered and mcdico-lcgally examined al 12 state mortuaries in
homicide victims, 0.8% were aged 10-14, there were 7.7%
(forthcoming)
metropolitan and urban centres
aged 15-19, 16.1% aged 20-24 and 20.1% aged 25-29.
Firearms were involved in 56.9% of all cases and there were
6.6 male victims for every female victim.
South Africa, 1990-199216
Clinical-forensic analysis of 1198 traumatic cardiac injuries
Campbell el al, 1997
presenting to hospitals and mortuaries in the Durban metropole
All but 6% of victims died. The mean age was 30.5 years
(range 5 to 77 years) and 91 % were male. The proportion of
eases involving firearm injuries increased from 34% in 1990
to 50% in 1992.
South Africa, 199717
6194 known and suspected non-natural deaths registered and
Homicide accounted for 2065 (33.3%) of all eases. Of the
Phillips, 1999
mcdico-lcgally examined in the Cape Town mctropolc
homicide victims 1.5% were 0-14 years, 31.4% were 15-24
years and 35.2% were 25-34 years. There were 11.7 males for
every female victim. Firearms accounted for 38.6% of the
deaths, and sharp instruments for 45.2%.
South Africa, 1992-199618
Retrospective study of mortuary, police and hospital records of
A total of 1 736 children and adolescents sustained firearm
Wigton, 1999
firearm injuries in children from 0 to 18 years of age resident in the
injuries, of which 86% were between 13 and 18 years of age.
Cape Town mctropolc.
One fifth of these eases were fatal, and the annual number of
gunshot injuries in persons under 19 years presenting at
hospitals tripled from 142 to 421 between 1992 and 1996.
Nigeria, 1991-199319
Prospective study of 876 consecutive coroner’s autopsies performed
A total of 27 (3.1%) of all cases were due to homicide, of
Amakiri ct al, 1997
at the University College in Ibadan.
which gunshots accounted for 55.6%. No homicides were
found in victims aged 0-14.
Nigeria, 1977-198820
Prospective study of 202 homicides examined at the Ilc-Ifc
Nwosu & Odcsanrni, 1998
Teaching Hospital Department of Morbid Anatomy.
Most victims were aged 20-40 years, and there were 4.6 males
for every female. F irearms accounted for 37% of the deaths
and sharp instruments for 35%.
Tanzania, 1992-199811
Descriptive analysis of 25 548 deaths recorded through annual
Of all deaths 1474 (5.8%) were due to injuries of which 26%
Moshiro ct al, submitted for publication (check to sec if published
household health and demographic surveys of one urban population,
were a result of violence. Male homicide rates in 5-14 year
yet and if so replace with proper reference - if not special
permission to use the results will be required).
one wealthy rural population and one poor rural population.
olds varied from 0 to 4.1 per 100 000, and in 15-59 years olds
Fiji, 1969-198922
Retrospective study of Republic of Fiji Ministry of Health official
No age-related findings given. Overall homicide rate of 0.4 per
Pridmorc, Ryan & Blizzard, 1996
mortality and morbidity reports.
100 000 population.
Taiwan, 1965-199423
Retrospective analysis of Taiwan official vital statistics, data
Among 10-14 years olds, homicide rates per 100 000 varied
Loe, Lee & Chou, 1998
aggregated into five year periods for males and females in ages 10-
between 0.16 and 0.35 in males, and in females from 0.17 to
14 and 15-19.
0.3. In ages 15-19 male rates ranged from 2.33 to 4.15, and
from 22.3 to 32.1. Female homicide rates for ages 10-14
ranged from 0 to 4.3 and in ages 15-59 from 4.4 to 6.1.
female rates from 0.25 to 0.9. There were no clear trends over
time.
Table 3. Prevention Strategies by Developmental Stage (Infancy Through Middle Childhood) and
Ecological Context
Developmental Stage
Ecological Context
Infancy
(ages 0-3)
Early Childhood
(ages 3-5)
Middle Childhood
(ages 6-11)
Individual
Increase access to
prenatal/postnatal
services to minimize
birth trauma, provide
adequate nutrition and
well-baby care
Social development
training
Preschool
enrichment
programs including
Head Start
Social development
training in anger
management, social
skills, and problem
solving
Close Interpersonal
Relations (e.g.,
family, peers)
Home visitation
services to strengthen
families
Parenting training for
new parents
Respite day care
centers or drop-in
programs
Parenting training
Mentoring
Home-school
partnership
programs to promote
parental involvement
Proximal Social
Contexts (e.g.,
schools, churches)
Lead monitoring and
toxin removal
Foster care programs
Lead monitoring
and toxin removal
Foster care
programs
Create safe havens
for children on highrisk routes to and
from school
Provide after-school
programs to extend
adult supervision
Recreational
programs
Social Macro-systems
(e.g., cultural norms,
economy)
Deconcentrate lowerincome housing
Reduce income
inequality
Public information
campaigns to
promote pro-social
norms
Reduce levels of
media violence
Public information
campaigns to
promote pro-social
norms
Table 4. Prevention Strategies by Developmental Stage (Early and Late Adolescence, Early
Adulthood) and Ecological Context
Developmental Stage
Ecological Context
Early Adolescence
(ages 12-14)
Individual
Social development training in anger
management, perspective taking, moral
development, social skills, and problem solving
Close Interpersonal
Relations (e.g.,
family, peers)
Mentoring
Academic enrichment programs
Peer mediation
Temporary foster care programs for serious and
chronic delinquents
Provide education to promote healthy relations
with the opposite sex and decrease dating
violence
Bullying
prevention
programs
Proximal Social
Contexts (e.g.,
schools, churches)
Social Macro-systems
(e.g., cultural norms,
economy)
•
•
•
Late Adolescence
(ages 15-19)
Early Adulthood
(ages 20-24)
Vocational training
Family therapy
Recreational Programs
Multi-component gang prevention programs
Train health care professionals in identification
and referral of high-risk youth
Establish adult
recreational
programs
Public information campaigns to promote pro
social norms
Reduce levels of media violence
Educational incentives for at-risk, disadvantaged
high school students
Enforce laws prohibiting illegal transfers of guns
to youth
Establish meaningful
job creation
programs for the
chronically
unemployed
Provide incentives
for post-secondary
education or
vocational training
References
1.
Reza A, Krug EG, Mercy JA. The Epidemiology of Violent Deaths in the World.
Unpublished Manuscript, 2000.
2.
Ad Hoc Committee on Health Research Relating to Future Intervention Options. Investing
in Health Research and Development. World Health Organization, Geneva, 1996
(Document TDR/Gen/96.1).
3.
Williams KR, Guerra NG, Elliott DS. Human Development and Violence Prevention: A
Focus on Youth. Boulder, CO: Center for the Study and Prevention of Violence, 1997.
4.
Fagan J, Browne A. Violence between spouses and intimates: physical aggression between
women and men in intimate relationships. In: Reiss AJ, Roth JA (eds.) Understanding and
preventing violence: panel on the understanding and control of violent behavior. Vol 3:
Social influences. Washington, DC: National Academy Press 1994:114-292.
5.
Widom CS Child abuse, neglect, and violent criminal behavior. Criminology 1989;244:160166.
6.
Gartner R. The victims of homicide: a temporal and cross-national comparison.. American
Sociological Review 1990; 55: 92-106.
7.
Briggs CM, Cutright P. Structural and cultural determinants of child homicide: a cross
national analysis. Violence and Victims 1994; 9:3-16.
8.
Smutt M, Miranda JLE. El Salvador: Socializacion y violencia juvenil. In Ramos CG(ed).
America Central en los noventa: Problemas de Juventud. FLAGSO Programa El Salvador,
julio 1998.
9.
World Health Organization. Violence: A Public Health Priority. Working document
EHA/SPI/POA.2 December, 1996.
10.
Farrington DP. Predictors, causes, and correlates of male youth violence. In Tonry M, Moore
MH (eds.) Youth Violence. Chicago: University of Chicago Press, 1998: 421-475.
11.
Rivara F P, Shepherd J P, Farrington D P, Richmond P W, Cannon P. Victim as offender
in youth violence. Annals of Emergency Medicine 1995; 26, 609-614.
12.
Loeber R, DeLamatre M, Tita G, Cohen J, Stouthamer-Loeber M, Farrington D P. Gun
injury and mortality: The delinquent backgrounds ofjuvenile victims. Violence and Victims
1999; in press.
13.
South African Police Services, Crime Information Centre, unpublished data, 1988.
34
14.
Kahn K, Tollman, SM, Garenne, M Gear, JS. Who dies from what? Determining cuase of
death in South Africa’s rural north-east. Tropical Medicine and International Health 1999;
4: 433-441.
15.
Violence and Injury Surveillance Consortium, unpublished data, 1998.
16.
Campbell NC, Thomson SR, Muckart DJ, Meumann CM, Van Middelkoop I, Botha JB.
Review of 1198 cases of penetrating trauma. British Journal of Surgery 1997; 84: 17371740.
17.
Phillips R. The economic cost of homicide to a South African city. Unpublished Masters
of Business Administration dissertation 1999; Graduate School of Business, University of
Cape Town.
18.
Wygton A. Firearm-related injuries and deaths among children and adolescents in Cape
Town - 1992-1996. South African Medical Journal 1999; 89: 407-410.
19.
Amakiri CN, Akang EE, Aghadiuno PU, Odesanmi, WO. A prospective study of coroner’s
autopsies in University College Hospital, Ibadan, Nigeria. Medicine, Science and Law 1997;
37: 69-75.
20.
Nwosu SE, Odesanmi WO. Pattern of homicides in Nigeria - the Ile-Ife experience. West
African Medical Journal 1998;17: 236-268.
21.
Moshiro C, Setel PW, Whiting DR, Unwin N, Mclarty DG, Alberti KGMM, AMMP Project
Team. The importance of injury as a cause of death in sub-Saharan Africa: results of a
community-based study in Tanzania. Unpublished manuscript submitted for publication
consideration.
22.
Pridmore S, Ryan K, Blizzard L. Victims of violence in Fiji. Australian and New Zealand
Journal of Psychiatry 1995; 29: 666-670.
23.
Lu T-H, Lee M-C, Chou M-C. Trends in injury mortality among adolescents in Taiwan,
1965-94. Injury Prevention 1998; 4: 111-115.
24.
Laub JH, Lauritsen JL. Violent criminal behavior over the life course: A review of the
longitudinal and comparative research. In Ruback RB, Weiner NA (eds.) Interpersonal
Violence Behaviors: Social and Cultural Aspects. New York, NY: Springer Publishing
Company, 1995,43-61.
25.
Brennan P, Mednick S, John R. Specialization in violence: Evidence of a criminal subgroup.
Criminology 1989; 27: 437-453.
35
26.
Farrington D P. The family backgrounds of aggressive youths. In Hersov L, Berger M,
Shaffer D (eds.) Aggression and antisocial behavior in childhood and adolescence. Oxford:
Pergamon 1978: 73-93.
27.
Capaldi D M, Patterson G R. Can violent offenders be distinguished from frequent
offenders? Prediction from childhood to adolescence. Journal of Research in Crime and
Delinquency 1996; 33: 206-231.
28.
Farrington D P. Childhood aggression and adult violence: Early precursors and later life
outcomes. In Pepler D J, Rubin K H (eds.) The development and treatment of childhood
aggression. Hillsdale, NJ: Lawrence Erlbaum 1991: 5-29.
29.
Hamparian D M, Schuster R, Dinitz S, Conrad J P. The violent few: A study of dangerous
juvenile offenders. Lexington, MA: D.C. Heath 1978.
30.
Elliott D S. Serious violent offenders: Onset, developmental course, and termination.
Criminology 1994; 32: 1-21.
31.
Hamparian D M, Davis J M, Jacobson J M, McGraw R E. The young criminal years of the
violent few. Washington DC: Office of Juvenile Justice and Delinquency Prevention 1985.
32.
Farrington D P. Predicting adult official and self-reported violence. In Pinard G-F, Pagani
L (eds.) Clinical assessment of dangerousness: Empirical contributions. New York:
Cambridge University Press 2000: in press.
33.
Stattin H, Magnusson D. The role of early aggressive behavior in the frequency, seriousness,
and types of later crime. Journal of Consulting and Clinical Psychology 1989; 57: 710-718.
34.
Pulkkinen L. Offensive and defensive aggression in humans: A longitudinal perspective.
Aggressive Behavior 1987; 13: 197-212.
35.
Loeber R, Wung P, Keenan K, Giroux B, Stouthamer-Loeber M, van Kammen W B,
Maughan B. Developmental pathways in disruptive child behavior. Development and
Psychopathology 1993; 5:103-133.
36.
Raine A. The psychopathology of crime: Criminal behavior as a clinical disorder. San
Diego, CA: Academic Press 1993.
37.
Kagan J. Temperamental contributions to social behavior. American Psychologist
1989;44:668-74.
38.
Wadsworth M E J. Delinquency, pulse rates, and early emotional deprivation. British
Journal of Criminology 1976; 16: 245-256.
36
39.
Farrington D P. The relationship between low resting heart rate and violence. In Raine A,
Brennan P A, Farrington D P, Mednick S A (eds.) Biosocial bases of violence. New York:
Plenum 1997: 89-105.
40.
Kandel E, Mednick S A. Perinatal complications predict violent offending. Criminology
1991;29:519-529.
41.
Brennan P A, Mednick B R, Mednick S A. Parental psychopathology, congenital factors,
and violence. In Hodgins S (ed.) Mental disorder and crime. Newbury Park, CA: Sage
1993:244-261.
42.
Denno D W. Biology and violence: From birth to adulthood. Cambridge, England:
Cambridge University Press 1990.
43.
Henry B, Caspi A, Moffitt T E, Silva P A. Temperamental and familial predictors of violent
and nonviolent criminal convictions: Age 3 to age 18. Developmental Psychology 1996;
32: 614-623.
44.
Caspi A, Moffitt T E, Silva P A, Stouthamer-Loeber M, Krueger R F, Schmutte P S. Are
some people crime-prone? Replications of the personality-crime relationship across
countries, genders, races, and methods. Criminology 1994; 32: 163-195.
45.
Klinteberg B A, Andersson T, Magnusson D, Stattin H. Hyperactive behavior in childhood
as related to subsequent alcohol problems and violent offending: A longitudinal study of
male subjects. Personality and Individual Differences 1993; 15: 381-388.
46.
Hogh E, Wolf P. Violent crime in a birth cohort: Copenhagen 1953-1977. In van Dusen
K T, Mednick S A (eds.) Prospective studies of crime and delinquency. Boston: KluwerNijhoff 1983:249-267.
47.
Lipsey M W, Derzon J H. Predictors of violent or serious delinquency in adolescence and
early adulthood: A synthesis of longitudinal research. In Loeber R, Farrington D P (eds.)
Serious and violent juvenile offenders: Risk factors and successful interventions. Thousand
Oaks, CA: Sage 1988: 86-105.
48.
Moffitt T E, Henry B. Neuropsychological studies of juvenile delinquency and juvenile
violence. In Milner J S (ed.) Neuropsychology of aggression. Boston: Kluwer 1991: 131146.
49.
Seguin J, Pihl R O, Harden P W, Tremblay R E, Boulerice B. Cognitive and
neuropsychological characteristics of physically aggressive boys. Journal of Abnormal
Psychology 1995; 104: 614-624.
37
50.
McCord J. Some child-rearing antecedents of criminal behavior in adult men. Journal o
Personality and Social Psychology 1979; 37: 1477-1486.
51.
McCord J. A comparative study of two generations of native Americans. In Meier R F (ed.
Theory in criminology: Contemporary views. Beverly Hills, CA: Sage 1977: 83-92.
52.
McCord J. Family as crucible for violence: Comment on Gorman-Smith et al. (1996).
Journal of Family Psychology 1996; 10: 147-152.
53.
Gorman-Smith D, Tolan P H, Zelli A, Huesmann L R. The relation of family functioning
to violence among inner-city minority youths. Journal of Family Psychology 1996; 10: 115129.
54.
Thomberry T P, Huizinga D, Loeber R. The prevention of serious delinquency anc
violence: Implications from the program of research on the causes and correlates of
delinquency. In Howell J C, Krisberg B, Hawkins J D, Wilson J J (eds.) Sourcebook on
serious, violent, and chronic juvenile offenders. Thousand Oaks, CA: Sage 1995: 213-237.
55.
Wadsworth M E J. Delinquency prediction and its uses: The experience of a 21-yeai
follow-up study. International Journal of Mental Health 1978; 7: 43-62.
56.
Malinosky-Rummell R, Hansen D J. Long-term consequences of childhood physical abuse.
Psychological Bulletin 1993; 114: 68-79.
57.
Eron L D, Huesmann L R, Zelli A. The role of parental variables in the learning of
aggression. In Pepler D J, Rubin K J (eds.) The development and treatment of childhood
aggression. Hillsdale, NJ: Lawrence Erlbaum 1991: 169-188.
58.
Widom C S. The cycle of violence. Science 1989; 244: 160-166.
59.
Smith C, Thomberry T P. The relationship between childhood maltreatment and adolescent
involvement in delinquency. Criminology 1995; 33: 451-481.
60.
Morash M, Rucker L. An exploratory study of the connection of mother's age at
childbearing to her children's delinquency in four data sets. Crime and Delinquency 1989;
35:45-93.
61.
Nagin D S, Pogarsky G, Farrington D P. Adolescent mothers and the criminal behavior of
their children. Law and Society Review 1997; 31: 137-162.
62.
Perales A, Sogi C. Conductas violentas en adolescentes: identification de factores de riesgo
para diseno de programa prevent!vo.” In Pimentel Sevilla C. (ed.) Violencia, familia y ninez
en los sectores urbanos pobres. Cecosam. Peru. 1995.
38
63.
Reiss A J, Farrington D P. Advancing knowledge about co-offending: Results from a
prospective longitudinal survey of London males. Journal of Criminal Law and
Criminology 1991; 82: 360-395.
64.
Elliott D S, Menard S. Delinquent friends and delinquent behavior: Temporal and
developmental patterns. In Hawkins J D (ed.) Delinquency and crime: Current theories.
Cambridge, England: Cambridge University Press 1996: 28-67.
65.
Elliott D S. Huizinga D, Menard S. Multiple problem youth: Delinquency, substance use,
and mental health problems. New York: Springer-Verlag 1989.
66.
Giamni RJ, Litvoc J, Neto JE. Agressao fisica e classe social.” Revista de Saude Publica.
1999;33(2): 180-6.
67.
Hawkins J D, Herrenkohl T, Farrington D P, Brewer D, Catalano R F, Harachi T W. A
review of predictors of youth violence. In Loeber R, Farrington D P (eds.) Serious and
violent juvenile offenders: Risk factors and successful interventions. Thousand Oaks, CA:
Sage 1998: 106-146.
68.
Howell J C. Juvenile justice and youth violence. Thousand Oaks, CA: Sage 1997.
69.
Farrington D P, Loeber R. Major aims of this book. In Loeber R, Farrington D P (eds.)
Serious and violent juvenile offenders: Risk factors and successful interventions. Thousand
Oaks, CA: Sage 1998: 1-9.
70.
Blumstein A. Youth violence, guns and the illicit-drug industry. Journal of Criminal Law
and Criminology 1995; 86: 10-36.
71.
Thomberry T P. Membership in youth gangs and involvement in serious and violent
juvenile offending. In Loeber R, Farrington D P (eds.) Serious and violent juvenile
offenders: Risk factors and successful interventions. Thousand Oaks, CA: Sage 1998: 147166.
72.
Lizotte A J, Tesoriero J M, Thombeny T P, Krohn M D. Patterns of adolescent firearms
ownership and use. Justice Quarterly 1995; 11: 51-73.
73.
van Kammen W B, Loeber R. Are fluctuations in delinquent activities related to the onset
and offset in juvenile illegal drug use and drug dealing? Journal of Drug Issues 1994; 24:
9-24.
74.
De Souza Minayo MC. Fala, galera: juventude, violencia e cidadania.” Garamond. Rio de
Janeiro, 1999.
39
75.
Cohen L E, Felson M. Social change and crime rate trends: A routine activity approach
.American Sociological Review 1979; 44: 588-608.
76.
Ekblom P. Getting the best out of crime analysis. London: Home Office Police Departmen
1988.
77.
Barker M, Geraghty J, Webb B, Kay T. The prevention of street robbery. London: Home
Office Police Department 1993.
78.
LeBlanc M, Frechette M. Male criminal activity from childhood through youth. New York
Springer-Verlag 1989.
79.
LeBlanc M. Changing patterns in the perpetration of offenses over time: Trajectories fron
early adolescence to the early 30's. Studies on Crime and Crime Prevention 1996; 5: 151165.
80.
Agnew R. The origins of delinquent events: An examination of offender accounts. Journal
of Research in Crime and Delinquency 1990; 27: 267-294.
81.
Farrington D P. Motivations for conduct disorder and delinquency. Development anc
Psychopathology 1993: 5: 225-241.
82.
Wikstrom P-0 H. Everyday violence in contemporary Sweden. Stockholm: National
Council for Crime Prevention 1985.
83.
Miczek K A, DeBold J F, Haney M, Tidey J, Vivian J, Weerts E M. Alcohol, drugs oi
abuse, aggression and violence. In Reiss A J, Roth J A (eds.) Understanding and preventins
violence. Vol. 3: Social influences. Washington DC: National Academy Press 1994: 377570.
84.
Wolfgang M E. Patterns in criminal homicide. Philadelphia: University of Pennsylvania
Press 1958.
85.
McClintock F H. Crimes of violence. London: Macmillan 1963.
86.
Pallone N J, Hennessy J H. Tinderbox criminal violence: Neurogenic impulsivity, risk
taking, and the phenomenology of rational choice. In Clarke R V, Felson M (eds.)
Advances in criminological theory. Vol 5: Routine activity and rational choice. New
Brunswick, NJ: Transaction 1993.
87.
Sampson R J, Lauritsen J L. Violent victimization and offending: Individual, situational,
and community-level risk factors. In Reiss A J, Roth J A (eds.) Understanding and
preventing violence. Vol. 3: Social influences. Washington, DC: National Academy Press
1994:1-114.
40
88.
Hough M, Sheehy K. Incidents of violence: Findings from the British Crime Survey. Home
Office Research Bulletin 1986; 20: 22-26.
89.
Moser C, Shrader E A Conceptual framework for violence reduction. LCR Sustainable
Development Working Paper, No. 2. The World Bank. Washington, D.C. August 1999.
90.
Fajnzylber P, Lederman D, Loayza N Inequality and violent crime. Regional Studies
Program, Office of the Chief Economist for Latin America and the Caribbean. The World
Bank. Washington, D.C., December 1999.
91.
Noronha, C.V. et al. “Violencia, etnia e cor: um estudo dos diferenciais na regiao
metropolitana de Salvador, Bahia, Brasil.” In Panamerican Journal of Public Health, Vol
5(4/5). 1999.
92.
Sanjuan AM. Juventude e violencia em Caracas. Paradoxos de um process© de perda da
cidadama. In: Paulo Sergio Pinheiro (ed) Sao Paulo Sem Medo: Um Diagnostico Da
Violencia Urbana. Editora Garamond. Rio de Janeiro, Brasil. 1998.
93.
Aitchinson J. Violencia e juventude na Africa do Sul: causas, li^oes e solu^oes para uma
sociedade violenta. In Paulo Sergio Pinheiro (ed). Sao Paulo Sem Medo: Um Diagnostico
Da Violencia Urbana. Editora Garamond. Rio de Janeiro, Brasil. 1998.
94.
Pampel FC, Gartner R. Age structure, socio-political institutions, and national homicide
rates. European Sociological Review, 1995; 11 (3):243-260.
95.
Messner SF, Rosenfeld R. Political restraint of the market and levels of criminal homicide:
a cross-national application of institutional-anomie theory. Social Forces, 1997;75(4)-13931416.
96.
Lederman D, Loayza N, Menendez AM Violent crime: does social capital matter? The
World Bank, Washington, D.C., July, 1999.
97.
Wilkinson RG, Kawachi I, Kennedy BP. Mortality, the social environment, crime and
violence. Sociology of Health and Illness. 1998;20(5):578-597.
98.
Ayres RL. Crime and violence as development issues in Latin America and the Caribbean.
The World Bank. Washington, D.C. 1998.
99.
Moser C, Holland J. Urban poverty and violence in Jamaica.” World Bank Latin American
and Caribbean Studies: Viewpoints. Washington, D. C. 1997.
100.
Orpinas, P. “Who is violent?: factors associated with aggressive behaviors in Latin America
and Spain.” Pan American Journal of Public Health, Vol. 5, No. 4. 1999.
101.
Bedoya Marin DA and Jaramillo Martinez J. De la barra a la banda. Editorial El Propio
41
Bolsillo. Medellin. 1991.
102.
Rodgers D. Youth gangs and violence in Latin America and the Caribbean: a literature
survey.” LCR Sustainable Development Working Paper No.4, Urban Peace Program Series.
The World Bank. Washington, D.C. August 1999.
103.
Schneidman M. Targeting at-risk youth: rationales, approaches to service delivery and
monitoring and evaluation issues. LASHC Paper Series No. 2. July 1996.
104.
Lauras-Loch T, Lopez-Escartin N. Jeunesse et demographie en Afrique. In d'Almeida-Topor,
H. et al. Les Jeunes en Afrique: Evolution et role (XE-C-XX6 siecles).
105.
UNICEF and WHO. A Picture of Health? A review and annotated bibliography of the health
of young people in developing countries. WHO/ADH/95.14. Geneva. 1995.
106.
Diallo Co-Trung, M. "La crise scolaire au Senegal: crise de 1'ecole, crise de I'autorite?" In
d'Almeida-Topor, H. et al. Op. cit.
107.
Rarrbo, K. Op. cit.
108.
Dinnen, S. “Urban Raskolism and Criminal Groups in Papua New Guinea.” In Hazlehurst,
K. and Hazlehurst, C. Op. cit.
109.
UNICEF. “Children at Risk in Central and Eastern Europe: Perils and Promises.” The
Monee Project, Regional Monitoring Report No. 4. International Child Development
Centre. Florence, Italy. 1997.
110.
Dahlberg LL, Potter LB. Youth violence: developmental pathways and prevention
challenges. American Journal of Preventive Medicine. (In press).
111.
Tolan PH, Guerra NG. What works in preventing adolescent violence: an empirical review
of the field. Boulder, CO: Center for the Study and Prevention of Violence, 1994.
112.
Elliott DS. Editor’s Introduction: Blue Prints for Violence prevention.. Boulder, CO: Center
for the Study and Prevention of Violence, 1997.
113.
Loeber R, Hay DF. Developmental approaches to aggression and conduct problems. In:
Rutter M and Hay DF (eds.) Development through life: a handbook for clinicians. Malden,
MA: Blackwell Scientific, 1994:488-516.
114.
Mirsky AF, Siegel A. The neurobiology of violence and aggression. In: Reiss AJ, Miczek
KA, Roth JA (eds.) Understanding and preventing violence. Vol 2: Biobehavioral
influences. Washington, DC: National Academy Press, 1994:59-172.
42
115.
Loeber R, Stouthamer-Loeber M. Family factors as correlates and predictors of juvenile
conduct problems and delinquency. In: Tonry M, Morris N (eds.) Crime and Justice. Vol
7. Chicago, IL: University of Chicago Press, 1986:29-150.
116.
McCord J. A forty year perspective on the effects of child abuse and neglect. Child Abuse
and Neglect 1983; 7:265-70.
117.
Tolan PH. Socioeconomic, family, and social stress correlates of adolescent antisocial and
delinquent behavior. J Abnormal Child Psychol 1988; 16:317-31.
118.
Kruttschnitt CL, Ward D, Scheble MA. Abuse-resistant youth: some factors that may inhibit
violent criminal behavior. Social Forces 1987; 66:501-19.
119.
George C, Main M. Social interactions of young abused children: approach, avoidance, and
aggression. Child Develop 1979; 50:306-18.
120.
Reidy TJ. The aggressive characteristics of abused and neglected children. J Clinical
Psychol 1977;33:1140-45.
121.
Reiss AJ, Roth JA (eds.). Understanding and preventing violence. Vol 1. Washington, DC:
National Academy Press, 1993.
122.
Attar BK, Guerra NG, Tolan PH. Neighborhood disadvantage, stressful life events, and
adjustment in urban elementary-school children. J Clinical Child Psychol 1994; 23:391-400.
123.
National Research Council. Losing generations:
Washington, DC: National Academy Press, 1993.
124.
Kellermann AL, Fuqua-Whitley DS, Rivara FP, Mercy JA. Preventing youth violence: what
works? Annu Rev Public Health 1998;19:271-292.
125.
Johnson DL, Walker T. Primary prevention of behavior problems in Mexican-American
children. Am J Commun Psychol 1987;15:375-85.
126.
Lally JR, Mangione PL, Honig AS. The Syracuse University Family Development Research
Project: long-range impact of an early intervention with low-income children and their
families. In Annual Advances in Applied Developmental Psychology: Parent Education as
an Early Childhood Intervention, Powell DR (ed.), Norwood, NJ: Ablex, 1988: 59.
127.
Seitz V, Rosenbaum LK, Apfel NH. Effects of a family support intervention; a 10-year
follow-up. Child Dev 1985;56:376-391.
128.
Olds DL et. al. Long-term effects of nurse home visitation on children’s criminal and
antisocial behavior: 15-year follow-up of a randomized controlled trial. JAMA 1998; 123844.
43
adolescents in high-risk settings.
129.
Farington DP, Welsh BC. Delinquency prevention using family-based interventions.
Children and Society 1999;13(37):287-303.
130.
Patterson GR, Capaldi D, Bank L. An early starter model for predicting delinquency. In
Pepler DJ, Rubin KH (eds.). The Development and Treatment of Childhood Aggression.
Hillsdale, NJ: Lawrence Erlbaum, 1991: 139-168.
131.
Patterson GR, Reid JB, Dishion TJ. Antisocial Boys. Eugene, OR: Castalia, 1992.
132.
Hawkins JD, Von Cleve E, Catalano RF. Reducing early childhood aggression: results of
a primary prevention program. J Am Acad Child Adolesc Psychiatr 1991; 30:208-217.
133.
Tremblay RE, Vitaro F, Bertrand L, LeBlanc M, Beauchesne H, et al. Parent and child
training to prevent early onset of delinquency: The Montreal Longitudinal Experimental
Study. In McCord J, Tremblay RE (eds.). Preventing Antisocial Behavior: Interventions
from Birth Through Adolescence. New York: Guilford, 1992:117-138.
134.
Greenwood PW, Model KE, Rydell CP, Chiesa J. Diverting children from a life of crime:
measuring costs and benefits. Santa Monica, CA: RAND, 1996.
135.
Berrueta-Clement JR, Schweinhart LJ, Barnett WS, Epstein AS, Weikart DP. Changed
Lives: The Effects of the Perry Preschool Program on Youth Through Age 19. Ypsilanti,
MI: High/Scope Press, 1984.
136.
Schweinhart LJ, Bames HV, Weikart DP. Significant Benefits: The High/Scope Perry
Preschool Project Study Through Age 27. Ypsilanti, MI: High/Scope Press, 1993.
137.
Ingersoll GM. Adolescents in school and society. Lexington, MA: DC Heath & Co 1982.
138.
Patterson GR, Stouthamer-Loeber M. The correlation of family management practices and
delinquency. Child Develop, 1984;55:1299-1307.
139.
Huizinga D, Loeber R, Thomberry TP. Urban delinquency and substance abuse: initial
findings. Washington, DC: US Department of Justice, Office of Juvenile Justice and
Delinquency Prevention, 1994:1-27.
140.
Thomberry TP, Lizotte AJ, Krohn MD, Famworth M, Jang SJ. Delinquent peers, beliefs,
and delinquent behavior: a longitudinal test of interactional theory. Criminology 1994;
32:601-37.
141.
Thomberry TP, Lizotte AJ, Krohn MD, Famworth M, Jang SJ. Testing interactional theory:
an examination of reciprocal causal relationships among family, school, and delinquency.
J Criminal Law Criminol 1991; 82:3-35.
44
142.
Lipsey MW, Derzon JH. Predictors of violent or serious delinquency in adolescence and
early adulthood: a synthesis of longitudinal research. In: Loeber R, Farrington DP (eds.)
Serious and violent juvenile offenders: risk factors and successful interventions. Thousand
Oaks, CA: Sage 1998:86-105.
143.
Maguin E, Loeber R. Academic performance and delinquency. In: Tonry M. (ed.) Crime
and justice. Vol 20. Chicago, IL: The University of Chicago Press, 1996:145-264.
144.
Richards BA, Dodge KA. Social maladjustment and problem solving in school-aged
children. J Consult. Clin Psychol 1982;50:226-33.
145.
Guerra NG, Williams KR. A Program Planning Guide for Youth Violence Prevention: A
Risk Focused Approach. Boulder, CO: Center for the Study and Prevention of Violence,
1996.
146.
Hawkins JD, Catalano RF, Kosterman R, Abbott R, Hill KG. Preventing adolescent health
risk behaviors by strengthening protection during childhood. Archives of Pediatrics &
Adolescent Medicine. 1999; 153(3):226-34.
147.
Howell JC, Bilchick S (eds.). Guide for Implementing the Comprehensive Strategy for
Serious Violent and Chronic Juvenile Offenders. Office of Juvenile Justice and Delinquency
Prevention, Washington, D.C. NCJ-153681. 1995.
148.
Mihalic SF, Grotpeter JK. Blueprints for Violence Prevention: Big Brothers/Big Sisters of
America, Book Two. Boulder, CO: Center for the Study and Prevention of Violence, 1997.
149.
Grossman JB, Garry EM. Mentoring -- A Proven Delinquency Prevention Strategy. Juvenile
Justice Bulletin. U.S. Department of Justice, Office of Justice Programs, 1997.
150.
Shadish WR. Do family and marital psychotherapies change what people do? A meta
analysis of behavior outcomes. In Cook TD, Cooper H, Cordray DS, Hartmann H, Hedges
LV, Light RJ (eds.) Meta-Analysis for Explanation: A Casebook. New York: Russell Sage
Foundation, 1992.
151.
Hazelrigg MD, Cooper HM, Borduin CM. Evaluating the effectiveness of family therapies:
an integrative review and analysis. Psychological Bulletin 1987;101:428-442.
152.
Klein NC, Alexander JF, Parsons BV. Impact of family systems intervention on recidivism
and sibling delinquency: a model of primary prevention and program evaluation. J Consult
Clin Psychol 1977;45:469-474.
153.
Aos S, Phipps P, Bamoski RA, Lieb R. The Comparative Costs and Benefits of Programs
to Reduce Crime. Olympia, WA: Washington State Institute for Public Policy.
45
154.
Henggler SW, Schoenwald SK, Borduin CM, Rowland MD, Cunningham PB.
Multisystemic Treatment of Antisocial Behavior in Children and Adolescents, New York:
Guilford, 1998.
155.
Paniter KA, Farrington DP. Evaluating situational crime prevention using a young people’s
survey. British Journal of Criminology. (In press).
156.
Farrington DP. The development of offending and antisocial behavior from childhood: Key
findings from the Cambridge study in delinquent development. J Child Psy and Psychiatry
1995; 36:929-64.
157.
Guerrero R, Concha-Eastman A. Programme summary prepared for the World Report on
Violence and Health. 2000.
158.
ESSOR. Avaliasao 1998 do trabalho realizado pela ESSOR com os adolescentes de dois
bairros de Maputo/ Mozambique. Programmes de Developpement ESSOR. Maputo,
Mozambique. 1999.
159.
Barker G, Fontes M. Review and Analysis of International Experience with Programs
Targeted on At-Risk Youth. LASHC Paper Series No. 5. The World Bank, Washington,
D.C., 1996.
160.
Munoz Chacon, S. Politicas hacia la adolescencia y juventud en Costa Rica. In Ramos CG
(ed). America Central en los noventa: Problemas de Juventud. FLACSO Programa El
Salvador, julio 1998.
46
Chapter 3
Child Maltreatment
Status:
Draft: 3rd
Peer Reviewed: Yes
Date of Current Draft: 12 October 2000
i
TABLE OF CONTENTS
I.
IL
III.
IV.
V.
VI.
VII.
vm.
IX.
X.
XI.
Introduction
International Issues in Defining Maltreatment
Epidemiology
Dynamics of Maltreatment
A.
Children Characteristics
B.
Caregiver Characteristics
C.
Family and Societal Characteristics
Etiology
Consequences of Maltreatment
A.
Health Burden
B.
Child Death
C.
Financial Burden
Recognition / Training of Health Professionals
Intervention
Prevention
Policy Responses
Summary
2
INTRODUCTION
An international literature has emerged to document that violence against children, by their
caretakers, is a international health concern (e.g.: 4, 6, 13, 19, 24, 39, 42, 43, 50, 51, 73, 80,
92). While child maltreatment has a long history; well documented in literature, art; and
science in many parts of the world; its recognition as a global public health problem is a
recent phenomenon (42)(92). Investigations, in both developed and lesser-developed
countries, have demonstrated significant initial and long-term harm (e.g. 2, 6, 8-10, 30, 33,
67). There is some evidence that wide public recognition and public policy interventions have
reduced the incidence of maltreatment in some countries (19)(26) but the scale of the problem
remains massive even in those countries. The medical literature demonstrates that clinical
skills in the of child maltreatment recognition by medical professionals in countries with a
longer history of awareness are much less than ideal and that even in developed countries,
there are few hours devoted to child maltreatment in medical curricula (1)(44)(71).
Professional education for health professionals will need to be upgraded to include the
recognition and appropriate response to child maltreatment. Prevention of child maltreatment
on a global scale will require not only education of health professionals but education of the
citizenry. Public policy efforts directed at educating parents and setting up systems to
respond to children in need are needed in every country. There is a clear need for public
health research to document the epidemiology and consequences of maltreatment in many
developing countries as well as research into effective prevention and intervention strategies.
INTERNATIONAL ISSUES IN DEFINING MALTREATMENT
While there is widespread agreement that child abuse should end; there is less agreement,
both within and between societies, about which specific acts are abusive (42, 48). There is
also widespread disagreement about which acts of omission in care of a child constitute
unacceptable child neglect. Some experts define abuse and neglect by focusing on the
behaviours or actions of the parents while others include the consequences of maltreatment
for the children or the intention of the parent as part of the definition. Some observers argue
that only violence from parents or caretakers should be included, others include children
harmed by other children or in other social relationships, such as the poor treatment of
children in government institutions. In a recent survey of key respondents in 47 countries, 20
potentially abusive or neglectful actions were listed and respondents were asked to indicate
whether these actions were defined as child abuse and neglect (42). Every listed action had at
least 14 respondents who reported that this action would be classified as abusive or neglectful
in his or her country; however the highest rate of endorsement (adult use of a child for sexual
gratification) was endorsed by just 95% of the respondents. Physical beating of a child was
considered abusive in the country of the respondent for 75% of the countries.
3
Although a universally agreed-upon definition of child maltreatment may not be attainable,
discussions among child advocates from 27 countries have produced the following definition:
‘Child abuse or maltreatment constitutes all forms of physical and/or
emotional ill-treatment, sexual abuse, neglect or negligent treatment or
commercial or other exploitation, resulting in actual or potential harm to the
child's health, survival, development or dignity in the context of a relationship
ofresponsibility, trust or power.' (96)
This broad definition reflects the observation that child maltreatment is commonly
conceptualised as including four specific sub-types of maltreatment: physical abuse,
emotional abuse, sexual abuse and neglect. Definitions are offered below for these sub-types.
There is growing, but not universal, acknowledgement of the damaging effects of child
exploitation in occupations such as prostitution, labour in dangerous or debilitating
conditions, and as soldiers, child exploitation is explicitly included in the general definition.
The Convention on the Rights of the Child (see box) addresses some of these issues.
Physical abuse: Physical abuse of a child results from an act of commission which produces
actual physical harm, or holds the potential for harm which is reasonably within the control of
a parent or person in a position on responsibility, power or trust. There may be single or
repeated incidents.
Figure 1: The International Convention on the Rights of the Child
The CRC asserts the right of children to be protected from abuse and maltreatment and
contains a number of articles which amplify this right. The CRC provides an international
framework for addressing and responding to violence against children. It sets down principles
that can guide efforts to prevent violence against and amongst children. Among these
principles are: Children should be treated without discrimination (Article 2); their best
interests should be the primary concern (Article 3); and their views should be taken into
account in a manner consistent with the maturity and evolving capacities of the child (Article
12). Article 19 specifies that States have an obligation to protect children from all forms of
violence and outlines measures that might be taken in this regard. Other articles draw
attention to groups of children who are at particular risk from violence: children who are
sexually exploited (Article 34), or exploited through work (Article 32), children who misuse
substances (Article 33), children in armed conflict (Article 38) and children who are disabled
(Article 23). The Convention explicitly recognises that many adults play a part in protecting
children from violence, drawing attention to the responsibilities of parents (Article 5), and the
vulnerability of children deprived of their parents (Article 9). It recognises that sometimes
children may need to be cared for outside the family (Article 20), and sets down conditions
for that care (Article 25). Article 39 sets out the importance for the child of having the
opportunities to recover from violence and be able to reintegrate into society in an
environment that fosters the health, self-respect and dignity of the child. Other Articles in the
Convention emphasise the important role of the health-care community in monitoring and
reporting child abuse, as a channel for advocacy and direct technical support to countries. The
convention offers a broad international perspective in an area complicated by different
cultural conceptions of appropriate discipline and child care standards. The CRC indicates a
number of aspects of the treatment of children that are almost universally accepted standards
while defining other actions as completely unacceptable.
4
Sexual abuse: Sexual abuse involves acts which use a child for sexual gratification of an
adult either by inappropriate exposure to adult sexuality, by direct sexual contact, or by
making the child available to others either directly for immoral and illegal acts, such as
through pornography.
Emotional abuse: Emotional abuse involves intentional acts which lead to a failure to
provide a developmentally appropriate, supportive environment, including the availability of
a primary' attachment figure, or acts which themselves have an adverse effect on the
emotional health and development of a child. A supportive environment is necessary for the
child to develop a full range of emotional and social competencies commensurate with her or
his personal potential and in the context of the society in which the child lives. Emotionally
abusive acts include: restricting movement, belittling, denigrating, scapegoating, threatening,
scaring, discriminating, ridiculing or other non-physical forms of hostile or rejecting
treatment.
Neglect and negligent treatment: Neglect is the failure to provide for the development of
the child in one or more of the following spheres: health, education, emotional development,
nutrition, shelter, and safe living conditions. Neglect is distinguished from poverty, by many
observers, in that neglect requires the failure to provide for the child when there are resources
reasonably available to the family or caretakers and the neglect causes, or has a high
probability of causing, harm to the child’s health or physical, mental, spiritual, moral or social
development. This definition explicitly includes the failure to properly supervise and protect
children from harm when supervision and protection is feasible.
EPIDEMIOLOGY
Despite the difficulties noted above in defining abuse and neglect, data have been collected in
many countries of the world on the epidemiology of child maltreatment. This research has
focused on physical and sexual abuse; much less is known about the epidemiology of neglect,
emotional abuse and other forms of maltreatment. At least 26 countries collect official
statistics on reported maltreatment (42). A WHO European Region initiative identified
current statistics on child abuse for 10 countries,1 either in health or legal systems (29).
There is an epidemiological tradition of international comparative research generating new
ideas about prevention and treatment for major public health problems (e.g. 47). Similarly
epidemiological studies of child maltreatment may help mobilise professional attention and
public opinion. Comparative studies among cultures and countries may lead to new
hypotheses about the roots of maltreatment and suggest promising directions for intervention.
However, research on child abuse is harder than most other social science research because of
methodological difficulties and special ethical issues (77)(81). Studies which are needed (56),
and which have only recently been attempted (43) are studies which use common definitions
and instruments to capture similar information across cultures and countries.
Legal and social systems with responsibility for responding to, and counting, child abuse and
neglect reports are not found in many countries (42). Because the legal and cultural
definitions of abuse and neglect vary among countries, and the majority of countries don’t
1 Bosnia Hezegovina, Croatia, Hungary, Italy, Kazakhstan, Latvia, Macedonia, The Netherlands, Poland, Russia.
5
require mandatory reports, there is little to be gained by directly comparing rates of official
maltreatment in different countries. Case reports and case registries are most useful as a
source of local data to guide local action but the use of this type of data assumes public or
professional recognition of maltreatment. More compelling data may come from surveys of
parental behaviours. Sound population-based surveys of the frequency of potentially abusive
behaviours may lead to broader public and professional recognition of child abuse and neglect
and influence both professional education and public policy.
With more than two decades of research, the most common approach to measuring the
prevalence of abusive behaviours has been the Conflict Tactics Scale (CTS) (84)(86).
Developed from conflict theory, the CTS scale ascertains the frequency of specific behaviours
used to resolve conflict. The original scale was based around three areas of conflict
resolution: rational discussion or ‘reasoning ’, the use of verbal and non verbal acts to
symbolically hurt the other or ‘verbal aggression’ and the use of physical force,. A recent
revision of the CTS has produced an instrument to specifically measure parent-child conflict
(PCCTS) (90). The scale rates behaviours according to their frequency both over the past year
and ever, providing estimates of both incidence and lifetime prevalence. It is sensitive to
different levels of violence severity, is relatively easy to complete and avoids confounding
measurement of the behaviour with either causes of the behaviour (such as dominance or
aggressiveness) or the consequences (i.e. injury). The hazard of defining abuse by bad
outcomes could otherwise lead to an abusive act not being considered problematic unless it
results in injury. Contexts and causes are not included in the CTS but these issues can be
included with it in population surveys, Population surveys using the CTS may be effective
tools to increase awareness in societies that have not previously recognised child abuse and
neglect. Although the original CTS was developed to measure partner violence, it has been
widely used to assess abusive behaviours toward children with over 130 publications using it
as a measure of child maltreatment (88). The 1995 version of the CTS dramatically improved
its use as a tool to assess child abuse (90).
The behaviours specified in the CTSPC (see Figure 2) require refinement for different
languages and cultural contexts but provide a foundation for the collection of baseline data.
The scale is designed as a questionnaire for parents although it has been adapted to obtain
child or adult recall of childhood experiences (13)(20)(39). The legal environment differs
between countries; in some countries it is illegal to hit children (for example, Sweden,
Norway, & Denmark) whereas in others it is legal to physically punish using objects (for
example, Germany). The criteria for determining prevalence rates of physical abuse may vary
according to statute in the country where the study is undertaken. One advantage to
ascertaining the prevalence of a variety of behaviours intended for discipline is that
investigators may use develop differing definitions of what is abusive while depending upon
the same instrument to collect the data. There are certain criteria that must be met before the
scale is applied since it requires a standardised introduction and set of response options 2
Applying the CTSCP: The introductory explanation and the response categories need to be added. Both
have been shown to make an important difference in the findings. These are as follows:
Children often do things that are wrong, disobey, or make their parents angry. We would like to know what you
have done when your (SAY age of referent child) year old child, did something wrong or made you upset or
angry.
6
Fig. 2: CTSPC items
Scale and Itenv;
■
-
•
.
.
•
■
-
'
'
1.,.
•
:
<
■
'
'
:
\
...........................
:
■
'•
~~
■■
■
■
’
<
.
?■'■■■
‘
■
,
•
■
'
'
'
•
■
■
.
’•
.
Non-violent Discipline
A. Explained why something was wrong
E. Gave him/her something else to do instead of what he/she was doing
Q. Took away privileges or grounded him/her
B. Put in ‘timeout’ (or sent to room).
Psychological Aggression
F. Shouted, yelled, or screamed at him/her
N. Threatened to spank or hit but did not actually do it
J. Swore or cursed at him/her
U. Called him/her dumb or lazy or some other name like that
L. Said you would send him/her away or kicked him/her out of the house
C-.-1
•
1Sil?a*lkedA hini711c1r °“ U1C 1’0tt°'vith yoar bare hand
p.^hm^r on the bottom with a hard oly
P-SJapped him/her on the hand, arm or leg
C^ho^hrmtoiGhrldaged^orolder
Severe Assault (Physical Abuse)
. ..
.^Slapped on the face, head or ears ; .
• Q^Hit someotherpart of the body besides the bottom with a hard object
TThrewortockeddown'^; /
,
Very Severe Assault (Severe Physical Abuse)
K/Beat up,'that is you hit him/her over and over as hard as you could ' '.
I Grabbed around neck and choked: ; . .
.
a;
M. Burned or scolded on purpose
S^e^enedwrthaknifeorgun IS
I am going to read a list of things you might have done in the past year and I would like you to tell me whether
you have: done it once in the past year, done it twice in the past year, 3-5 times, 6-10 times, 11-20 times, or more
than 20 times in the past year. If you haven’t done it in the past year but have done it before that, I would like to
know this, too.”
Coding Responses:
1 = Once in the past year
2 = Twice in the past year
3 = 3-5 times in the past year
4 = 6-10 times in the past year
5 = 11-20 times in the past year
6 = More than 20 times in the past year
7 = Not in the past year, but it happened before
0 = This has never happened
7
Neglect
NA. Had to leave your child home alone, even when you thought some adult should
be with him/her
NC Were not able to make sure your child got the food he/she needed
NE. Were so drunk or high that you had a problem taking care of your child
ND. Were not able to make sure your child got to a doctor or hospital when he/she
needed it
NB Were so caught up with problems that you were not able to show or tell your
child that you loved him/her
Copyright, 1995. For permission to use this instrument, contact Murray A.
Straus at MAS2@CISUN1X.UNH.EDU(Source: (90))
8
An identically modified version of the parent-child CTS was used in 4 countries by the
WorldSAFE (World Studies of Abuse and the Family Environment) consortium of the
International Clinical Epidemiology Network to increase attention to the problem of
maltreatment among health professionals in lesser-developed countries (43). WorldSAFE is a
project of a team of clinical epidemiologists in paediatrics, internal medicine, and psychiatry,
as well as statisticians and social scientists from 15 countries, (see http://wwvv.inclen.org for
additional information.) WorldSAFE developed a common core protocol to ascertain the
prevalence of abusive child discipline behaviours and domestic conflict behaviours in
population-based samples. Pilot projects have been fielded in the Philippines, India, Egypt,
Chile and Brazil.
Focus groups in India, Chile and Egypt were used to confirm the
appropriateness of the questions in the instrument; a number of additional questions were
added by the focus groups. A table of preliminary results from individual communities in
that project along with comparative data from a national random-digit dial telephone poll in
the United States, which used the PCCTS, appears below.
The CTS does not eliminate the difficulties inherent in the international study of violence
directed against children. Difficulties remain despite the use of common instrumentation
because different sample populations and techniques for data collection have been used. Data
from the US suggest that there is wide variation in rates by family composition and social
class (90). Further, the experience of the WorldSAFE investigators is that there are some
behaviours that are relatively unique to specific countries or which carry different meanings
for parents in other countries. For example WorldSAFE data, shown in table 1 below,
indicate that slapping on a child’s face or head was a relatively common disciplinary method
in India while it is uncommon in the United States, Egypt, Chile or the Philippines.
Screaming at children appeared to be universal while hitting the child on the buttocks with an
object appears at a lower frequency in most countries. Analyses to examine confounding by
social class and education are still underway. Comparable US data reveal very strong
differences by social class and whether the family is a single parent household (90).
Stratification by social class may reveal more similarities than differences. Table 1 presents
WorldSAFE findings comparing the prevalence of self-reported specific parental behaviours
in communities in Chile, Egypt, India, and the Philippines with previously collected data
from the US collected in a national telephone survey (90).
These data demonstrate that child maltreatment is not isolated to a few countries or just one
hemisphere. The WorldSAFE investigators chose not to define an overall physical abuse rate
by combining rates for specific items on the Conflict Tactics Scale but instead report the
variations in frequency of specific acts. If hitting a child with an object, elsewhere besides
the buttocks were defined as physical abuse, three of the commumties surveyed have much
higher rates than those published for the US. Other authors have made conclusions about
overall rates. Data are presented in Appendix 1 which reveal published prevalence rates for
physical abuse ranging from 14.7% (46) using a self assessed definition which asked
respondents whether they thought they had been physically abused to 93.8% (20) using a
standardised questionnaire. An estimated mean prevalence rate for physical abuse across
studies reported in the 1990’s is 31% (see Appendix 1).
9
Tablet: Reports of Disciplinary Practice Use in Specific Communities in 4
countries and in a US Telephone Survey Using the Modified PC-CTS#
Non-violent discipline
Explained why
Took privileges
Told to stop
Gave something to do
Stay in one place
Chile
91%
60%
88%
71%
37%
Verbal Punishment
Yelled or screamed
84%
Called names
15%
Cursed
3%
Refused to speak
17%
Threatened to kick-out
5%
Threatened abandonment
8%
Threatened evil spirits
12%
Moderate
Physical
Punishment
51%
Spanked butt (with hand)
18%
Hit on butt (with object)
Slapped on head
13%
24%
Pulled hair
Shook him or her
39%
12%
Hit with knuckles
Pinched
3%
27%
Twisted ear
Forced stand burdened
0%
Put hot pepper in mouth
0%
2%
Locked out
Physical
Severe
Punishment
4%
Hit with object (not butt)
Kicked
0%
Burned
0%
Beat-up
0%
0%
Knife-gun threat
0%
Choked
80%
27%
69%
43%
50%
Philippines Rural
USA
India(43) (90)
94%
94%
90%
3%
43%
77%
*
91%
66%
27%
75%
58%
5%
75%
72%
44%
51%
48%
0%
10%
6%
82%
24%
0%
15%
26%
48%
24%
29%
28%
41%
29%
59%
25%
45%
31%
6%
2%
1%
26%
2%
2%
25%
0%
1%
Egypt
70%
29%
85%
17%
24%
31%
*
6%
20%
20%
*
*
75%
51%
21%
23%
20%
8%
60%
31%
4%
1%
12%
58%
23%
58%
29%
12%
28%
17%
16%
2%
3%
47%
21%
4%
21%
6%
0%
3%
1%
1%
36%
10%
1%
*
9%
*
5%
*
*
■k
*
1%
2%
4%
0%
0%
0%
0%
0%
* Not collected in 1995 Gallup Survey of US (90).
# WorldSAFE (World Studies of Abuse and the Family Environment) is a consortium of investigators from the
International Clinical Epidemiology Network representing medical schools in India, Chile, the Philippines,
Egypt, Brazil and the United States who have been fielding population based studies of domestic violence and
child abuse using a common protocol, (see 45 & http://www.INCLEN.ORG for information about the
investigators and project).
10
Surveys estimating prevalence rates for sexual abuse vary dramatically in their estimates
depending upon the definitions and methods used in the study. The US literature has revealed
wide variations in rates depending upon how questions are asked. Because sexuality and
sexual acts with children are difficult to talk about in many societies, there are little
international data on the prevalence of child sexual abuse. Surveys about sexual abuse may
be compromised in that parents may not be aware of what has happened to a child or a parent
may refuse to report acts that may place the parent in legal jeopardy. For example, incest is
punishable by death in the Philippines. Most existing prevalence surveys of sexual abuse ask
adults about their own childhood while physical abuse surveys ask parents about acts toward
their own children... Few surveys have children directly about their sexual experiences.
Among published studies, prevalence rates for men, reporting about their own childhood,
range from 1% (58), using a narrow definition of sexual contact with pressure or force, to
19% (33), using a broader definition. Lifetime prevalence rates for child sexual victimisation
among adult women range from 0.8% (14) using rape as the definition to 45% (33) with a
much wider definition. Using studies reported in international journals over the 1990’s a
mean lifetime prevalence rate of childhood sexual victimisation of women as girls is 19% and
men as boys is 7%.
Obviously, the wide variations in published prevalence estimates could result from real
differences in risk in different cultures or from differences in study design (definitions,
sampling and method of data collection) (37). For child sexual abuse, including peer abuse
can increase prevalence by 9% (30) and including contact and non-contact can raise rates by
approximately 16% (72).
Prevalence studies can provide a means of evaluating changes over time. A national study
was repeated in the US at ten-year intervals from 1975, 1985 and 1995 (86). Reductions in
levels of severe physical punishment to children were noted although overall rates of physical
pumshment did not appear to reduce at the same pace. The cross-sectional surveys did not
address the causes of the decrease in rates but hypotheses include; changes in family
structure, economic prosperity, public awareness of child abuse, and growth in treatment and
intervention programmes (86).
Using public health statistics on the prevalence of child
deaths and children in need of protective intervention, reductions in levels of physical
violence against children have been noted and ascribed to changes in laws about corporal
punishment in Sweden (25). There continues to be debate about the reliability of these claims
(22) which serve to emphasise the importance of and need for comprehensive measures to
evaluate change as awareness of child abuse develops.
DYNAMICS OF MALTREATMENT
Child maltreatment is an entirely socially constructed phenomenon. A variety of theories and
models have been developed to explain its existence. The most successful and predictive
explanatory model or theory is the “ecological model” (5)(37). In this model, there four
contributory components: 1) the characteristics of the individual child, 2) the characteristics
of the caregiver or perpetrator, 3) the characteristics of the family and the immediate domestic
environment of the child, and 4) the community and society in which the child lives which
includes the social, economic and cultural features of the child’s environment. One of the
most consistently associated risk factors for all forms of child abuse is that of a context of
domestic violence and previous victimisation emphasising the links between violent contexts
11
for both adults and children. Some findings are fairly consistent across countries. The factors
specified below for each area have been linked to child maltreatment in more than one study
and most have been associated in more than one country. However, the factors listed may be
only statistically associated and not causally linked (16). Factors Associated with Child
Maltreatment include:
Children Characteristics
•
•
•
•
•
•
Neglect is most commonly reported among the youngest children; they are the most
vulnerable.
There appears to be an increased risk for physical abuse for premature infants,
handicapped children, and twins. The risk of abuse climbs during early childhood.
Physical abuse peaks for children between 9 and 12 years of age in the United States (82)
while in China the risk is highest for 3-6 year olds (87).
Girls are at higher risk for infanticide, educational and nutritional neglect, kidnapping,
sexual abuse, and forced prostitution (18)(33)(45).
Gender differences in risk for physical abuse are not consistent. In China, boys are more
strictly disciplined, perhaps due to the higher parental expectations (83). In neither the
US nor India is there evidence of a difference in harsh physical punishment by gender
(41)(85).
Adolescents appear to be at greatest risk for verbal abuse (41) and sexual abuse (33) (83).
A gender difference exists for victims of sexual abuse with boys having lower rates than
girls do; girls are at two to three times higher risk compared to boys (33).
Caregiver Characteristics
•
•
•
•
Women self-report harsher physical discipline use than men in the US (82) and India (45).
Single mothers are 3 times more likely to self-report the use of harsh physical discipline
than mothers in two-parent families (82). However, in the United States, life-threatening
head injuries, abusive fractures, and fatal child abuse are more commonly perpetrated by
men (75) (76).
In most countries, gender roles influence responsibility for child rearing and determine
responsibility for discipline and care. In focus groups, Indian mothers report nearly sole
responsibility for discipline (43).
Sexual abuse offenders, for both female and male victims, are predominantly men. Rates
of male perpetrators for female victims range from 92.0% (55) to 99.2% (5). For male
victims the range is between 63.2% (38) and 85.7% (3) (33).
Parents maltreated as children are at higher risk of maltreating their own children
although in the US, data suggest that the majority of maltreating parents were not
themselves maltreated (66).
Family and Societal Characteristics
•
•
•
•
•
Poverty is strongly associated with both physical abuse and sexual abuse in many
countries and in many studies (41 )(85)(86).
Child abuse appears more likely in households with parental divorce or separation (31)
Child abuse appears less likely in extended family households or where child care is
shared (41 )(54)(84).
Household crowding appears to increase the risk for children (41)(69).
Substance abuse and family violence are both clear risk factors for maltreatment
(33)(41)(62)(70)(71).
12
ETIOLOGY
Child maltreatment is entirely socially constructed. Maltreatment arises from the family
social environment around the child; it cannot be explained without a thorough consideration
of the role that culture plays in its definition, recognition, cause and effects. Perpetrators and
victims cannot be viewed in isolation; both are subject to social forces. A society may
internalise patterns of social organisation which reinforce violence such as ethnic segregation,
gender-based discrimination and age oppression (43). It may be difficult for cultural insiders
to recognise cultural norms as harmful or violent. Certain practices are taken for granted as
acceptable, even though they may induce pain, injury and trauma.
Initially, research attention focused on simplistic associations between factors, such as
poverty, mental illness, and history of abuse in the family. As the research base developed
there was a realisation that there is no one single cause or set of causes of child maltreatment.
It is not possible to establish simple causal relationships between specific caregiver factors
and different forms of abuse or between specific forms of abuse and specific consequences.
Child abuse is the result of many factors coming together in a complex interplay. An
accumulation of risk factors increases the likelihood of maltreatment. Protective factors also
appear to be additive; they diminish both the likelihood of maltreatment and negative
consequences (76). Social capital, the social networks and support systems that a child and
his or her family can access, appear to provide an important resource to ameliorate risk.
THE CONSEQUENCES OF MALTREATMENT
There are both immediate consequences of maltreatment, including physical and
psychological damage, and harms which can last a lifetime. Research, mostly from the
developed world, which has a longer history of recognition of maltreatment, has demonstrated
acute and long-term physical injuries. Similarly, there are many studies demonstrating short
and long-term psychological harms. A number of developed country studies have illustrated
serious societal consequences including delinquency, school failure, teen-age pregnancy,
suicide, and drug abuse. Relatively limited data exist demonstrating long-term consequences
in lesser developed countries. The consequences for different forms of maltreatment may be
different but there does not appear to be any syndrome or pattern by type of abuse. A recent
review of the harms from sexual abuse concludes that 1/3 or children have no detectable
difficulties following sexual abuse (79). However, two-thirds appear to have difficulties.
Some children have a few symptoms that do not reach clinical levels of concern or are at
clinical levels but which are not as high as children generally seen in clinical settings. Other
children have serious psychiatric symptoms such as depression, anxiety, substance abuse,
aggressivity, shame, or cognitive impairments. Finally some children meet full criteria for
psychiatric illness including post-traumatic stress disorder, major depression, overanxious
disorder and sleep disorder.
Health Burden
Injuries, intentional and unintentional, are a large and neglected health problem in all regions,
accounting for 16% of the global burden of disease in 1998 as measured in DALYs
13
(Disability Adjusted Life-Years). Violence and self inflicted injuries, including suicide, (both
related to child maltreatment (29)) are major public health concerns because of their
increasing contribution to the global disease burden. The burden of ill health caused by child
abuse forms another significant portion of the total burden. While some health consequences
have been researched (8)(9), others have only recently been brought to attention including:
behavioural psychopathology (30)(65) More importantly, there is now serious evidence that
major adult forms of illness including ischemic heart disease, cancer, chronic lung disease,
skeletal fractures (2), irritable bowel disease, and fibromyalgia (59) have origins in child
maltreatment, presumably acting through behavioural risk factors such as increased smoking
and other risky behaviours. It is quite clear, therefore, that the health burden is dramatically
under estimated. WHO statistics (see Appendix to report) show that 26.9% of the total
amount of injuries for boys aged 0-4 and 33.5% for girls aged 0-4 are caused by intentional
injury and interpersonal violence. For boys and girls aged 5-14 these figures are 22.2% and
22.6%. There are variations between regions and income levels. The percentage of total
amounts of injuries sustained through intentional injury are highest, for example, for children
in the Eastern Mediterranean and European regions in low/middle income groups (Appendix
X).
Physical, behavioural and emotional manifestations vary between children, depending on the
child’s developmental status when the abuse occurs, its severity, the relationship of the
perpetrator to the child, the length of time the maltreatment goes on for and supportive or
buffering factors in the child’s environment (10)(79)(97). The effects of maltreatment can be
serious, life threatening and long lasting. Of those that live, some claim never to recover
(97). However, it is not just the child who will be affected. The immeasurable damage in
family disruption and individual trauma to family members and other affected people
compounds effects and can continue into the next generation. Because health is more than
the absence of disease, the suffering and decreased quality of life resulting from child abuse is
significant.
14
Fig. 3: Other Health Consequences of Child Maltreatment
Physical
Sexual
' : ; ..
■
r
-Jr J
Emotional/Behavioural
Bruises and welts
Bums/scalds
Ocular damage
Lacerations and abrasions
Fractures
Abdominal/thoracic injuries
Poisoning
Asphyxia
Central nervous system injuries
Munchausen Syndrome by Proxy
Unwanted pregnancy
STDs
HIV/AIDS
♦ Morbidity due to adverse reproductive health outcomes
-................ ♦■■ ...............
Eating
disorder
J
♦ PTSD
♦ Depression, anxiety
♦ Drug/Alcoholmisuse
Poor self-esteem
Hyperactivity
Self-inflicted injuries
Poor peer relationships
Feelings of shame/guilt
Somatic disorders
Deterioration in school performance
Eating disorder
Depression, anxiety
Drug/alcohol misuse
1 Developmental effects
/
>
' -'M
Eating disorders
Alcohol/drug misuse
• Depression/anxiety
' ■ Delinquency, violent behaviour
.Selfdestructiveness
..
IT
■
'
Fatal
4
?
,■ ;
..... J
‘
; '7.;.
.:<L.
;
<
Long tern reproductive health outcomes
Sexual dysfunction y.
•
•. ■ ' M
:
'
> Infertility;
- . •. - • •■ ?
-VV.
;
> Ischemic heart disease;
. ■> Cancer
> Skeletal fractures
..J.> ..Uyerdisease
Suicide
HIV/AIDS
Mortality due to adverse reproductive health outcomes
;
(Sources: (2)(10)(30)(33)(34)(65)(97))
15
Child Death
There are wide discrepancies in the numbers of children thought to have died as a
consequence of maltreatment. Several states in the US have demonstrated significant levels
of misclassification in the cause of death as many deaths attributed to other causes have been
shown, on re-investigation, to be homicides (63). Most commonly deaths discovered to have
been homicides were originally attributed to SIDS (sudden infant death syndrome) or to
“accidents.” Unfortunately, in many jurisdictions death scene investigations and autopsies
are not routine. Despite the apparent widespread misclassification, there is general agreement
that fatalities from child maltreatment are more frequent than official estimates from vital
records. Data from the United States indicate that the rate of known maltreatment deaths has
increased from 1.30 to 1.81 per 100,000 between 1985 and 1995. The increase may represent
greater diligence in investigation and classification. The forms of maltreatment leading to
death have remained similar over time; between 1993 and 1995, 37% of dead maltreatment
victims died from neglect, 48% died from abuse and 15% died as a result of both types of
maltreatment (23).
International estimates of the number of children dying from
maltreatment vary according to the source of information. WHO data estimate 88,000 deaths
to children under 14 as a result of homicide and 281,000 deaths to children under 14 as a
result of intentional injury across the world. As noted above in the section on child
characteristics, infants and very young children are at greatest risk, with rates per 100,000 of
the population for the 0-4 age group more than double those of 5-14 year olds. Income and
the global region of origin are related to variation in risk. Among children aged 0-4, the rate
per 100,000 for mortality caused by homicide is 2.5 for boys and 2.6 for girls living in high
income groups, whereas in low/middle income groups it is 8.4 for boys and 9.9 for girls; a
nearly four-fold difference in risk by income. Children under 4 from the Eastern
Mediterranean low/middle income group have the highest homicide mortality rates at 14.8 for
boys and 16.4 for girls. India is next (10.1 for girls and 13.6 for girls). The greatest gender
difference in homicide mortality rates is found in China with rates of 15.7 for girls in
comparison to 7.9 for boys aged 0-4.
Financial Burden
While the obvious financial costs for both the short and long term care of victims of child
maltreatment are high, the hidden costs may be extensive. Some examples are:
•
•
•
•
•
•
•
•
Medical care and complications
Mental health and substance misuse care for victims, perpetrators and families
Inappropriate medical care for unrecognised abuse
Criminal justice system expenditures
Other legal costs
Social welfare organisations costs
Cost to the education system caused by poor school performance
Years of life lost because of death, disability and long term effects
16
This list is not exhaustive, but reflects some of the components that are taken into account to
calculate the financial burden of maltreatment. This list of costs is not all-inclusive; it does
not include potential social and economic multiplier effects to address the impact of
maltreatment on economic productivity and impact on the quality of life. The financial cost
associated with child maltreatment was estimated at $12,410,000,0003 in the US in 1996 (96).
This figure included estimates for future lost earnings, educational costs and adult mental
health services. In the UK an estimated annual cost of $1,176,000,000 is expended for
immediate welfare and legal services alone (74). Additional research studies and data are
urgently needed in this area. Preventive costs are likely to be many times less than the
combination of initial and long-term costs of maltreatment to the individual, family and
society.
RECOGNITION / TRAINING OF HEALTH PROFESSIONALS
It is apparent that many health professionals around the world lack the skills or inclination to
identify and intervene when confronted with a child maltreatment victim. Data from the
United States, a country with a relatively long history of recognition of the problem of child
maltreatment, are illustrative. While a survey of American medical schools observed that
95% of medical school deans reported that their schools included information about child
abuse in the medical school curriculum, the median amount of class-time spent on child
maltreatment was two hours. When the medical students were asked, they reported child
maltreatment coursework in only 80% of American medical schools although they agreed
with the estimate of 2 hours of instruction (1). It seems unlikely that medical students will
develop adequate skills at the recognition and response to child maltreatment with just 2
hours of instruction. Collateral evidence bears out the failure to adequately prepare the health
care workforce to recognise child maltreatment. Recent studies have demonstrated
tremendous lack of agreement between physicians of findings indicative of child sexual abuse
and a high rate of missed diagnosis on abusive head trauma presenting to a major medical
centre (43)(70). Anecdotal reports from the faculty of the 27 lesser developed country
medical schools participating in the International Clinical Epidemiology Network suggest that
the problem may be even greater in other regions of the world. Little or no formal instruction
in child maltreatment has been provided at any of a group of some of the most prestigious
medical schools in Asia, Africa, and South America.
Physicians need to be trained to pursue histories of child maltreatment, interpret histories in
the context of the child’s developmental status, recognise the signs and symptoms of child
maltreatment and distinguish suspicious findings from innocent findings. The medical
evaluation of suspected child abuse needs to have both great sensitivity and specificity,
meaning that the likelihood of detecting abuse must be high and the likelihood of detecting
abuse when no abuse has occurred must be very low (75). Social service organisations rely
extensively on medical expertise and are not in a position to question medical interpretations
of physical findings. High rates of medical error and physician unwillingness to pursue the
diagnosis of child maltreatment complicate societal efforts to protect children and leave
children at risk. Unfortunately, available data suggest that the preparation of physicians for
work in this area is inadequate.
3 US dollars
17
The 1998 publication of the International Resource Book (42) provides estimates of the forms
of child maltreatment that consume the greatest amount of professional energy. Respondents
reported that sexual abuse consumed the greatest amount of professional time followed by
physical abuse. Relatively few respondents thought that neglect, abandonment and
psychological maltreatment consumed the most professional time. Sexual abuse is one area
that many health professionals report great reluctance to pursue (76) and which is hard to
even research (32). Training in the techniques of examination in this area needs to be
specialised and there are clearly differences in reliability of diagnosis based on experience
(71).
Table 2: Forms of Child Maltreatment That Consume the Greatest Percentage of
Time and Professional Energies across 47 countries
32
Percent
.
69.6
13
28.3
7
15.2
4
Number
Type ofAbuse
Physical Abuse . /
Sexual Abuse
Physical Neglect ' ■
Psychological Neglect
Abandonment
Other '
'
(Source: (47))
INTERVENTION
Mandatory reporting of suspected child maltreatment appears to be the law for a minority of
the world’s population (42). A survey completed by professionals in 47 countries noted that
at most 24 countries have mandatory reporting of suspected child maltreatment and another
17 countries have voluntary systems for reporting. These same respondents report that in 39
of the 47 countries foster care is the most prevalent service offered. Case management
services and therapy for the child were also mentioned as frequent services. The same survey
also noted that investigation and rescue were more common than preventive services despite
evidence of the effectiveness of early home visiting and other forms of prevention
(67)(68)(69). Few other data are available about the frequency of different interventions and
the success of these interventions in different countries (56)(95).
Is treatment effective? There are very limited studies of this issue, most in the US and more
for sexual abuse. A review of the effectiveness of treatment for sexual abuse noted that the
older non-experimental studies suggested that some problems, such as aggressiveness and
sexualised behaviour were resistant to change (79). However, the same review noted that
newer randomised trials have concluded that many of the symptoms can be responsive to
professional intervention although other factors such as parental reaction and children’s
attributions can influence outcomes.
18
PREVENTION
The Haddon Matrix is a useful tool for organising prevention planning and policy
development in child maltreatment (41). Haddon proposed that the prevention of motor
vehicle injuries could be organised into A) primary prevention or prevention prior to the
occurrence of the event, B) secondary prevention or prevention that came into place at the
time of the event, and C) tertiary prevention or rehabilitation after the event. Further, Haddon
proposed that prevention could be directed alone or together at the host (child), at the agent
(caregiver), or the environment. The intersection of these two dimensions produces a matrix
with 9 cells. A multi-faceted prevention program requires strategies for every cell of the
matrix. Careful parenting education and home visiting (69) can be seen as primary
prevention directed at the caregivers. Teaching a parent to pick up the phone or walk away
when he or she is angry with a child is secondary prevention and post-maltreatment parenting
classes or the use of foster care constitute tertiary prevention. Teaching a child to avoid
situations that could be risky is primary prevention while education to say “no” is secondary
prevention and instructing the child to tell an adult if someone has touched her or him is
tertiary prevention. Providing social support and income support for new parents is a primary
environmental strategy and making sure that medical providers recognise maltreatment is a
tertiary environmental strategy for prevention. Other prevention strategies that fit into this
matrix include altering social norms for acceptable parenting behaviour, increasing
recognition of the potential harm from shaking a child, and ensuring the availability of
counselling services for victims. Interventions that target “high-risk” groups prior to the
occurrence of maltreatment should be considered primary prevention. Some of the data from
the home visiting programs in the United States suggest that targeted prevention activities
may be the most cost-effective (69).
On a global scale, a major reason that maltreatment continues to flourish is that maltreatment
is not even recognised as a problem. The Haddon Matrix approach suggests that building
public awareness of child maltreatment can be one important starting point for primary
prevention for both the parent and the environmental. The ISPCAN world report on
maltreatment (42) revealed wide variation in public awareness across different countries (see
Table 3). Only a minority of countries and a minority of the world’s people appear to
recognise the problem of maltreatment.
Table 3: The General Public’s Level of Awareness Regarding
Child Abuse and Neglect Issues across 47 countries
Neutral
Aware
3hl%15 ... 33.3%,,,
E^e<pf CliiM Abi^c
14
27.3%
6
13.6%
Causes of Child Abuse 25
56.8%
12
'2
‘
How
Can
Society ■31
68.9%
12
26.7%
... 4.4%
Prevent Child Abuse
•
%
What Can Individuals 28
62.2%
14
31.1%
3
6.7%
Do to Prevent Child
Abuse
(Source: (42))
19
Research examining the effectiveness of different prevention strategies has been limited even
in the two countries indicating that there is awareness of how society can prevent
maltreatment. In the US, attention has focused on home visitation as an approach. These
programs use a variety of providers (nurses, graduate students, and paraprofessionals) and
they emphasise different areas (health care, child development, and social supports). Whilst
not all home visitation has been shown to be effective, most of the studies document positive
alterations in parental attitudes and behaviours as well as a reduction in the likelihood of child
physical abuse and neglect (22)(40)(54)(55)(68)(69)(70). Home visiting is particularly
effective with very vulnerable families including those who are reticent, highly isolated,
lacking in social skills and unlikely to reach out for help themselves.
Centre-based and parent education preventive interventions have documented gains
including: increased positive parent-child interactions, more extensive use of social supports,
less use of corporal punishment, higher self esteem and personal functioning. For teenagers
in particular evaluations have demonstrated fewer subsequent births and higher employment
rates. Centre-based programs are particularly effective for new parents who are motivated and
self confident enough to step outside their own homes and seek support.
Prevention strategies directed simultaneously at multiple intersections of the Haddon Matrix
provide the greatest promise of success. However, the strategies themselves must be serious
and sustained. Research has identified a number of key characteristics of successful
preventive services.
•
The importance of duration. If the goal of a program is crisis intervention, short-term
services may be effective. Short-term programs may be useful in identifying risk,
improving knowledge, reducing immediate stress and temporarily reducing isolation.
However, improving parenting practices, and thus reducing the risk for abuse, generally
requires significant (and long-term) intervention. Longitudinal studies confirm that gains
grow over time.
•
The greatest gains are seen with high-risk families. The likelihood of maltreatment in the
future is clearly higher in families with multiple risk factors. Prevention efforts targeted
at populations of families at high risk are most likely to show significant benefits since
the baseline rate of events in lower risk populations will be near zero.
•
The importance of intensity. Interventions with new parents, especially high risk parents,
need to be frequent (once a week or more often) in order to be effective (22).
•
The qualities and skills of the service providers are important. Only a few home visiting
prevention programs have shown clear benefit. The most successful to date have very
clear but flexible educational curricula and use well-trained professionals such as trained
nurses (68)(69). Other home visiting programs that have failed to show benefit appear to
suffer form very modest interventions by visitors without extensive special preparation.
As discussed earlier in this chapter, the ability of the health system to recognise trauma and
make accurate diagnoses is an important part of the rehabilitative response. Very few children
report themselves for help, even where they have the verbal or physical ability to do so (35).
They may fear they will not be believed, fear the consequences of telling others or think that
the way they are being treated is normal. Pre-verbal infants are particularly at risk. Following
detection, tertiary interventions are best underpinned by a co-ordinated inter-agency, multi20
disciplinary approach. Children may need immediate medical attention and counselling.
Caregivers may need therapeutic help, information and education, skill development,
material, practical and social support. Children, safe carers and the community may need
legal protection from perpetrators, who themselves may need treatment and support.
Fig. 4: Prevention Activities
Tertiary
Primary
______ _ Secondary
•
Environment
Perinatal
and
ongoing
:H Public
awareness
Adequate
child
identification of at risk
activities (i.e. through
protection
laws
and
children
and
families
>■ •. and
media
child friendly courts
Community-based,
campaigns}
family
centred
support,
Community ■
assistance and networks
education,
programmes on CRC
Availability
andd
accessibility of social ...
services,
: —
supports
services,1 >.
;
and networks
Family support such as
Caregiver
C^giver
......
home visiting
"f
=>,■. Pre-natal, pennatai
"■
i
Treatment
and support
established
and early childhood '; Clearly
referral ’system ;, of
for perpetrators
health-care. that
support services
improves; pregnancy i
Substance
:
abuse
A^d
outcomes
treatment programmes
early
strengthens'
Information ' \ about
-attachment
community resources
Promoting ; ■ good
and safety planning
parenting practices
-■
'■
■
Z-A
”,4
Child
; School-based
. ; activities towards non
violence
s
School based social
services: for high stress
environment
■4
• r/7't’-■. .d-'-'i.’i4';1
SOW
OS
(Source: (41)(95)
21
Early diagnosis
Proper inter-disciplinary
services to ensure ?
' Medical treatment, caref
counselling, ? <Vmanagement
; and
support
of
yictims/families
. .Reintegration in a ctiild• friendly ‘
community/schools
The Haddon Matrix approach to prevention suggests multiple levels for targets (41).
Children are likely to need a range of help including therapeutic and relationship work to
compensate for poor attachment formation and/or early sexualisation, educational support and
personal safety training. Caregivers can need help with self-esteem, anger control and
conflict resolution, sexual dysfunction, parenting skills and addiction related behaviour.
Preventive interventions can also be directed at the family level on such issues as social
isolation and exclusion, economic difficulties and marital or relationship problems including
domestic violence. At the community level interventions can be targeted on housing,
employment and social support services. Finally, at the cultural or societal level interventions
can focus on the acceptability of corporal punishment and gender inequalities related to child
rearing and family life (69).
POLICY RESPONSES
Governmental policies addressing the prevention and response to child maltreatment appear
to range from ‘laissez faire’ to strong ‘paternalist’ policy, and from pro-birth family to child
rights perspectives (36). The World Perspectives on Child Abuse resource book (42)
suggests that governmental policies generally lean toward the laissez faire with either no or
voluntary reporting and no officially recorded statistics for the majority of countries providing
a response. Further, only 47 countries even participated. It is very likely that the majority of
non-respondent countries also have either no or very laissez faire policies.
The appropriate set of government policies are the same as the core functions of public
health: governments should undertake a) surveillance, b) assurance, c) monitoring, and d)
policy development (66). Governments must monitor the occurrence of child maltreatment
and the potential for resulting harm. Monitoring may consist of collecting case reports,
periodic surveys, or other methods appropriate to ascertaining the incidence and impacts of
maltreatment in its different forms. In the United States registries of reported cases (19) have
been supplemented with three national incidence studies (56) as well as three academic
surveys of child violence (85)(90). Because of the lack of recognition and training of
professionals, and lack of governmental programs, reliance upon official reports is likely to
be even less successful in other countries. Periodic population-based surveys of the public
are likely to be needed to help raise professional and public awareness.
Assurance means ensuring that response systems are working and providing systems where
private systems are lacking or cannot respond to the need. For example, in the Philippines,
private and public hospitals provide the first line of response to child maltreatment and the
government follows with the criminal justice system (73). If private systems are adequate for
the task, there may be no need for governmental systems. It is absolutely essential that
children receive thoughtful and expert services at every stage in the process. Investigations,
medical evaluations, medical and mental health care, family intervention and legal services
are all activities that children and families will require to be safe. In countries where there is
a tradition of private children’s aid societies providing these services, it may be necessary
only to monitor care. However, governments must assure the quality and availability of
services and provide them if no other provider is available. Assuring the training and
availability of child protective service workers, health care personnel, court personnel, police,
and child care workers clearly falls under government policy. If medical schools continue to
ignore the problem of child maltreatment, government must assure that doctors are
22
sufficiently skilled in this area by directing curricula, altering licensing examinations, or
making other demands on the profession.
Monitoring remains a government function. Even in situations where private child advocacy
organisations are providing services, the core function of continuing to monitor services and
study the epidemiology of maltreatment remains a governmental task. A focus on outcomes
and an allegiance to developing better interventions through the scientific method should be
fostered by the government regardless of whether the service providers are public or private.
Finally, policy development is another primary governmental task. Policies may be required
to assure a trained workforce, a multi-disciplinary response, alternative placements for
children, access to health resources, and resources for families. Governments must take
seriously their responsibilities to assist families care for children and to assist local agencies
with careful and effective protective service interventions. Careful consideration of
mandatory reporting, provision of training of professionals in child protection, and ensuring
the medical and mental health needs of victims cannot be done by any other agency in society.
However, a necessary condition for the development of policy is that child maltreatment must
have risen to the level of public and professional awareness.
Responsibility of the Health Care System
Medical and public health providers have a special responsibility. Researchers in these fields
have the skills to design and conduct investigations of the epidemiology and consequences of
maltreatment which can drive public awareness and governmental responses. Where health
professionals turn a blind eye to maltreatment, it is unlikely that the public will recognise and
demand assistance for maltreated children. Medical schools must include attention to the
problem of child maltreatment within their curricula.
SUMMARY
Although the predominance of western publications about the problem of child maltreatment
could lead to the interpretation that child maltreatment is a western problem, there are
mounting data to suggest that the problem is of equal or greater importance in non-westem
countries. There are great difficulties with definitions of maltreatment in different societies.
There are also variations in patterns of caring for children and in the strengths and resources
of families around the globe. We have evidence that there is little public or health
professional recognition of child maltreatment in most of the countries of the world. Wider
public and health professional recognition would be the start of effective prevention policies.
Health professionals must learn to recognise child abuse and neglect and governments and
academics must ascertain the epidemiology. Effective intervention strategies can be
developed. Effective prevention efforts and policies must be directed at the children, the
caregivers, and the environment before, during, and after occurrences of abuse or neglect.
Health professionals must take on the task learning to recognise and respond to child
maltreatment. Medical schools must include training in this field as a part of medical
education.
23
Box 1
Implementation of the UN Convention on the Rights of the Child (CRC)
in Relation to Child Maltreatment
Marcus Stahlhofer
Article 19 of the CRC recognises the child’s right to protection from all forms of violence. It acknowledges the
range of situations that are potentially abusive, and that violence may be attributable to active mistreatment, or
else to neglect. It notes that violence and abuse occur both within and without the family and the home. This
article affirms the central need to provide protection for all children. In contrast, articles 32, 33, 34 and 38
amongst others identify the special protection needs that occur in very particular situations.
Article 19.2 identifies the many domains in which violence can be responded to, and the responsibility that falls
to the State to strengthen all these actors. This is not simply through the introduction of protection measures, but
through prevention, identification, investigation and follow-up. Article 19.2 highlights the importance not only
of legislative measures in responding to violence and abuse, but also the central place of social programmes and
the provision of support to carers in order to prevent it.
In this way, article 19 provides a framework for thinking about the context of violence and abuse and for
approaches to responding to it, always bearing in mind the general principles of the Convention.
The Committee on the Rights of the Child has drawn on all the aspects of article 19 given above in its discussion
with State Party delegations on matters of violence and abuse. It has highlighted the fact that violence and
abuse often goes unacknowledged and has urged countries to investigate the issue. In its comments to the
government of Romania, the Committee urged ‘that research be undertaken on the issue of child abuse and
neglect within the family.’ In the light of article 19.2, the Committee frequently calls upon States to introduce
legislation to address the issues of violence and abuse. ‘The Committee recommends that ...the State Party
considers the possibility of introducing more effective legislation and follow-up mechanisms to prevent violence
within the family...’ but equally, that legislative reform must be accompanied by other measures if violence and
abuse are to be properly addressed. In its comments to Costa Rica, the Committee noted that the legislation that
was in place had not been adequately enforced. It further urged that...’the government intensify... its information
and advocacy campaign at the community and family level.’
Jordan passed before the Committee on the Rights of the Child in 1993. In its Concluding Observations and
Recommendations, the Committee stated its concern at ‘the rates of domestic violence and abuse, the lack of
discussion of these problems, and facilities to address them.’ It recommended that studies be conducted into on
the nature and extent of domestic violence, and that appropriate follow-up measures should be taken in the field
of family awareness and social support. In the light of these recommendations and others, Jordan had, by 1999,
introduced a national plan of action for children, and a task force to study matters relating to the welfare and
rights of children including the issues of domestic violence and abuse. Appropriate legislative reform was
undertaken, and in addition, training and information was provided to increase public awareness and discussion
of this otherwise taboo subject. Recognising that social and cultural barriers inhibited women and children from
seeking assistance in case of domestic violence, special units of women police officers were trained to provide
appropriate assistance and support.
24
Box 2
Learning from People with Direct Experience of Child Maltreatment
Corrine Wattam
The National Commission of Inquiry into the Prevention of Child Abuse in the UK asked survivors what they
thought child abuse was and what could be done to prevent it. These are some of the responses^:
Definitions
"Child abuse is, mental abuse, i.e., telling a child that he is useless, thick or unable to thinkfor himself. Physical
abuse, i.e., beating him for absolutely no reason. Sexual abuse i.e., touching intimately, intercourse, or other
sexual acts ”
"I realize that when some people use the expression ‘child abuse ’ they mean only sexual abuse. That of course, is a
horrific crime. Some would take the view that physical abuse is ‘nothing’ compared with it. I don’t take that view
becausefrom experience I know that it damages, especially when verbal abuse accompanies the beatings"
"the nature of child abuse and neglect comes in many different forms. Children can be deprived of love, food,
clothes to wear, a warm place to sleep, stimulation with learning skills, the list is endless. A child needs tofeel safe,
ifyou can’t be safe with yourfamily or parents then who can you be safe with?
Causes
"It [abuse] is a way that some people deal with their hurt. It is a way ofcontrolling others ”
"my suggestion comes from my own situation. My parents were both pretty grim: a cruel seemingly unloving
mother and a semi-absent father who at times resorted to leathering, punching and kicking. It took me forty years
to realize the problem was not with me but was a) their marriage; b) their inability to communicate and c) their
inability to resolve difficulties in a mature way ”
“I believe in my own experience, couples who lack sexual communication can be a major cause ofchild abuse. My
mum weighed over 40 stone, my dad used my mums weight as one of his excuses to come onto me. He often said
my mum was barren, which Ifound cruel of him to say about her. Ifelt very sorry for my mum. My dad always
said he wanted to keep it in thefamily, he said he trusted me as I wouldn 't have any diseases. He said he loved my
mum and made me feel guilty by saying it would destroy my mum ifI ever told her about my dad So for years I
kept it a secret and chose to say nothing to spare my mums feelings ”
Prevention
“community andfamilies want to believe that abuse is rare and cannot happen to them. They need educating to
the facts, that even the most respectable household can hide an abuser and a victim. Education, would let a
victim know he/she was not alone and that help was available, also abusers would learn ofways to seek help ”
“I don’t honestly know if child abuse would ever stop, whether it be sexual or otherwise. It has been going on
for generations ”
“If asked today by anyone how to avoid being abused my advice would be shout, yell, scream, kick and tell
someone (I wish I had) ”
“Tender loving care, a listening ear and lots of it, in my view, are the only way to try and help ”
“maybe if it was spoken about more, accepted that this sort of thing does happen in all walks oflife by men and
women, then this subject would not be ‘taboo!. I think ifpeople were more open and stopped burying their
heads in the sand we could start to prevent it”
4 Letters were received from over 1000 respondents. For the full report see (69).
25
Box 3
Infanticide in Hungary
Maria Herczog
When we talk about domestic violence we usually think of battered women and children, sometimes men. And if
we hear about a mother killing her new-born child, we do not want to believe it. In Hungary more than two
dozen cases of infanticide are recorded each year. In 1999 there were 29 and on average, an infant is killed
every two weeks of each year.
Research started in 1998 by the National Institute of Family and Children in co-operation with the Association of
Visiting Nurses aimed to find out about women who commit infanticide and why they do not receive support in
crisis situations that could act as preventative measure. 4500 questionnaires were distributed to visiting nurses.
The research has revealed a picture that does not relate to the monster-mother model. All the women who
committed infanticide kept the whole period of their pregnancy secret. They could not or dared not share the
fact that they were expecting a baby with anyone. The cases were primarily characterised by isolation and the
anticipation of a miracle that would somehow sort out the pregnancy. In general the women did not injure their
children, but left them alone and let them die.
The typical picture is of a very lonely woman, not having one person to trust, let alone to receive help from when
giving birth. Remarkably enough, almost half of them were married or living in a permanent relationship and
two-thirds live in very small communities where everybody knows them. Before receiving the typical sentence
of 2-3 year imprisonment, they stand alone again in front of the judge. Family, neighbours, colleagues, doctors
and district nurses — who in all cases have known about or at least suspected the concealed pregnancy but could
not or did not want to get involved in such ‘private’ issues - are heard as wimesses. These wimesses are never
considered as responsible or implicated. Society seems to point its finger to the deliberately ‘wicked’ woman
who does not fit our ideas about motherhood. Researchers point to social causes such as a mother’s poverty and
isolation, and lack of sex education, but no comprehensive evaluation or action plan has been introduced to
make the necessary changes.
Infanticide is a preventable problem if it is considered as the responsibility of the wider community as well as the
mother herself. Following the research, recommendations have been made about a concerted response to hidden
pregnancy. Anyone who is suspicious of a hidden pregnancy is advised to approach the local child welfare
services ■where the social worker can organise a case conference after talking to the pregnant woman and trying
to convince her to ask for help of any kind. In cases where the prospective mother denies the pregnancy an
obligatory report should be made in accordance with health legislation. This states that after the 24th week of
pregnancy there is a right to intervene despite non-consent of the woman in the best interests of the unbom child.
Although there are many doubts concerning ethical issues related to pressurising anyone to be examined and
helped - even if she does not want to be - the conflict of interest, the risk of death to the child and the potential
tragedy for the entire family make it acceptable for most professionals. In parallel to these recommendations a
media campaign has been launched along with training for helping professionals and a handbook on the subject
is in preparation.
26
BOX 4
ZIMBABWE: INSIGHTS INTO A
COMPREHENSIVE AND EFFECTIVE RESPONSE
Naira Khan
The training and Research Support Centre (TARSC) of the Child and Law Project in Zimbabwe undertook a
participatory and multi-sectoral response to child sexual abuse. Recognition of child sexual abuse has only
evolved recently as a grave and perpetuating problem in Zimbabwe. No systematic national research has been
conducted to determine prevalence rates. Only small, local research has been documented. Building on this
information, TARSC embarked on participatory research among rural and urban groups across the country. A
reference group of individuals and professionals from the affected communities first established the particular
aspect of child abuse to be studied. It was decided that using a new education philosophy, 'education for social
change'. Areas of concern would be identified by the communities, enabling reflection on causes and
participatory strategies for action. Role-play, drama, pictures and conversations were used to draw out the
communities' views, experiences and perceptions around child sexual abuse. The groups consulted during this
process were youth in and out of school, adult men and women, professionals, community leaders and
government, and NGO's active in children's issues. A two stage process collected views about children and their
rights, the extent, nature, forms and causes of sexual abuse of children as well as what can be done to prevent,
detect, report and manage the problem.
During this research process, consultative meetings were held with the reference group to monitor progress and a
leaflet detailing stage one results was prepared in English and Shona, the local language. These groups later
developed and implemented action programmes. Professionals, having been involved from the development
process, instead of being brought in at the end to deliver a service. Following evaluation findings, many
activities were implemented to prevent child abuse and neglect. Two examples are the school information
programme and the legal programme. TARSC implemented the school information programme with the Ministry
of Education and Culture, focusing on training, capacity building and development of training materials for
school psychologists, heads and teachers, administrative staff and children. It is envisaged that by the end of
2000 the majority of personnel in this Ministry will be conversant with issues of Child Sexual Abuse and the
program will begin the second phase of their initiative to prepare children to recognise potentially abusive
situations, to be aware of reporting channels.
Together with the Ministry of Justice, Legal and Parliamentary Affairs, TARSC initiated a legal programme.
Plans included the establishment of multisectoral training on victim-friendly courts for vulnerable witnesses for
nurses, police and NGOs, and courses for trainee lawyers and reporting protocols. For abusers, a training of
probation officers, police and public prosecutors on management of young sexual offenders, training and a
review of sentencing patterns.
Innovative initiatives have also been launched on the subject of public awareness, health professional awareness
and crisis intervention. For example, the Mabvuku Cultural Drama Group uses drama to initiate community
dialogue and problem solving on child sexual abuse. During the last two years the group has been conducting
performances at places where men are in the majority such as beerhalls and burial societies.
27
BOX 5
TRAINING FOR HEALTH PROFESSIONALS IN GERMANY
Reiner Frank and Maru Kopecky-Wenzel
There may be debate about whether a particular case can be labelled as abusive or not. However, the
developmental status, behavioural problems of children, relationships within a family and the needs and
strengths of families can be recognised with some certainty. In Germany research projects in hospitals as well as
in private practices have demonstrated that child abuse is a widespread phenomenon encountered within the
health system. Local and regional initiatives have unified professionals from practice and from research in
different fields. In the absence of a mandatory reporting system training must address knowledge and skills
within a given profession and knowledge and skills for co-operation between professions.
Teaching within a profession:
Patterns of identification vary according to the groups of professionals, e.g. paediatricians, paediatric surgeons,
child psychiatrists, nurses. They also depend on the professional context, such as hospital or private practice.
Knowledge: For all medical professions it is essential to be familiar with signs and symptoms suggestive of
physical abuse, sexual abuse and neglect and with concurrent explanations. Behaviour of children, family
situations and pattern of relationships have to be assessed. For the best use of resources it is useful to tap the
experience of children of health professionals within their daily practice. To enhance competence in recognising
and naming problems in behaviour and relationships short instruments were developed or adapted for use in the
clinical routine. In a controlled study researchers found that nurses in paediatric and paediatric surgical hospitals
were able to distinguish between normal and behaviourally disturbed children and to identify parental rejection,
a strong indicator of child abuse and neglect. On the basis on this research practice parameters were developed
and published by the board of paediatricians and by the board of child psychiatrists. Guidelines stemming from a
version from Hamburg paediatricians were adapted regionally in different counties.
Skills: In the county of Bavaria a survey among practitioners in paediatrics and child and adolescent psychiatry
was done to evaluate the Bavarian version. From the view of the practitioners the best way to improve practice is
to offer an opportunity to discuss their own cases. In training sessions in small circles it was possible to train on
issues such as how to describe a given child and its development, how to assess hyperactivity, how to deal with
parents perceived as demanding, hostile and hopeless. The readiness of physicians in hospitals to identify signs
and symptoms as a result of family violence depends strongly on the existence of a support group within that
hospital.
Teaching between professions
Knowledge: The correlation between the amount of family adversities and behaviour problems of children is
well established. The co-operation of mental health professionals such as psychologists, social workers, child
psychiatrists should be sought. In Germany, the largest gap exists between the medical system and the system of
youth welfare. Different professional belief systems impede a successful co-operation. Knowledge on the
expertise of other professions is a prerequisite for co-operation. Information must be given on child abuse from a
medical point of view, from the view of the youth welfare system and from a legal perspective.
Skills: Personal contacts and long term dialogue can build up mutual understanding and confidence. Teaching
elements are common sessions conjointly guided by a physician and a social worker. Local or regional groups
meeting regularly over time are useful to create a good working atmosphere.
To implement the existing body of knowledge into practice, an ongoing effort is necessary. The ingredients
necessary for effective functioning are knowledge, skills in communication and a personal network. A local or
regional feed back system would be very useful to show that long term targeted interventions are successful.
28
BOX 6
EARLY INTERVENTION TO PREVENT CHILD ABUSE IN THE UNITED STATES
Anne Cohn Donnelly and Deborah Daro
For the better part of the last decade, professionals and others in the United States have sought ways to offer new
parents help in getting off to a good start as a way of preventing child abuse and strengthening families.
More than half of the nation’s 50 states now have state-wide parent support initiatives underway. In addition, a
number of carefully crafted family support programs (such as HIPPY, MELD, Parents as Teachers and Olds’
Prenatal and Infancy Home Visitation Program) aimed at new parents have been promoted by NGO’s across the
country. Also, several of the nation’s largest foundations, have launched national initiatives to explicitly
promote a more comprehensive and coordinated system of support for young children and their parents. The
Federal government has new funding streams in the areas of child health, early childhood education and child
welfare services which are being used for new parent and family support assistance (such as the Early Head Start
Initiative).
Of all these relatively new initiatives the one that perhaps best captures what we have learned from research is
Healthy Families America. Launched by an NGO in collaboration with a corporate foundation and building upon
the experiences of one state (Hawaii’s Healthy Start), the goal of HFA is to offer all new parents nation-wide
support around the time their first baby is bom.
The service typically begins in the hospital and continues with home visits. The home visitor may be a nurse, a
social worker or a paraprofessional from the community. The purpose of services offered to the families is to
address the full gambit of needs new families face (e.g. facilitating bonding, parenting skills and child
development, health care for both mother and baby, and assisting with housing and employment and other social
needs).
The structure of the service is guided by critical elements derived from research. These include: early
intervention (begin as close to the time of birth as possible, if not prenatally), intensity (at least once a week at
first and perhaps up to 6-9 months), longevity (2 years or more for the most vulnerable families), comprehensive
(services should address a variety of social, emotional and concrete needs including ensuring that the family has
a medical home), flexible (tailoring services to each individual family and their particular needs), and
appropriate (both the method of service delivery and the content is sensitive to the particular cultural and ethnic
values of the family being served). Also important is the degree to which the services offered are coordinated
with others the family may be receiving.
The effort, now in more than two thirds of the states and over 300 communities, is attempting to offer all parents
some support (with long term intensive support for high risk parents) while establishing an integrated family
support system in the community. Enthusiasm for this intervention has been keen from all quarters. Securing
funding and setting up home visitation programs has been relatively easy. Finding well qualified workers has
been more difficult necessitating various efforts to offer on the job training. A fundamental challenge has been to
weave existing efforts into new developments and competition between programs with similar goals is
significant.
The Healthy Families America effort has been the focus of more than 30 separate evaluative research efforts,
most of which are still in process. The lead researchers from these various efforts have been meeting in a
Research Network for a number of years to share information about methodology, measures used and findings.
Early results, while uneven, are encouraging.
29
Box 7
Corporal Punishment: State-authorised Violence to Children
The most common direct experience of violence by children globally is being hit, slapped,
punched, kicked and beaten by their parents and other carers. Corporal punishment of
children is socially and legally accepted in most states. In many, it is not only parents who can
hit their children. In schools, other institutions and in penal systems for young offenders,
corporal punishment remains common: state-authorised violence, often severe, on a massive
scale.
In many states, corporal punishment of children is now the only form of inter-personal
violence which remains legal. It is at the least ironic that adults have designed laws which
give children - the smallest and most vulnerable of people - less, not more, protection from
assault.
It is through the assertion that children, too, are holders of human rights that the legality of
corporal punishment is now being globally challenged. Hitting children is a breach of the
universal rights that all people share: to respect for physical integrity and human dignity. The
existence of special legal defences justifying corporal punishment - “reasonable
chastisement”, “lawful correction” - breaches children’s right to equal protection under the
law.
The UN Convention on the Rights of the Child requires states to protect children from “all
forms of physical or mental violence” while in the care of parents and others. The Treaty
Body for the Convention, the UN Committee on the Rights of the Child, has highlighted that
corporal punishment is incompatible with the Convention. In examining reports from states in
all continents it has recommended prohibition of all corporal punishment, however light.
In 1979 Sweden became the first country to prohibit all corporal punishment of children.
Since then, at least 10 more states have banned it (Germany being the most recent, in July
2000). There have also been key judgments from constitutional or supreme courts
condemning corporal punishment in schools and penal systems (for example, in Namibia,
Zimbabwe, South Africa and Zambia). Ethiopia’s 1994 Constitution asserts the right of
children to be free of corporal punishment in schools and care institutions. In 1998 the
European Court of Human Rights found the beating of a young English boy by his stepfather
amounted to inhuman or degrading punishment and that domestic law allowing “reasonable
chastisement” failed to protect the child. It ordered the UK government to pay the boy
£10,000 and his legal costs.
More recently, in 2000 the European Court threw out an application from a group of UK
Christian private schools claiming that the ban on school corporal punishment breached their
rights and parents’ rights to religious freedom, etc. (echoing a similar decision by South
Africa’s Constitutional Court). Protection of vulnerable individuals from deliberate violence
cannot be diluted by considerations of religion or culture.
30
In January 2000 Israel’s Supreme Court declared all corporal punishment to be unlawful:
If we allow Tight’ violence, it might deteriorate into very serious violence. We must not
endanger the physical and mental well-being of a minor with any type of corporal
punishment. A truth which is worthy must be clear and unequivocal and the message is that
corporal punishment is not allowed”.
The consistent recommendations of the UN Committee on the Rights of the Child are
undoubtedly leading to accelerating progress to end corporal punishment. In Bangladesh, the
Ministry of Women and Children Affairs designated the theme for “Child Rights Week 2000”
as “Banning corporal punishment”. In Sri Lanka, the National Child Protection Authority has
prepared a booklet for parents and teachers as a prelude to a legal ban. Recently, school
beating has been banned in Korea, New Zealand, Uganda and Thailand.
Nevertheless, international surveys suggest that corporal punishment - flogging, whipping or
caning - for juvenile offenders remains legal in at least 60 states; that it is legal in schools and
other institutions in at least 65 states, and in the family home in all but 11. Where corporal
punishment has not been consistently challenged by legal reform and public education,
scattered prevalence studies suggest that it remains extremely common. For example, a
survey of Egyptian children published in 1998 found more than a third of the children
reported being beaten with hands, sticks, belts and shoes; a quarter of these children reported
that harsh discipline led to physical injuries. In the UK, research in the 1990s found three
quarters of a large sample of mothers admitted to smacking their baby before the age of one; a
quarter of the young children in the survey had been hit with an implement and a third
punished “severely”.
The imperative for ending corporal punishment of children is one of universal human rights.
But there are strong supporting arguments from public health and violence prevention
perspectives. Corporal punishment is dangerous to children in the short-term: in every state
harsh discipline kills some children, injures and handicaps many more. In the longer term, it
is identified by a very large body of research as a significant factor in the development of
violent attitudes and actions in childhood and later life.
Action to end coporal punishment is long overdue. States need to adopt explicit legislation
prohibiting all corporal punishment and link implementation to comprehensive public
education campaigns involving adults, adolescents and children.
Peter Newell, Co-ordinator EPOCH-WORLD WIDE
Akila Belembaogo, Senior Adviser on Child Rights, UNICEF’s Regional Office for West and
Central Africa, former Member and Second Chairperson, Committee on the Rights of the
Child
31
References
1. Alpert, E.J., Tonkin, A.E., Seeherman, A.M. Holtz, H.A., 1998, Family Violence
Curricula in U.S. Medical Schools. American Journal of Preventive Medicine 14(4):
273-278
2. Anda, R., Croft, J., Felitti, V., Nordenberg, D., Giles, W., Williamson, D., Giovino, G.
1999, Adverse Childhood Experiences and Smoking during Adolescence and Adulthood.
Journal of the American Medical Association 282:1652-1658
3. Baartman, H.E.M., 1997, Home-based services: to each his own? In Hellinckx, W.,
Colton, M., and Williams, M., (eds.) International Perspectives on Family Support.
Aidershot: Arena.
4. Barthauer, L.M., Leventhal, J.M. 1996, Child Sexual Abuse in a Poor, Rural Salvadorian
community. Eleventh International Congress on Child Abuse and Neglect. Abstracts, p
74
5. Belsky, J., 1980, Child Maltreatment: An Ecological Integration. American
Psychologist, 35, pp320-335.
6. Bendixen, M., Muss, K.M., and Schei, B., 1994, The impact of child sexual abuse - A
study of a random sample of Norwegian students. Child Abuse & Neglect, 18, pp837847.
7. Besharov, D., 1993, Overreporting and Underreporting are Twin Problems. In Gelles,
R.J. and Loseke, D.R. Current Controversies on Family Violence, Newbury Park: Sage.
p257-272
8. Bifulco, A. and Moran, A. (1998). Wednesday’s Child: Research into Women’s
Experience of Neglect and Abuse in Childhood, and Adult Depression. London:
Routledge.
9. Briere, J.N. and Elliott, D.M. (1994). Immediate and long-term impacts of child sexual
abuse. The Future of Children, 4, 54-69.
10. Briere, J.N., 1992, Child Abuse Trauma: Theory and Treatment of Lasting Effects.
London: Sage
11. Browne, K., 1995, Predicting Child Maltreatment in P.Reder and C.Lucey (eds.)
Assessment of Parenting: Psychiatric and psychological contributions. London:
Routledge pp 118-13 5
12.CAPCAE, 1996, An Overview of Child Maltreatment Prevention Strategies in
Europe. Report for the Concerted Action for the Prevention of Child Abuse in Europe of
the BIOMED 2 Programme Grant (BMH4-CT96-0829). Available from CAPCAE CoOrdinator, Faculty of Health, University of Central Lancashire, Preston UK.
32
13.Cawson, P., Wattam, C., Booker, S., Kelly, G., 2000, The Prevalence of Child
Maltreatment in the UK. London: NSPCC.
14. Chalk, R. and King, P.A. (eds.), 1998, Violence in Families: Assessing Prevention and
Treatment Programs. Committee on the Assessment of Family Violence Interventions
Board on Children, Youth and Families, Commission on Behavioural and Social Sciences
and Education National Research Council and Institute of Medicine. Washington, D.C.:
National Academy Press.
15.Choquet, M., Darves-Bomoz, J.M., Ledoux, S., Manfredi, R., Hassler, C., 1997, Self
reported health and behavioural problems among adolescent victims of rape in France:
results of a cross-sectional survey. Child Abuse & Neglect, 21, 9, 823-32.
16. Cicchinelli, L.F., 1991, Proceedings of the Symposium on Risk Assessment in Child
Protective Services, National Center on Child Abuse and Neglect, December, 1991,
Washington DC.
17. Cohn, A.H., 1987, How Do We Deal With Research Findings? Journal of
Interpersonal Violence, June, 1987. pp228-232.
18. Cook, R., 1989, Reducing maternal mortality: A priority for human rights law. In Sheila
McLean (Ed.) Legal Issues in Human Rights Reproduction. Brookfield: Gower
ppi 85-212.
19. Children’s Bureau (1999) The National Child Abuse and Neglect Data System 1998.
Washington, DC:
20. Creighton, S. and Russell, N., 1995, Voices from Childhood: A survey of childhood
experiences and attitudes to childrearing among adults in the United Kingdom.
London: NSPCC Research Policy and Practice Series.
21. Daro, D and Cohn, A., 1987, Is Treatment Too Late? What Ten Years of Evaluative
Research Tell Us. Child Abuse & Neglect, Vol. 11, pp 433-441
22. Daro, D., McCurdy, K. and Harding, K., 1998, The Role of Home Visiting in
Preventing Child Abuse: An Evaluation of the Hawaii Healthy Start Programme.
National Committee to Prevent Child Abuse, Chicago.
23. Daro, D., 1996, Current Trends in Child Abuse Reporting and Fatalities: NCPCA’s 1995
Annual State Survey. APSAC Advisor, Vol. 9. No.2, pp21-24.
24. De Paul, J., Milner, J.S., Mugica, P., 1995, Childhood maltreatment, childhood social
support and child abuse potential in a Basque sample. Child Abuse & Neglect, 19, 8,
907-20.
25. De Zulueta, F. 1998, From Pain to Violence: The Traumatic Roots of Destructiveness.
London: Whurr Publishers
33
26. Durant, J., 1998, The Status of Swedish Children and Youth Since the Passage of the
1979 Corporal Punishment Ban. Department of Family Studies, University of
Manitoba, Canada.
27. Egeland, B., 1993, A History of Abuse Is a Major Risk Factor for Abusing the Next
Generation. In Gelles, R.J. and Loseke, D.R. Current Controversies on Family
Violence, Newbury Park: Sage.pp 197-208.
28.Enzmann, D., Pfeiffer, C and Wetzels, P., 1998, Youth Violence in Germany: A Study
of Victimization and Delinquency in Four Major Cities. Criminological Research
Institute of Lower Saxony, Hannover, Germany.
29.Facchin, P., Barbieri, E., Boin, F., Canton, L., Ferrante, A., Manea, S., Ranzato, C.,
Vianello, A., Zorzi, M., 1998, European Strategies on Child Protection: Preliminary
Report, First Workshop Meeting, Padua, 29-31 October, 1998. Epidemiology and
Community Medicine Unit, University of Padua, Italy.
30. Felitti, V., Anda, R., Nordenberg, D., Williamson, D., Spitz, A., Edwards, E., Koss, M.,
Marks, J., 1998, Relationship of Childhood Abuse and Household Dysfunction to Many
of the Leading Causes of Death in Adults. American Journal of Preventive Medicine
14(4): 245-258.
31. Fergusson, D.M. and Mullen, P.E., 1999, Childhood Sexual Abuse: An Evidence Based
Perspective. Thousand Oaks: Sage.
32. Fergusson, D.M., Lynsky. M.T., Horwood, L.J., 1996, Childhood Sexual Abuse and
Psychiatric Disorder in Young Adulthood: I. Prevalence of Sexual Abuse and Factors
Associated with Sexual Abuse. Journal of the American Academy of Child and
Adolescent Psychiatry 34:1355-1364
33. Fergusson, D.M., Horwood, Lynsky. M.T., L.J., 1996, Childhood Sexual Abuse and
Psychiatric Disorder in Young Adulthood: II. Psychiatric Outcomes of Childhood Sexual
Abuse. Journal of the American Academy of Child and Adolescent Psychiatry
34:1365-1374
34.Finkelhor, D., 1986. A Sourcebook on Child Sexual Abuse. London: Sage.
35.Finkelhor, D., 1990, Is Child Abuse Over Reported? Public Welfare 48: 23-29
36. Fox Harding, L., 1991, Perspectives in Child Care Policy. Harlow:Longman.
37. Garbarino, J., 1977, The Human Ecology of Child Maltreatment: a Conceptual Model for
Research. Journal of Marriage and the Family 39: 721-35.
SS.Gelles, R.J., 1993, Current Controversies on Family Violence. Newbury Park: Sage
39. Goldman, J.D., Padayachi, U.K., 1997. The Prevalence and Nature of Child Sexual Abuse
in Queensland, Australia. Child Abuse & Neglect, 21,5, 489-98
34
40. Gray,J., Cutler, C., Dean, J., and Kempe, C.H., 1979, Prediction and Prevention of Child
Abuse. Seminars in Perinatology 3, No. 1, pp 85-90
41. Haddon, W., 1980. Options for the Prevention of Motor Vehicle Crash Injury. Israeli
Journal of Medical Science 16: 45-65
42. Hiatt, S., Miyoshi, T.J., Fryer, G.E., Miyoshi, P., Krugman, R.., 1998, World
Perspectives on Child Abuse: The Third International Resource Book. Elsevier
Science Ltd
43. Hunter, W.M., Jain, D., Sadowski, L., Sanhueza, A., 2000. Risk Factors for Severe Child
Discipline Practices in Rural India. Journal of Pediatric Psychology 25(6): 435-447
44. Jenny, C., Hymel, K.P., Ritzen, A., Reinert, S.E., Hay, T.C., 1999. Analysis of Missed
Cases of Abusive Head Trauma. Journal of the American Medical Association
281(7):621-626
45. Kaplan, S., Labruna, V., Pelcovitz, D., Salzinger S., Mandel, F., Weiner, M., 1999,
Physically abused adolescents: behaviour problems, functional impairment, and
comparison of informants reports. Pediatrics 104:43-49
46. Kaufman, J. and Zigler, E., 1993, The Intergenerational Transmission of Abuse is
Overstated. In Gelles, R.J. and Loseke, D.R. Current Controversies on Family
Violence, Newbury Park: Sage.pp 209-221
47. Keys, A.. 1980. Seven Countries, A Multivariate analysis of Death and Coronary
Heart Disease. Cambridge, MA: Harvard University Press
48.Korbin, J., 1991, Cross Cultural Perspectives and research directions for the 21st Century.
Child Abuse & Neglect, 15, 67-77.
49.Korbin, J., 1993, Culture, Cultural Diversity and Child Maltreatment, APSAC Advisor,
Vol.6,No.3,pp 23-25.
50. Kotch, J.B.; Chalmers, D.B.; Fanslow, J.L.; Marshall, S.; Langley, J.D. 1993. Morbidity
and Death Due to Child Abuse in New Zealand. Child Abuse & Neglect 17(2): 233-47
51 .Krugman, S., Mata, L., and Krugman, R., 1992, Sexual abuse and corporal punishment
during childhood: A pilot retrospective survey of university students in Costa Rica.
Pediatrics, 90, ppi57-161.
52. La Fontaine, J., 1990, Child Sexual Abuse. Cambridge: Polity Press
53. Lamer, M., Halpren, B., Harkavy, O., 1992, Fair Start for children: Lessons Learned
from Seven Demonstrations. New Haven: Yale University Press.
35
54. Larson, C., 1980, Efficacy of Prenatal and Postpartum Home Visits on Child Health and
Development. Pediatrics, 66, pp 191-197
55. Leventhal, J.M., 1996, Twenty Years Later: We do Know How to Prevent Child Abuse
and Neglect, Invited Commentary, Child Abuse and Neglect, Vol. 20, No. 8, pp 647653.
56. Leventhal, J.M., 1998, Epidemiology of sexual abuse of children: old problems, new
directions. Child Abuse & Neglect 22(6): 481-491
57. Levinson, D., 1989, Family violence in a cross-cultural perspective. Frontiers of
Anthropology. Vol. 1. Newbury Park, Ca. Sage
58. Levesque, R.J.R., 1999, Sexual Abuse of Children: A Human Rights Perspective.
Bloomington: Indiana University Press.
59. MacLeod, 1996, Talking with Children about Child Abuse: ChildLine’s first ten
years. London: ChildLine.
60. McBeth, J., Macfarlane, G.J., Benjamin, S., Morris, S and Silman, A.J., 1999, The
Association Between Tender Points, Psychological Distress, and Adverse Childhood
Experiences, Arthritis and Rheumatism, Vol. 42, No. 7 ppi397-1404
61 .McCauley-J, Kern, D.E., Kolodner, K., Dill, L., Schroeder, A.F., DeChant, H,K, Ryden,
J., Derogatis, L.R., Bass, E.B., 1997, Clinical characteristics of women with a history of
childhood abuse: unhealed wounds. Journal of the American Medical Association,
277, 17, 131-5
62. MacIntyre, D. and Carr, A., 1999, The Epidemiology of Child Sexual Abuse. Journal of
Child Centred Practice, pp57-86
63. Meadow, R., 1999, Unnatural Sudden Infant Death, Archive of Diseases in Childhood,
Vol. 80, pp7-14
64. Miller, A., 1985, Though Shalt Not Be Aware: Society’s Betrayal of the Child.
London: Pluto Press
65. National Research Council, 1993, Understanding Child Maltreatment. Washington
DC: National Academy of Sciences Press
66. National Research Council 1988, The Future of Public Health. Washington, DC:
National Academy of Sciences Press
67.NCIPCA, 1996, Childhood Matters, The Report of the National Commission of
Inquiry into the Prevention of Child Abuse, Vol. 1. London: HMSO.
68. Olds, D., 1997, The Prenatal/Early Infancy Project: Fifteen Years Later. In Albee, and
Gullota, T.P. (eds.) Primary Prevention Works. Thousand Oaks: Sage. Pp 41-67
36
69. Olds, D., Eckenrode, J., Henderson, C.R., Kitzman, H., Powers, J., Cole, R., Sidora, K,
Morris, P., and Pettit, L.M., 1997, Long-term effects of home visitation on maternal life
course and child abuse and neglect: Fifteen-year follow up of a randomized trial. Journal
of the American Medical Association, 278, 8, 637-643.
70. Olds, D., Henderson, C., Chamberlin, R., and Tatelbaum, R., 1986, Preventing child
abuse and neglect: A randomized trial of nurse home visitation. Pediatrics, 78, 65-78.
71. Paradise J, (agreement among examiners) Pediatrics
72. Pederson, W. and Skrondal, A., 1996, Alcohol and sexual victimization: A longitudinal
study of Norwegian girls. Addiction, 91 (4): 565-581
73. Ramiro L, Madrid B, Amarillo M. (2000) Final Report: The Philippines WorldsAFE
Study. Final report submitted to the International Clinical Epidemiology Network,
(unpublished manuscript).
74. Roberts, H., Smith, S.J. and Bryce, C., 1995, Children at Risk? Safety as a Social
Value. Buckingham: Open University Press
75. Runyan, D.K., 1998, Prevalence, Risk, Sensitivity and Specificity: a Commentary on the
Epidemiology of Child Sexual Abuse and the Development of a Research Agenda. Child
Abuse & Neglect 22(6): 493-498
76. Runyan, D.K., Hunter, W.M., Socolar, R.R.S., et. al., 1998, Children who prosper in the
face of adversity: the relationship to social capital. Pediatrics
77. Runyan, D.K., 2000 the Ethical, Legal and Methodological Implications of Directly
Asking Children About Abuse. Journal of Interpersonal Violence 15(7): 675-681
78. Russell, D.E.H., 1986, The Secret Trauma: Incest in the Lives of Girls and Women.
New York: Basic Books
79. Saywitz, .J., Mannarino, A.P., Berliner L, Cohen, J.A. 2000, Treatment for Sexually
Abused Children and Adolescents. American Psychologist 55: 1040-1049
80. Shalhoub-Kevrkian N. (1999) The politics of disclosing female sexual abuse: a case
study of Palestinian society. Child Abuse & Neglect 23(12): 1275-1293
81. Socolar RRS, Runyan DK, Amaya-Jackson L. (1995) Methodological and Ethical Issues
related to studying child maltreatment. Journal of Family Issues 16, pp 565-586.
82. Starling, Suzanne Abusive Head Trauma paper
83. Starling, S., 2000, Perpetrators of Abusive fractures. Paper presented to the Helfer
Society Annual Meeting, September 21, 2000, Midway, Utah.
84. Straus, M.A., 1995, Manual for the Conflict Tactics Scales. Durham, NH: Family
Research Laboratory, University of New Hampshire.
37
85. Straus, M.A. and Gelles, R.J. (1990). Physical Violence in American Families: Risk
Factors and Adaptations to Violence in 8,145 Families. New Brunswick, New Jersey:
Transaction Publishers
86. Straus, M.A. 1979, Measuring intrafamily conflict and violence: The Conflict Tactics
(CT) Scales. Journal of Marriage and the Family, 41, pp 75-88.
87. Straus, M.A. and Hamby, S.L., 1997, Measuring Physical and Psychological
Maltreatment of Children With the Conflict Tactics Scale. In K. Kantor et.al. (eds.) Out
of the Darkness: Contemporary Perspectives on Family Violence. Thousand Oaks:
Sage. Ppi 19-135
88. Straus, M.A., 1990, Measuring Intrafamily conflict and violence: The Conflict Tactics
CT) Scales. In Straus, M.A., & Gelles, R.J., 1990, Physical Violence in American
Families: Risk Factors and Adaptations to Violence in 8,145 Families. New
Brunswick, NJ: Transaction Publishers. pp29-47
89. Straus, M.A., Hamby, S.L., Boney-McCoy, S and Sugarman, D.B., 1996, The Revised
Conflict Tactic Scales (CTS2). Journal of Family Issues, Vol. 17 No.3, pp 283-316
90. Straus, M.A., Hamby, S.L., Finkelhor, D., Moore, D.W. and Runyan, D., 1998,
Identification of child maltreatment with the Parent-Child Conflict Tactics Scales:
Development and psychometric data for a national sample of American parents, Child
Abuse & Neglect, Vol. 22, No. 4, pp249-270.
91. Straus, M.A., personal communication
92. Tang, C.S., 1998 the rate of child abuse in Chinese families: a community survey in
Hong Kong. Child Abuse & Neglect 22: 381-391
93. Ten Bensel RW, Rhineberger MM, Radbill SX, Children in a world of violence: the roots
of child maltreatment. In Helfer ME, Kempe R, Krugman R.(eds). The Battered Child.
(5th Edition) Chicago: U of Chicago Press 1997
94.Trowell, J., Ugarte, B., Kolvin I., Berelowitz, M., Sadowske, H. and Le Couteur, A.,
1999, Behavioural psychopathology of child sexual abuse in school girls referred to a
tertiary centre - a North London study, European Journal of Child and Adolescent
Psychiatry, pp
95. Wattam, C. and Woodward, C., 1996, ‘...And did I abuse my children. NO! - Learning
about prevention from people who have experienced child abuse. In Childhood Matters,
The Report of the National Commission of Inquiry into the Prevention of Child
Abuse, Vol. 2, Background Papers. London: HMSO
96. WHO, 1999, Report of the Consultation on Child Abuse Prevention. Geneva:
PVI/WHO
38
97. Wolfe, D. A., 1999, Child Abuse (2nd Edition): Implications for Child Development
and Psychopathology. Thousand Oaks: Sage
39
Chapter 4
Violence Against Women by Intimate Partners1
Status:
Draft: 2nd
Peer Reviewed: Yes
Date of Current Draft: 4 July 2000
1 Portions of this manuscript were prepared under contract to the Center for Communications Programs at Johns
Hopkins University. A modified version of the text appears in: Heise, Lori, Ellsberg, Mary, & Gottemoeller,
Megan. Ending Violence Against Women , Population Reports, Series L, No. 11. Baltimore, Johns Hopkins
University School of Public Health, Population Information Program, December, 1999.
1
Table of Contents
I.
n.
in.
IV.
v.
VI.
VII.
VIII.
IX.
X.
Introduction
Nature and Magnitude of the Problem
The Dynamics of Abuse
Women’s Response to Abuse
Explaining Intimate Partner Violence
The Consequences of Abuse for Health and Well-Being
A.
Impact on Health
B.
Violence and Reproductive Health
C.
Use of Health Services
D.
Economic Costs of Violence
E.
Impact on Children
Prevention and Policy Response
A.
Support and Refuge for Victims
B.
Legal Remedies and Judicial Reforms
C.
Transforming Police Practice
D.
Health Service Interventions
E.
Outreach and Advocacy
F.
Alternative Sanctions
G.
Coordinated Community Interventions
H.
Perpetrator Treatment
I.
Prevention Campaigns
J.
Anti-violence Education Programs
Principles of Good Practice
Conclusions and Recommendations
Boxes
1. Enhancing Comparability of Data on Domestic Violence
2. "La Ruta Critica:" An Evaluation of Institutional Responses to Domestic Violence
3. Examples of Innovative Prevention Programs
4. How Women Experience Violence: Women's Voices from North and South
2
Introduction
Worldwide, one of the most common forms of violence against women is abuse by a husband or
intimate male partner. A signal feature of the global epidemiology of violence is that women are
most at risk from family members and male intimates. This is in stark contrast to the risk pattern
for men, who on average are much more likely to be attacked by a stranger or an acquaintance
than by someone within their primary circle of relationships (1-5). The fact that women are
often emotionally involved with and economically dependent on those who victimize them has
major implications /for both the dynamics of abuse and the approaches available for addressing it.
Relationship violence is a global phenomenon that disproportionately affects women. Domestic
violence occurs in all countries and transcends social, economic, religious and cultural groups.
Although women can be violent in relationships and violence exists in some same-sex
partnerships, the vast majority of partner violence is perpetrated by men against their female
partners (6). As a result, this chapter will restrict itself to discussing violence by men against
female intimates.
Thanks largely to the efforts of women’s movements worldwide, violence against women, and
domestic violence in particular, is now firmly placed on the international agenda. Initially
conceptualized primarily as a women’s human rights issue, partner violence is increasingly also
seen as an important public health problem. This chapter briefly examines what is known about
the nature and magnitude of partner violence and describes its consequences for women’s health
and well-being. In addition, it examines the range of interventions being tried globally and
reviews the limited data available on the effectiveness of these efforts. Despite a scarcity of
formal evaluations, the field of domestic violence has over two decades of experience to inform
practice. Drawing on this experiential base, the chapter highlights lessons learned about how to
conduct and organize domestic violence interventions and offers insights on areas in need of
strengthening. The chapter concludes with a list of recommendations for future work on violence
against women in intimate partnerships.
Nature and Magnitude of the Problem
Intimate partner violence refers to any number of behaviors that serve to undermine the physical,
psychological, and/or sexual integrity of women within intimate relationships. It includes acts of
physical aggression, such as slaps, hits, kicks or beatings; psychological abuse such as constant
belittling, intimidation, and humiliating treatment; forced intercourse and other forms of sexual
coercion; and a variety of controlling behaviors, such as isolating a woman from family and
friends, monitoring her movements, and restricting her access to resources. When different types
of abuse occur repeatedly in the same relationship, the phenomenon is often referred to
“battering.”
In nearly 50 population-based surveys from around the world, 10 percent to over 50 percent of
women report being hit or physically harmed by an intimate male partner at some point in their
lives (See Table 1). The percent of women who have been assaulted by a partner in the last 12
months varies from less than 3 percent of adult women in the United States and Canada, to 27
3
percent of ever-partnered women in Leon, Nicaragua; 38 percent of currently married women in
Korea, and 52 percent of currently married Palestinian women in the West Bank and Gaza strip.
For many of these women, physical aggression is not an isolated event but an ongoing
phenomenon, often accompanied by debilitating sexual and psychological abuse.
Research suggests that physical violence in intimate relationships is almost always accompanied
by psychological abuse, and in one-third to over one-half of cases by sexual violence (3, 7-9). For
example, among 613 ever-abused women in Japan, 57% had suffered all three types of abuse—
emotional, physical and sexual. Only 8% of women had experienced physical abuse alone (7).
Likewise in Monterrey, Mexico, 52% of physically assaulted women had also been sexually
abused by their partner (10). Figure 1-1 graphically illustrates the overlap among types of abuse
among ever-partnered women in Leon, Nicaragua (8).
/ Sexual
Never
= 97
/
/
\
\ 3 A
M 74
)
\
a
Physical/
Emotional V
109 / abuse /
abuse
71
5
I
Figure 1-1: Overlap between sexual, physical and psychological violence
experienced by women in Leon, Nicaragua (n=360 ever-married
women).
Most women who suffer any physical aggression generally experience multiple acts over time. In
the Leon study, for example, 60% of women abused in the previous year were abused more than
once, and 20% experienced severe violence more than six times. Among women reporting any
physical aggression 70% reported severe abuse (11). The average number of physical assaults in
the previous year among currently abused women surveyed in London was seven (12); in the
United States in 1997, it was three (5).
Although types of abuse generally overlap in the same relationship, the data presented in Table 1
refer exclusively to the percentage of women who have experienced physical assault. Prevalence
studies of domestic violence represent such a new area of research that comparable figures on
other types of partner violence are not yet available. Because of methodological differences,
even the data on physical violence from these well-designed studies are not directly comparable.
Reported rates of abuse are highly sensitive to the definitions used, how questions are asked, the
population being investigated, and the degree of privacy achieved during the interview (13). As a
result, differences among countries—especially small to moderate differences—might well
represent methodological variations rather than true differences in rates [See Box 1: Enhancing
the Comparability of Data on Domestic Violence].
4
In surveys of partner violence, women usually are asked whether or not they have experienced
any of a list of specific behaviors, such as being slapped, pushed or shoved, hit, punched or
kicked, beaten, or threatened with a weapon. Research has shown that asking behaviorally
specific questions—such as "Have you ever been forced to have sexual intercourse against your
will?"—yields higher rates of disclosure than questions that ask women whether they have been
“abused” or “raped.” (13). Behaviorally specific instruments also allow researchers to assess the
relative severity and frequency of the abuse that women suffer. Surveys generally define
physical acts more severe than slapping, pushing, shoving or throwing an object as “severe
violence.”
At times the focus on “acts” can mask the atmosphere of terror that often permeates these
relationships. For example, in Canada’s national survey of violence against women, one third of
women who had been physically assaulted by a partner said that they had feared for their lives at
some time in the relationship (14). Although international studies have concentrated on physical
violence because it is more easily conceptualized and measured, women routinely say that it is
the psychological abuse and degradation that they find the most difficult to bear (1, 15, 16) (see
Box 2 - Women’s Experiences of Violence).
The Dynamics of Abuse
Many cultures hold that men have the right to control their wives’ behavior and that women who
challenge that right-even by asking for household money or by expressing the needs of the
children—may be punished. Studies from countries as different as Bangladesh, Cambodia, India,
Mexico, Nigeria, Pakistan, Papua New Guinea, Tanzania and Zimbabwe indicate that violence is
frequently conceptualized as physical chastisement—the right of a husband has to “correct an
errant wife. (17-25). As the authors of the Pakistani study note, “Beating a wife to chastise or to
discipline her is seen as culturally and religiously justified...Because men are perceived as the
‘owners’ of their wives, it is necessary to show them who is boss so that future transgressions are
discouraged (p. 39).”
Justifications for violence frequently evolve from traditional notions of the proper roles and
responsibilities of men and women. Men are given relatively free reign as long as they provide
financially for the family. Women are expected to tend the house and mind the children, and to
show their husbands obedience and respect. If a man perceives that his wife has somehow failed
in her role, overstepped her bounds, or dared to challenge his prerogative, then violence can
ensue.
A wide range of studies from both the industrial and developing world identify a remarkably
consistent list of events that are said to “trigger” violence (17-22). These include: not obeying
the husband, talking back, not having food ready on time, failing to care adequately for the
children or home, questioning the man about money or girlfriends, going somewhere without his
permission, refusing the man sex, or suspicions of infidelity. All of these represent transgression
of dominant gender norms.
5
In many developing countries, women share the notion that men have the right to discipline their
wives using force if necessary (See Table 2—from Pop Report). In rural Egypt, upwards of 80
percent of rural women say that beatings are justified under certain circumstances (26).
Significantly, one of the reasons that women most often cite as just cause for beatings is refusing
a man sex (26-29). Not surprisingly, denying sex is also one of the reasons women cite most
often as a trigger for beatings (18, 30-32). This reality clearly has implications for women’s
ability to protect themselves from unwanted pregnancy and sexually transmitted infections.
Societies often distinguish between “just” and “unjust” reasons for abuse as well as between
“acceptable” and “unacceptable” amounts of violence. Thus, certain individuals (usually
husbands or elders) are given the right to physically chastise a woman, within limits, for certain
transgressions. Only if a man oversteps these bounds—for example, by becoming too violent or
for beating a woman without just cause—will others intervene (17, 21, 33, 34).
This notion of “just cause” permeates qualitative data on violence in many parts of the
developing world. As one indigenous woman in Mexico observed, “I think that if the wife is
guilty, the husband has the right to hit her...If I have done something wrong...nobody should
defend me. But if I haven’t done something wrong, I have a right to be defended” (21). Similar
sentiments are expressed among focus group participants in North and South India. “If it is a
great mistake,” notes one husband in Tamil Nadu, “then the husband is justified in beating his
wife. Why not? A cow will not be obedient without beatings.” (25).
Even where culture itself grants men substantial control over female behavior, abusive men
generally exceed the norm (27, 35, 36). For example, data from the Nicaragua Demographic and
Health Survey (DHS) show that, among women who were abused physically, 32% had husbands
who scored high on a scale of marital control compared with only 2% among women who were
not abused physically. The scale included such behavior as the husband’s continually accusing
his wife of being unfaithful and limited her access to family and friends (27).
Women’s Responses to Abuse
Qualitative studies confirm that most abused women are not passive victims but use active
strategies to maximize their safety and that of their children. Some women resist, others flee, and
still others attempt to keep the peace by capitulating to their husband’s demands (3, 37-39).
Thus, what may seem to an observer to be lack of response to living in a violent relationship may
in fact be strategic assessment of what it takes to survive in the marriage and to protect herself
and her children.
A woman’s response to abuse is often limited by the options available to her (38). In-depth
qualitative studies with abused women in Africa, Asia, Latin America, Europe, and the United
States, confirm that fear of retribution, lack of alternative means of economic support, concern
for the children, emotional dependence, lack of support from family and friends, and an abiding
hope that he “will change” are common factors that keep women in abusive relationships. (8, 18,
20, 40, 41) In developing countries, women also cite the stigmatization of being unmarried as an
additional barrier to leaving abusive relationships (18, 34, 42).
6
At the same time, denial and fear of social stigma often prevent women from reaching out fo
help. Studies show that anywhere from 22% to nearly 70 % of abused women surveyed neve:
told another person prior to the interview (see Table 3). Those who do reach out do so primaril}
to family members and friends rather than formal institutions. Only a minority has eve:
contacted the police.
Table 3
Percentage of Physically Abused Women Who Sought Help From Different Sources,
Selected Countries
Never Told
Anyone
%
Bangladesh
Canada
Cambodia
Chile
Egypt
Ireland
Moldova
Nicaragua
UK
68
22
34
30
47
37
38
Contacted
Police
Told
Friends
Told
Family
~o
%
%
45
33
14
03
50
30
28
46
30
44
22
32a
44
37
31
34
31
26
1
16
20
6
17
22
Reference
(43)
(35)
(44)
(45)
(26)
(40)
(46)
(27)
(12)
a 32% told her family; 21% told his family
Despite the obstacles, many abused women eventually do leave violent partners—even if aftei
many years, once the children are grown. In Leon, Nicaragua, for example, 70% of women
eventually leave their abusers (47). The median time that women spend in a violent relationship
is 6 years, although younger women are more likely to leave sooner (8). Studies suggest a
consistent set of factors that propel women to separate permanently: the violence gets more
severe and triggers the realization that the abuser is not going to change, or the violence begins to
noticeably affect the children. Women also cite emotional and logistical support from family or
friends as pivotal in their decision making process. (39,41, 48-50).
Research likewise suggests that leaving an abusive relationship is a process, not a "once-off
event. Most women leave and return several times before they leave for good. The process of
detaching includes periods of denial, self-blame and endurance before women come to recognize
the abuse and to identify with other women in the same situation. This is the beginning of
disengagement and recovery from the abusive relationship (51). Recognizing this process can
help individuals be more understanding and less judgmental when they encounter a woman who
returns to an abusive situation.
Regrettably, leaving does not necessarily guarantee a woman’s safety. Violence sometimes
continues and may even escalate after a woman leaves her partner (52). In fact in the United
States and Canada, an abused woman’s risk of being murdered is greatest immediately after
separation (53) [insert Canada ref] [See Box 3: Domestic Homicide],
7
Explaining Intimate Partner Violence
Researchers have only recently begun to look for individual and community-level factors that
might serve to increase or decrease the rate of partner violence in different settings. Crosscultural research shows that, although violence against women is present in most societies, there
are examples of pre-industrial societies where partner violence is virtually absent (54, 55). These
societies stand as testament to the fact that social relations can be organized in such a way as to
minimize violence against women.
In many places the prevalence of domestic violence varies substantially among neighboring
areas. These local differences are often greater than differences across countries. For example,
in Uttar Pradesh, India the percentage of men who said that they beat their wives varried from
18% in Naintal District to 45% in Banda (56). The percentage that physically forced their wives
to have sex varied from 14% to 36% among districts (see Table 4). This variation raises an
interesting and compelling question: What is it about these different settings that accounts for an
almost three-fold difference in wife beating?
Table 4
Variations in Men’s Attitudes and Reported Use of Violence,
Selected Districts Uttar Pradesh, India (n = 6,695)
Admit to forcing
wife to have sex
Aligarh
Banda
Gonda
Kanpur Nag ar
Naintal
Admit
to hitting
wife %
Hit wife
in last year
%
Ifwife disobeys,
she should be
beaten %
31
17
36
14
21
15
50
27
11
10
29
45
31
22
18
17
33
20
10
11
%
(56)
Recently, researchers have become more interested in asking and answering such questions,
although the current research base is inadequate for the task. One of the few sociodemographic
factors that has consistently emerged as predictive of higher rates of abuse in population-based
studies is lower socioeconomic status. Studies in many countries show that, although violence
among intimates cuts across all socioeconomic groups, women living in poverty are
disproportionately affected (11, 14, 27, 44, 45, 57-60).
It is as yet unclear why poverty increases the risk of violence—whether it is due to low income
itself or other factors that accompany poverty, such as crowding or hopelessness. For some men,
living in poverty is likely to generate stress, frustration, and a sense of inadequacy for having
failed to live up to their culturally defined role of provider. It may also operate by providing
ready fodder for marital disagreements and/or by making it more difficult for women to leave
8
violent or otherwise unsatisfactory relationships. In all likelihood, low SES acts as a “marker
for a variety of life conditions that combine to increase women’s risk (33).
Another risk factor that appears especially robust across settings is witnessing or experiencin;
violence as a child. Studies in Nicaragua, Chile, Cambodia and Canada have all found that rate
of abuse are higher among women whose husbands were either beaten themselves as children o
witnessed their mothers being beaten (11, 35, 44, 45). Although men who physically abuse thei
wives frequently have violence in their background, not all boys who witness or suffer abus>
grow up to become abusive themselves. An important theoretical question is what distinguishe
those men who are able to form healthy, non-violent relationships despite childhood adversity
from those who become abusive?
Cross-cultural research suggests a variety of structural and sociocultural factors that contribute t(
higher rates of violence as well. Levinson’s ethnographic study of 90 pre-industrial societie
throughout the world identified four factors that, in combination, are strongly associated with ;
high prevalence of violence against women (55). These include economic inequality betweei
men and women; a pattern of using physical violence for conflict resolution; male authority anc
decision-making in the home; and divorce restrictions for women.
Likewise, in their comparative study of high versus low violence societies, Counts, Brown ant
Campbell found that societies with the least domestic violence were those that had community
sanctions against domestic violence (either formal legal sanctions or the cultural expectation tha
neighbors should intervene when a woman is beaten) and those where abused women had acces;
to sanctuary, either in the form of shelters or family support. The "Sanctions and Sanctuary'
framework further theorizes that violence will be highest in societies where women’s status is ii
transition. Where women have very little status, violence is not “needed” to enforce male
authority. Where women have high status, they have achieved sufficient power en masse t(
change the gender power dynamic. Domestic violence is highest when women begin to assume
non-traditional roles and/or enter the work force.
But the causes of violence are neither fully structural nor fully individual. Practitioners, activists
and researchers are increasingly using an “ecological model” to understand the interplay o:
personal, situation, and sociocultural factors that combine to cause abuse (33, 61). An ecologica.
approach to abuse argues that no one factor alone “causes” violence but rather that a number ol
factors combine to raise the likelihood that a particular man in a particular setting may aci
violently toward his partner.
The more risk factors present, the higher the likelihood ol
violence.
The model can best be visualized as four concentric circles. The innermost circle represents the
biological and personal history that each individual brings to his or her behavior in relationships.
The second circle represents the immediate context in which abuse takes place—frequently the
family or other intimate or acquaintance relationship. The third circle represents the institutions
and social structures, both formal and informal, in which relationships are embedded—
neighborhood, workplace, social networks, and peer groups. The fourth, outermost circle includes
the economic and social environment, including cultural norms.
9
A range of studies suggests various factors that appear to increase the likelihood of partner abuse
at each of these four levels.
•
At the individual level these include being abused as a child or witnessing marital
violence in the home (62, 63), having an absent or rejecting father (61), and frequent
use of alcohol (63-68).
•
At the level of the family and relationship, cross-cultural studies cite male control of
wealth and decision-making within the family (55, 64) and marital conflict as strong
predictors of abuse (58, 69).
•
At the community level women’s isolation and lack of social support, together with
male peer groups that condone and legitimize men’s violence, predict higher rates of
violence (43, 64, 70).
•
At the societal level violence against women appears most common where gender
roles are rigidly defined and enforced (33) and where the concept of masculinity is
linked to toughness, male honor, or dominance (54, 71). Other cultural norms
associated with abuse include attitudes that condone physical punishment of women
and children, acceptance of violence as a means to settle interpersonal disputes, and
the perception that men have “ownership” of women (33, 55, 63, 72).
Figure 2. An ecological framework for understanding gender-based violence
(adapted from Heise, 1998 (33)).
By combining individual level risk factors with cross cultural studies, the ecological model
contributes to understanding gender-based violence by explaining on one hand, why some
societies and some individuals are more violent than others and, on the other hand, why women
are so consistently the victims.
10
The Consequences of Abuse for Health and Well Being
The consequences of abuse are profound and extend beyond the health and happiness c
individual women to affect the wellbeing of entire communities. Living in a violent relationshi
affects women’s sense of self worth and their ability to act in the world. Studies show the
abused women are routinely prohibited from accessing resources, participating in public life, c
receiving emotional support from friends and relatives. Not surprisingly, such women ar
frequently less able to care for themselves and their children or to pursue jobs and careers as the
may have wanted.
Impact on health. A growing body of literature documents that living with an abusive partne
can have profound impacts on woman’s health. Violence has been linked to a host of differer
health outcomes, both immediate and long term. Figure 3 summarizes the various consequence
that have been associated with abuse in the scientific literature. Although violence can hav
direct health consequences, such as injury, victimization also operates through a number o
pathways to increase women’s risk offuture ill health. Like tobacco or alcohol use, victimizatio:
can best be conceptualized as a risk factor for a variety of diseases and conditions.
Studies show that women who have experienced physical and/or sexual abuse in childhood o
adulthood experience a higher frequency of negative health outcomes related to physica
functioning, psychological well-being, and risk behaviors, including smoking, physical inactivity
drinking and drug use. (68, 73-79). A history of violent victimization puts women at increase*
risk of depression, suicide attempts, chronic pain syndromes, psychosomatic disorders, injury
gastrointestinal disorders, irritable bowel syndrome; and a variety of reproductive healtl
consequences (see below). Taken together, existing research suggests several emerging
conclusions about the health consequences of abuse:
•
The influence of abuse can persist long after the abuse has stopped;
•
The more severe the abuse, the more severe its impact on women’s physical am
mental health; and
•
The impact of different types of abuse and multiple episodes over time appears to be c
Violence and Reproductive Health. Women who live with violent partners have a difficult time
protecting themselves from unwanted pregnancy or disease. Violence can lead directly to unwantec
pregnancy or sexually transmitted infections (STIs) including HTV/AIDS via coerced sex, oi
indirectly by interfering with a woman’s ability to use contraceptives and or condoms (80, 81)
Studies consistently show that domestic violence is more common in families with many children
(5, 25, 27, 28, 45, 57, 82). Researchers have long assumed that the stress of many childrer
increases the risk of violence, but recent data from Nicaragua suggests that the relationship maj
be the opposite. In Nicaragua, the onset of violence largely precedes many children, suggesting
that violence may be a risk factor for having many children (80% of violence begins within the
first four years of marriage) (8).
11
Violence also occurs during pregnancy, with consequences not just for the woman, but for the
developing fetus. Studies from Nicaragua, Egypt, Chile and Cambodia have found that as many as
one of every four women has been physically or sexually abused during pregnancy, usually by a
partner (26, 27, 35, 44, 45). In the United States estimate of abuse during pregnancy range from
3% to 11% among adult women and up to 38% among low-income, teenage mothers (83-87).
Violence during pregnancy has been associated with miscarriage, late entry into pre-natal care, still
birth, pre-term labour and birth, foetal injury and death (80) and low birth weight, which is a major
cause of infant death in the developing world. (85, 88-91).
In Leon, Nicaragua, for example,
abused women have a four times greater risk of having a low birth weight infant than non-abused
women, even after controlling for other factors. In this setting, 16% of low birth weight among
infants can be attributed to domestic violence (91).
Elsewhere, domestic violence accounts for a substantial but largely unrecognized proportion of
maternal mortality. A recent study among 400 villages and 7 hospitals in Pune, India, found that
16% of all deaths during pregnancy were due to domestic violence (92). The study also
demonstrated that 7 out of 10 maternal deaths in this region normally went unrecorded and 41%
of recorded deaths were misclassified.
Use of Health Services. Given the long-term impacts of violence on women’s health, it is not
surprising that victimization also increases women’s use of services, thereby increasing health
care costs. Studies in the United States, Zimbabwe and Nicaragua indicate that women who have
experienced physical or sexual assault in either childhood or adulthood use health services more
frequently than their non-abused peers (74, 76, 93-96). On average, abuse victims have more
surgeries, physician visits, hospital stays, pharmacy visits and mental health consultations over
their lifetimes than non-victims, even after controlling for potential confounding factors.
For example, in one study at a major HMO in the United States, researchers found that having
been raped or assaulted was a stronger predictor of health care use than was any other variable,
including a woman’s age or other health risks such as smoking (75). Women who had been
victimized sought medical attention twice as often as non-victimized women in the year of the
study (which was not the year the woman was victimized). The medical care costs of women who
were raped or assaulted were 2.5 times higher than the costs of non-victims, after controlling for
confounding factors (75).
Economic Costs of Violence. In addition to its human costs in terms of pain and suffering,
violence places an enormous economic burden on societies in terms of lost productivity and
increased utilization of social services. Among women in Nagpur, India, for example, 13%
reported that they had missed paid work because of abuse, missing an average of 7 workdays per
incident. Eleven percent reported that they were unable to perform household chores due to an
incident of violence (97).
Although domestic violence does not have a consistent impact on women’s overall likelihood to
be employed, it does appear to influence women’s earnings and their ability to keep a job (95, 98,
99). A study from Chicago found that women with histories of domestic violence were more
likely to have experienced spells of unemployment, to have job turnover, and to have suffered
more physical and mental health problems that could affect job performance. They also had
12
lower personal incomes and were significantly more likely to receive public assistance tha
women who did not report domestic violence (99). Similarly, in a study from Managui
Nicaragua, abused women earned 46% less than women who did not suffer abuse, even afte
controlling for other factors that affect earnings (95).
Impact on Children. Conflict in the home frequently has a “spill-over” effect on youn
children. Children who witness marital violence are at higher risk for a host of emotional an
behavioral problems, including anxiety, depression, poor school performance, low self esteen
disobedience, nightmares, and somatic health complaints (8, 100-102). They are also more likel
to act aggressively in both childhood and adolescence (103, 104). Indeed, studies from Nort
America indicate that children who witness violence between their parents frequently exhib:
many of the same behavioral and psychological disturbances as children who are themselve
abused (101, 105).
Regrettably, children are often present during domestic altercations. Sixty-four percent of abuse*
women in Ireland (40) said that their children routinely witnessed the violence, as did 49% o
battered women in Nicaragua (8), and 50% of abused women in Monterrey, Mexico (10). In thNicaragua study, children of battered women were more than twice as likely to suffer fron
learning, emotional and behavior problems, and almost seven times more likely to be abuse*
themselves (either physically, emotional or sexually) than were children of non-battered womei
(8).
Recent evidence suggests that violence may undermine child survival as well (106, 107). /
study in Leon, Nicaragua, for example, found that after controlling for other possibL
confounders, the children of women who were physically and sexually abused by a partner wer<
six times more likely to die before the age of five. Partner abuse accounted for as much as on<
third of child deaths in this region (107). Another study in the Indian states of Tamil Nadu an*
Uttar Pradesh found similar results. Women who had been beaten were significantly more likeb
than non-abused women to have experienced an infant death or pregnancy loss (abortion
miscarriage, still birth), even after controlling for well-established predictors of child mortality
such as women’s education, age and parity (106).
13
Prevention and Policy Response
To date, the majority of work on partner violence has been spearheaded by women’s
organizations, with occasional funding and assistance from government. Where government has
become active—as in North America, Australia, parts of Europe, and Latin America—it has
generally been in response to demands by civil society for more constructive action on this issue.
The first wave of reform has generally involved some combination of legislative reform, police
training, and the establishment of specialized services for victims. Scores of countries have
passed laws related to domestic violence, although many justice system officials are still unaware
of the changes or unwilling to implement them. Individuals working within the system
frequently share the same biases and prejudices that dominate the society at large. Experience
has repeatedly shown that without sustained efforts to change institutional culture and practice,
most legal and policy reforms remain cosmetic.
Despite over twenty years of activism against gender violence, remarkably few interventions
have been rigorously evaluated. In fact, in their recent review of family violence interventions in
the United States, the National Research Council identified only 34 studies that attempted to
evaluate interventions related to partner abuse. The majority of these (19 out of 34) focused on
law enforcement strategies, reflecting the strong bias that exists in the United States toward
criminal justice approaches to dealing with violence (108). The available research base on
interventions in developing country settings is even more limited. Only a handful of studies exist
that attempt to critically examine some of the interventions currently underway. Among these are
a review of programs dealing with violence against women in four states of India and an initiative
by UNIFEM to systematize lessons learned from projects funded through the United Nations
Violence Against Women Trust Fund (Roxanna Carrillo, personal communication, May 2000).
There is an overwhelming need in the field of partner violence for greater attention to evaluating
the impact of current interventions. Outlined below is a description of the most common
interventions being tried globally to reduce partner violence and to respond to the needs of
victims and perpetrators. Although commonly understood as “good practice,” there is little
concrete data available to support the effectiveness of these interventions in reducing violence or
mitigating its impact.
Support and refuge for victims. In the industrial world, women’s crisis centers and battered
women’s shelters have been the cornerstone of programs to assist victims of domestic violence.
In 1995, there were approximately 1,800 programs in the United States for abuse victims, 1200 of
which provided emergency shelter in addition to emotional, legal and material support to women
and their children. (109). Such centers generally provide support groups and individual
counseling, job training, programs for children, legal assistance, help negotiating social services,
and referrals for drug and alcohol treatment. Most refuges and women’s centers in Europe and
the United States were founded by women’s activists, although many shelters and crisis centers
today are run by professionals and receive government support.
Since the early 1980s, shelters and women’s crisis centers have sprung up in many developing
countries as well. Most countries have at least a handful of NGOs providing specialized services
and advocacy for victims of abuse and some—such as Nicaragua—have hundreds. Because of
the expense of maintaining shelters, many developing countries have avoided this model in favor
14
of telephone hotlines and/or non-residential crisis centers that provide self-help support group
legal services, counseling and advocacy services.
In settings where maintaining a formal shelter is not feasible, women have devised oth<
emergency housing options, such as organizing informal networks of “safe homes” where wome
in distress can seek temporary shelter in the homes of neighbors. Elsewhere communities ha\
designated the local church as a “sanctuary” where women can stay with their children over nig]
if their partner arrives home drunk or violent.
Legal remedies and judicial reforms. The 1980s and 1990s have witnessed a wave of refonr
directed at transforming how the law treats physical and sexual abuse by an intimate partner. 1
the last decade, for example, 24 Latin American and Caribbean countries have passed legislatio
specifically designed to address domestic violence (110, 111). The most common refonr
involve criminalizing physical, sexual and psychological abuse by intimate partners eithc
through special domestic violence laws or by amending existing penal codes. Laws vai
according to the range of perpetrators and acts covered, as well as the penalties imposed.
The logic behind such reforms is to communicate that domestic violence is a crime and will n<
be tolerated. By bringing domestic violence into the public sphere, advocates hope to counter th
notion that violence is a private, “family matter.” Legal reforms have been accompanied b
various approaches to increase women’s access to justice and improve successful prosecutioi
Industrial countries have tried various experimental reforms including specialized domesti
violence courts, training of police, court and prosecutorial personnel, and providing victii
advocates to assist women through the criminal justice system. Although little rigorous data ar
available to evaluate these measures, the recent National Academy of Sciences review of Famil
Violence Interventions concludes: “Anecdotal evidence suggests that specialized units an
comprehensive reforms in police departments, prosecutor’s offices and specialized courts hav
improved the experience of abused children and women (108).”
Similar experiments are underway in a number of developing countries. In India, for example
state governments have established Legal AID Cells, Family Courts, Lok Adalats or People'
Courts, and Mahila Lok Adalats or Women’s Courts. A recent evaluation notes that these bodie
primarily function as conciliatory mechansims, relying exclusively on mediation and counselin
to promote family reconciliation. Interviews with key informants suggest that even as
conciliatory mechansims, these entities leave much to be desired.
Mediators generail
subordinate women’s well-being and safety to the state’s interest in family preservation (112).
15
Transforming Police Practice. Next to support services for victims, efforts to reform police
practice are the second most common form of domestic violence intervention globally.
Advocates sought to make the police more responsive to the victims needs and more aggressive
toward arresting perpetrators. Early efforts focused on training; but when training alone proved
largely ineffective at changing police behavior, advocates began to seek mandatory arrest laws
and pro-arrest policies to force officers to take a more aggressive stance toward domestic
violence. Arrest for domestic violence is perhaps the best-studied intervention for family
violence.
Early support for arrest as a means to reduce recidivism came from a 1984 research experiment in
Minneapolis, Minnesota, that suggested that arrest cut in half the risk of future assaults over a six
month follow-up period, compared with separating couples or advising them to get help (113).
These results were widely publicized and led to a dramatic shift in police policies toward
domestic violence throughout the United States.
Efforts to replicate the Minneapolis findings in five additional jurisdictions, however, failed to
support the initial findings. The replication studies found that on average arrest produced no
discernable effect on recidivism (114, 115). Detailed analysis revealed that in the United States
the effect of arrest varied with characteristics of the perpetrator. When the perpetrator was
married, employed, or both, arrest reduced repeat assaults, but for unemployed and unattached
men, arrest actually increased abuse in some cities. The impact of arrest also varied by
community. Men living in communities with low unemployment were deterred by arrest
regardless of their individual employment status; suspects in high unemployment areas were
more violent following arrest than after a warning (116). These findings have led some US
researchers to conclude that mandatory arrest laws should be repealed in areas of concentrated
poverty (117).2
Elsewhere governments have experimented with “All Women’s Police Stations”—an innovation
that began in Brazil and has now spread throughout Latin America and parts of Asia (119, 120).
Although good in theory, evaluations show that such efforts to date have experienced many
problems (112, 120-124). While the presence of a women’s police station increases the number
of abused women coming forward, frequently the women require services—such as legal advice
and emotional counseling—that are not available at the stations.
Moreover, the assumption that female officers will be more sympathetic to victims has not
always proved true. Female officers assigned to all-women stations frequently have been
ridiculed by their peers and have become demoralized. In some settings, the creation of
specialized police cells for crimes against women has made it easier for other police units to
dismiss women’s complaints. As a review of All Women Police Stations (AWPS) in India
observes, “Women victims are forced to travel great distances to register their complaints with
the AWPS and can not be assured of speedy neighborhood police protection.” To be viable, this
strategy must be accompanied by sensitivity training for officers, mechanisms to reward and
legitimate the work, and provision of a wider array of services (112, 120, 122).
2 This is not to say that arrest serves no useful purpose in domestic violence cases, but only that it does not appear to
reduce recidivism across the board. Advocates’ original intent in promoting arrest was not to deter future violence,
but to interrupt cunent abuse and to ensure women’s equal protection under the law (118).
16
Health service interventions. In recent years advocates have turned their attention towart
reforming the response of health care providers to victims of abuse. Most women are likely t<
interact with the health system at some point in their lives—when they seek contraception, giv<
birth or seek care for their children. This makes the health care setting an important opportuniv
to identify women experiencing abuse and provide them with needed support and referral
Unfortunately, studies show that in most countries, doctors and nurses rarely ask women whethe
they are being abused, even when there are obvious signs of violence (125-132).
Existing health care interventions have focused on sensitizing providers, encouraging routine
screening for abuse, and institutionalizing protocols for the proper management of abuse. /
growing number of countries, including Brazil, Ireland, Malaysia, Mexico, Nicaragua, the
Philippines and South Africa have initiated pilot projects to train health workers to identify anc
respond to abuse (133-135). Several Latin American countries—including Peru, Bolivia
Ecuador, Mexico, and Nicaragua-have also incorporated guidelines for addressing domestic
violence into their national health sector policies (136).
Presently, the US Centers for Disease Control is evaluating a range of interventions aimed a
reforming the response of the health sector to abuse (137). Existing research suggests tha
making procedural changes in client care—including adding chart prompts, or integrating
questions on abuse into standard intake forms—have the biggest impact on provider behavioj
(138, 139). Confronting the underlying beliefs and attitudes is also important. In South Africa,
for example, the Agisanang Domestic Abuse Prevention and Training Project (ADAPT) and its
partner, the Health Systems Development Unit of the University of Witwaterstrand, developed z
reproductive health and gender training program with a strong domestic violence component
The course used role playing, popular sayings and wedding songs to help participants analyze
common notions about violence and the proper roles of men and women. Only then did training
turn to the nurses' responsibility as health professionals. A post-training survey found thai
participants no longer believed that beating a woman was justified and that most accepted the
concept of marital rape (131).
Although active screening for abuse (questioning clients about histories of victimization) is
generally considered “good practice” in the field of health care and intimate violence, there has
not yet been any systematic look at the impacts of this practice on women’s safety, satisfaction
with care, or help-seeking behavior. Studies repeatedly show that women are open to being
queried about violence in a non-judgmental way (126, 127, 140), but little information is
available to evaluate whether and under what conditions this strategy is helpful (141).
Outreach and Advocacy. Outreach and advocacy have been a cornerstone of virtually all
responses to domestic violence evolving from the non-profit sector. Outreach initiatives seek to
support domestic violence victims in their homes and communities, via peer promoters, women’s
defensoras, and other mechanisms designed to take information about women’s rights and
available services to women, rather than expect them to come to services. Frequently, NGOs
recruit and train peer advocates from the ranks of former domestic violence survivors who have
used their services.
17
Both government and non-profit projects frequently employ individual “advocates” to provide
survivors with support, information and advice. Advocacy recognizes that individuals coming
from a position of fear and isolation will often require assistance to negotiate the justice system,
family welfare bureaucracies, and benefit entitlements. It is the emphasis on rights and
entitlements that distinguishes advocacy from more familiar concepts like support. Some of the
more innovative projects involve collaborations where NGO-employed advocates work directly
out of state-run police stations, prosecutors’ offices, or hospitals, to help women negotiate these
systems and receive quality service.
Several advocacy and outreach schemes have been evaluated. The Domestic Violence Matters
(DVM) project in Islington, UK, for example, placed civilian advocates in neighborhood police
stations. These advocates would contact all victims within 24 hours of their call to the police.
Another UK initiative, the Domestic Violence Intervention Project (DVIP), combined an
education program for violent men with proactive responses and advocacy for their partners. A
recent review of these programs found that the DVM successfully decreased the number of repeat
calls to the police and by inference, repeat victimizations. It also increased women’s use of new
services, including refuges, solicitors and support groups. The DVIP accessed greater numbers of
ethnic minority and professional women than other domestic violence services. In both cases,
women welcomed outsiders making the first move and felt that the intervention helped them take
actions that increased their safety and accelerated the process of change (142).
Alternative Sanctions. Instead of threatening jail time, some communities are experimenting
with other means to raise the social cost of violent behavior. A common civil law approach is to
issue court orders that can prohibit a man from contacting or abusing his wife, mandate that he
leave the marital home, order him to pay maintenance and/or child support, and require him to
seek counseling or substance abuse treatment.
In multiple sites around the United States, researchers have found that although victims feel that
protective orders are effective, there is generally no difference in rates of victimization among
those with orders and those without (143). By contrast, Harrell (144) found that protective orders
-were effective in Denver and Boulder for at least a year in preventing repeat violence in
comparison with victims without orders. Multiple studies confirm that arrests for violation of a
protection order are rare, a fact that tends to undermine their effectiveness in preventing violence
(145). Other research shows that protective orders can enhance women’s self esteem but are less
effective against men with serious criminal records (146, 147).
Elsewhere, communities have explored techniques such as public shaming, picketing an abuser’s
home or workplace, or requiring community service to censure abusive behavior. Activists in
India often stage dharnas, a form of public shaming and protest, in front of the house or
workplace of abusive men (112). In the US state of Texas, an innovative judge is sentencing
batterers to “shame sentences,” ordering one abusive man to apologize to his wife publicly on the
steps of city hall, and another to carry a sign around a local shopping mall that read “I went to jail
for assaulting my wife. This could be you” (148).
18
Coordinated Community Interventions. An increasingly common model uses coordinating
councils or interagency fora to improve and monitor community-level responses to victims anc
perpetrators (147). Apart from networking and exchanging information, such councils, at theii
best: 1) identify and rectify gaps and bottlenecks in the provision of services; 2) promote gooc
practice via training and practice guidelines; 3) track the disposition of cases and conduci
institutional audits to the assess the practice of different agencies; and 4) sponsor community
awareness and prevention activities. Originally patterned after model programs in Duluth
Minnesota, Quincy, Massachusetts and San Francisco, this approach to intervention has spreac
widely throughout the United States, the UK, Canada and parts of Latin America.
For example, the Pan American Health Organization (PAHO) has initiated pilot projects in 16
Latin American countries to explore the utility of this type of approach in both urban and rural
settings. In rural settings, the coordinating councils include such individuals as the local priest,
the mayor, community health promoters, women’s groups, and the magistrate.
The PAHO
project began with a qualitative research project—known as the "Ruta Critica"—to identify whai
happens to women in these communities when they reach out for help. The results, summarized
in Box 4, document the repeated failure of systems to meet women’s needs and highlight the
critical need for better coordination among services.
Although there are likely many positive benefits of community interventions, these programs
have seldom been evaluated. One study that looked primarily at process variables, found a
statistically significant increase in the percentage of police calls that resulted in arrest and the
percentage of arrests that resulted in prosecution following the implementation of a community
intervention project in each of three communities (149). The study also found a significant
increase in the percentage of men sent to mandatory counseling in each of the communities,
although it is unclear what impact, if any, these changes had on rates of abuse.
Qualitative evaluations have noted that many of these interventions focus primarily on
coordinating refuges and the criminal justice system, at the expense of wider involvement from
faith communities, schools, the health system, or other social service agencies. A recent review
of multi-agency fora in the United Kingdom notes that while councils can improve the quality of
services that women and children receive, “interagency work can act as a “smokescreen’ and a
face-saver, while very little actually changes. To avoid this outcome, groups should commit to
self-evaluation criteria that include concrete changes in policies, practice and user satisfaction
(150).
Perpetrator Treatment. Batterer treatment programs are an innovation that began in the United
States that has now spread to throughout Canada, Europe, Australia and parts of the developing
world (151-153). Most programs use a group format and include skill building and reflection
around issues of gender role socialization, stress and anger management, empathy and taking
responsibility for one’s own actions. Although potentially beneficial for a small group of men.
studies show that the majority of men never complete the required counseling, even in those
programs where the court mandates their participation (154). Evaluation of Britain’s flagship
Violence Prevention Programme, for example, showed that 65% did not show up for the first
session, 33 /a attended fewer than six sessions, and only 33% went onto the second stage group
(155).
19
Unfortunately, only a handful of batterer treatment programs have been rigorously evaluated. In
the United States, research suggests that the majority of men (53%-85% ) who complete such
programs remain physically nonviolent for up to two years after treatment (156, 157). But
between one-third to one-half of men who enroll in such programs fail to complete them (156).
Moreover, while men may refrain from physical violence after treatment, many men continued
other types of threatening or coercive behavior toward their partners (158).
Nevertheless, a recent evaluation of programs in four US cities found that most abused women
felt “better off’ and “safe” after their partners entered treatment. This study found that after 30
months, nearly half the men had used violence once, 23% of men had been repeatedly violent and
continued to inflict serious injuries, and only 21% of men were neither physically nor verbally
abusive. Sixty percent of couples had split up and 24% had no contact (157).
According to a recent international review by researchers at the University of North London
(155), evaluations collectively suggest that treatment programs work best if they:
•
continue for longer rather than shorter periods;
•
can change men’s attitudes enough for them to discuss their behavior;
•
can sustain men in membership; and
are integrated with a criminal justice system which takes prompt, rigorous and agreed
upon action in cases of a breach of conditions.
In Pittsburgh, for example, the no-show rate dropped from 36% to 6% between 1994 and 1997
(albeit in the context of much reduced take-up overall), when the justice system began issuing
arrest warrants for men who failed appear at the program intake interview o (155).
Prevention Campaigns. Women’s organizations have long used communication campaigns,
small-scale media, and other community events to raise awareness about domestic violence and
attempt to change social norms. Although there is evidence that such campaigns achieve
considerable reach, there is little data on the impact of such efforts on attitudes or rates of
violence. During the 1990s, for example, a network of over 100 women’s groups in Nicaragua
mounted an annual mass media campaign to raise awareness of the impact of violence on women
(159). Using slogans such as “Quiero vivir sin violencia” (I want to live free of violence), the
campaign mobilized communities against abuse. According to a national health survey conducted
in 1998, more than half of the Nicaraguan population had heard at least one of the campaign’s
messages, and one-half of all women who had heard the messages were able to repeat at least one
of the slogans. (27) Likewise, UNIFEM, together with eight other UN Agencies, has been
sponsoring a series of regional campaigns against gender violence that organize around the slogan,
“A Life Free of Violence: It’s our Right” (160).
Anti-violence education programs. Despite a growing number of violence prevention
initiatives aimed at youth, only a small percentage has incorporated elements specifically
designed to address violence in intimate relationships. The vast majority of such programs
promote general conflict resolution skills without explicitly addressing the emotional, cultural,
20
and gender-related dynamics of violence in intimate relationships. Indeed, there is much room t<
integrate exercises that explore relationships, coercion and control, and gender norms int<
existing prevention programs aimed at reducing school violence, bullying, delinquency, and othe
“problem behaviors.” In developing country settings, similar exercises could be integrated inti
family life education curricula, youth development schemes, and reproductive and sexual healtl
programs.
The programs that do explicitly address dating violence and other forms of abuse tend to b<
“stand-alone’ initiatives sponsored by entities working to end violence against women (See Bo:
5: Innovative Prevention programs). Only a handful of these programs have been evaluated
including a dating violence prevention curricula in Canada (161), and a curricula entitled, Skill
for Violence-Free Relationships,” designed for 7th grade students (162) . Using pre-test, post-tes
designs, these evaluations found positive changes in knowledge and attitudes toward relationshi]
violence (see also (163)), but no longitudinal studies exist to assess the impact of such change:
on students’ future involvement in violent relationships.
Principles of “Good Practice”
As the above review makes clear, activists and program planners have had to rely more oi
experience and instinct than on sound research when setting priorities and/or designing
interventions. Despite the lack of rigorous evaluations, the field of domestic violence is rich ir
experience. The accumulated wisdom of hundreds of service providers, advocates anc
researchers suggests a number of principles to help guide “good practice” in the field of domestic
violence.
Actions to address violence must take place at both the national and the local level to be
effective
An enduring lesson to emerge from violence organizing to date is that actions to address violence
must take place at both a national and local level. At a national level, efforts must include actions
to improve the status of women, to establish appropriate norms, policies and laws for responding
to abuse, and to create a social environment conducive to non-violent relationships (See Box 6:
Agenda for Change). Countries in both the industrial and developing world have found it usefuJ
to create a formal mechanism or process for developing national action plans to achieve suck
goals. To be effective, such plans should include clear objectives, lines of responsibility anc
timeframes, and be backed by an adequate commitment of resources. The process of developing
such a plan provides an excellent means to engage diverse social actors, including representatives
from different government ministries, women’s NGOs, service providers, and professional
associations.
But experience suggests that national-level action alone will never be sufficient to transform the
landscape of intimate violence. Even in countries like the United States and Canada where
national movements against domestic violence have been active for over 25 years, the options
that an abused woman faces and the responses she receives from formal institutions like the
police, still largely depend on where she lives. In settings where the community has organized at
21
a local level, where groups exist to train and monitor the response of formal institutions, and
where efforts have been made to challenge the norms that perpetuate abuse, a woman
experiencing violence faces one reality. Absent such local organizing, she faces quite another
(164).
Promoting women 9s safety and autonomy should guide all decisions related to interventions
Interventions should be designed to work with women and respect their decisions, rather than
make assumptions about what is best for them. Indeed, a central tenet of feminist organizing
around violence is that women are generally the best judge of their own situation. Interventions
that adopt this stance are generally rated as more successful both by women themselves and by
external evaluators. Recent reviews of different programmatic responses to domestic violence in
the Indian states of Maharashtra and Madhya Pradesh, Karnataka and Gujurat, for example,
repeatedly emphasized that the success or failure of endeavors was defined largely by the
attitudes and perspective that organizers brought to their work (165). The same service - for
example a shelter home or counseling—had dramatically different consequences for women
depending on whether it was run by individuals who prioritized women’s safety and autonomy or
prioritized “rehabilitating fallen women” or family reunification (165).
In general most interventions that take control away from the women - like mandatory reporting
by health workers to the police - have proven to be counterproductive. They can jeopardize a
woman’s safety and make it less likely that women will come forward for care (166-169).
Concern has been raised, for example, about the recent proliferation of laws designed to require
health workers to report suspected cases of abuse. Such laws transform doctors into arms of the
justice system and fundamentally undermine women’s autonomy and the “emotional safety” of
the clinical encounter. Ironically, the template for this intervention, mandatory reporting for
suspected child abuse, has itself never been evaluated in terms of its positive or negative impact
on children’s safety and well-being (108).
Efforts to reform the response of institutional actors—such as the police, health workers, or
the judiciary—must go beyond “training” to include system-wide efforts to change
institutional culture
Experience has shown that little lasting change is achieved from short term efforts to “sensitize”
institutional actors, unless there is a real effort to engage the whole institution in which an
individual is embedded—the leadership, the way in which performance is evaluated and
rewarded, as well as cultural biases and beliefs (170, 171). In the case of reforming health care
practice, for example, training alone has seldom been sufficient to change providers’ behavior
toward victims of violence (172, 173). Although training can improve knowledge and practice in
the short term, the impact of training generally erodes unless accompanied by institutional level
changes in policies, protocols and performance criteria (172, 174). Even the most motivated
individuals cannot sustain new behaviors in the face of an indifferent or hostile institutional
culture.
22
Interventions must emphasize coordination and work with multiple sectors in the same locality
Different sectors such as the police, health, judiciary and social support services must work
together in order to meet the range of needs that women in violent relationships experience.
Historically the tendency of programs has been to concentrate effort on one sector—conducting
training of the police and the judiciary, for example—rather than emphasizing engaging all
relevant actors in a particular setting. Experience suggests that single sector interventions are
often ineffective because they address only one aspect of a dysfunctional social system (112).
Conclusions and Recommendations
The evidence available, although limited, shows violence against women by intimate partners to
be a serious and widespread problem in all parts of the world. There is also growing
documentation of the impact of violence on women’s physical and mental health and wellbeing.
In spite of the growing recognition of this, more needs to be done, particularly in the area of
primary prevention. Major investment is needed in both research and program development.
The following section provides some guidance as to where this investment would be most useful.
Donors and governments should invest heavily in violence related research over the next
decade.
The lack of a clear theoretical understanding of the causes of domestic violence and its
relationship to other forms of interpersonal violence has frustrated efforts to build an effective
global response. Studies to advance our theoretical understanding of violence are needed on a
variety of fronts, including:
> Studies that examine the prevalence, consequences, and risk and protective factors of partner
violence in different cultural settings, using standardized methodologies and measures;
> Longitudinal research that studies the developmental trajectory of violent behavior against
women and whether and how it differs from the development of other violent behaviors;
> Studies that explore the impact of violence from a life-course perspective, investigating the
relative impact of different types of violence on women’s health and well-being and whether
the effects of victimization are cumulative;
> Studies that explore the developmental life course of adults who form healthy, non-violent
relationships despite past traumas or experiences known to increase risk of abuse.
In addition, greater investment is needed in research to help advance intervention, both to make
the case for investment to policy-makers as well as to inform the design and implementation of
programs. In the next decade, priority attention should go to:
> Increased documentation of the range of strategies and interventions that exist around the
world to combat gender-based violence;
23
> Studies that calculate the economic costs of intimate partner violence, in terms of lost
productivity, health care costs, costs related to policing and social services, etc.;
> Studies designed to evaluate the immediate and long-term effects of programs designed to
prevent and respond to intimate partner violence, including school based education programs,
legal and policy changes designed to deter violence, services for victims, and campaigns
designed to change social norms.
Programs shouldfocus more energy on the primary prevention of violence
To date, the vast majority of energy and investment in partner violence programs have focused on
responding to the needs of abused women and their children. Although understandable, this
tendency to design programs to assist “victims,” means that the work of primary prevention is
often lost.
Lack of attention to prevention derives in part from the fact that women’s NGOs—the primary
architects and engines of work on domestic violence internationally—have traditionally been
under-funded and overworked. Although most groups working on partner violence are
committed to eliminating violence and empowering women, their ability to prioritize prevention
has frequently been overwhelmed by the sheer number of women in need of assistance. As one
activist observed, “When facing hundreds of women in need of emergency help now, the longer
term agenda tends to slip.”
The situation is complicated by the fact that the entire field of domestic violence has been
severely under funded, and the monies that are available have tended to be short term and focused
on either services or discrete “projects.” Although women’s groups have been incredibly creative
in their efforts to change social norms through campaigns, workshops, street theater and the like,
they have seldom had the technical expertise or financial backing necessary to move beyond
“awareness raising.”
True progress in the field of violence prevention awaits serious attention by policymakers and
activists to the task of empowering women, confronting the social dislocations and economic
upheavals that disempower men, and creating a social environment that supports and promotes
equitable, non-violent relationships between men and women. Fundamentally, preventing
domestic violence is about creating healthy families and healthy communities. Today’s violence
prevention interventions must aim to create a generation of children who come of age with new
skills to manage relationships and resolve conflict, expanded life opportunities, and different
expectations regarding gender roles and the sharing of power between men and women.
SIDEBAR BOXOne exercise used by the Indian NGO, Sakshi, to help domestic violence programs reorient their
activities more toward prevention is to envision an anti-violence agenda that evolves from the
needs of the daughters and sons of today’s battered women. Most domestic violence programs
design their interventions around assisting victims and transforming how social institutions—the
courts, health sector, the police—respond to victims and perpetrators. Instead, Sakshi encourages
groups to take children as their point of departure for planning. What interventions and programs
24
n
could be funded today that would help prevent future domestic violence? Such an exercise still
requires groups to consider the needs of victims because some children will nonetheless grow up
to experience violence. But this exercise helps redirect thinking toward primary prevention and
fundamental social change (Naina Kapur, personal communication, January, 2000)
END SIDE BARPrograms should place greater emphasis on equipping family, friends and faith communities
to respond constructively to issues of domestic violence.
Since many women will never access “official” services or systems, working to expand informal
sources of support through neighborhood and friendship networks, faith communities and
workplaces is highly important Most abused women reach out first to family members or
friends, not formal institutions (80, 128, 175, 176). How these individuals respond is highly
predictive of whether a woman continues toward empowerment and action or whether she
retreats once again into isolation and self-blame (175).
Whereas anti-violence efforts have focused considerable attention on reforming the response of
“formal” institutions such as the police, far less attention has been directed toward changing the
attitudes and response of trusted individuals on the “frontline.” There is much room for creative
programs aimed at combating harmful social norms that keep women trapped in abusive
relationships, and to model more constructive responses to abuse on the part of family and
friends. An innovative program in Ixtacalco, Mexico, for example, used community events,
small-scale media and 12-session workshops, to help women identify and name the abuse in their
lives and to model for friends and family members how best to respond to her situation. The
program worked specifically to counter victim-blaming attitudes and to provide concrete
examples of more constructive responses (177).
Domestic violence programs should make common cause with other programs aimed at
preventing youth violence, teen pregnancy, substance abuse, andjuvenile delinquency.
Evidence from the industrial world suggests that there is considerable overlap between the factors
that increase risk of a variety of problematic behaviors (178). There also appears to be
substantial continuity between aggressive behavior exhibited in childhood and various problem
behaviors in youth and early adulthood [ref]. Increasingly, the prevention insights from a variety
of problem areas overlap. All point to the need to intervene early with high risk families to
provide needed support, guidance, mentoring and services before a pattern of dysfunctional
parenting, harsh punishment, and/or maltreatment set the stage for abusive behavior in
adolescence or adulthood.
Regrettably, there is little overlap in the programmatic or research agendas of programs dealing
with youth violence, child abuse, substance abuse or domestic violence—despite the fact these
problems regularly co-occur in families. Although not all violent men come from troubled
families, if the violence against women movement is to get serious about prevention it must begin
to focus its attention on children—especially young children from troubled homes, and their
parents.
25
Domestic violence advocates have traditionally been wary of shifting attention to children for
fear that the all-important task of empowering women will be lost. Rather than adopt the
responsibility of working directly with children and youth, domestic violence advocates and
researchers should work with programs focusing on early childhood development, fatherhood,
home visitation, and youth violence prevention to integrate concern for domestic violence. Do
these programs address issues of gender role socialization? Do they explicitly address marital
conflict and violence? Are they in touch with local domestic violence programs and resources?
Those working directly with victims can also contribute by intensifying their work with the
children of abused women.
26
% of Adult Women Physically
Table 1
Sample
Physical Assault on
Women by an
Intimate Male
Partner
Selected PopulationBased Studies,
1982-1999
Region, Place & Year of Field
Work (Ref. No.)
Size
PopuLation*
Age
ARFICA. SUB-SAHARAN
Meskanena Woreda
673
2
15+
Kenya 1984-87 (362) ....
Kisii District
612
7
15+
Nigeria 1993P(331)
Not stated
Eastern Cape
Mpumalanga
Northern Province
1.000
396
418
465
3
3
3
18-49
18-49
18—49
11’
12’
5’
South Africa 1998 (281)....
National
5,077
2
15-49
6
Uganda 1995-96 (33)
Lira & Masaka Districts
1,660
2
20-44
Midlands Province
966
1
18+
South Africa 1998 (235).
"P” after year indicates the year of
Zimbabwe 1996 (464) ......
ASIA & PACIFIC
reporting the field work dates.
Australia 1996 (490)
National
6,300
1
Bangladesh 1992 (407)
National (villages)
1,225
2
3.611
2
Nasimagar Thana
Bangladesh 1993-95 (422).
15-65
9,938
3
15-49
14“
National
707
2
20*
38/12'
New Zealand 1994 (272)
National
2,000
6
17+
21’
Papua N. Guin. 1982 (437) ....
National, rural (villages)
628
Papua N. Guin. 1984 (366) ....
Port Moresby (low income)
298
Philippines 1993 (323)
National
8,481
5
15-49
10“
Philippines 1998 (57)
Cagayan de Oro City &
Bukidnon Province
1,660
2
15-49
26
Thailand 1994 (215)
Bangkok
619
4
India 1993-94 (233)
Tamil Nadu
Uttar Pradesh
859
2
983
India 1995-96(288)
Uttar Pradesh. 5 dist.
6.695
India 1999 (496)
6 states
Korea, Rep. of 1989 (253)
own use of violence against
spouse
women
of violence against partners
7 - married women: half with
47
32
4
2 * currently married/partnered
6 « all men reporting on own use
8'
19
2
3
outcome
3'
<50
37
2
1,374
5 ■ women with a pregnancy
17'
15-39
15-39
10,368
Phnom Penh & 6 prov.
4 ■ married men reporting on
13
41
42“
Jessore & Sirajgonj (rural)
Cambodia 1996P (325)
women
31'
20’
29’
20’
15-49
Bangladesh 1993 (255)
1 * all women
3 - ever-married/partnered
45
42
1
publication for studies not
'Population of respondents:
IO”
Ethiopia 1995 (110)
Percentages rounded to whole
numbers
Coverage
Assaulted by an Intimate Partner
In Cur
Ever (in
In Pre
Any Rela
vious 12
rent Re
lationship
tionship)
Months
16
45
30
40/26'
35’
67
56
20
pregnancy outcome, half
EUROPE
without
Moldova 1997(410)
National
4.790
3
15-44
14+
Netherlands 1986 (383)
National
989
1
20-60
21/1 r’
Norway 1989P (403)
Switzerland 1994-96 (178).
Trondheim
111
3
20-49
National
1.500
2
20-60
'Sample group included women
Turkey 1998 (223)
E and SE Anatolia
599
1
14-75
who had never been in a rela
United Kingdom 1993P (308)..
North London
430
1
16+
tionship and therefore were not in
LATIN AMERICA & CARIBBEAN
exposed group.
Antiquea 1990 (200)
National
97
1
29-45
30‘
‘Rate of partner abuse among
Barbados 1990 ( 494)
National
264
1
20-45
30m
"Nonrandom sampling tech
niques used.
ever-married/ partnered women,
Bolivia 1998 (338)
3 districts
recalculated from author's data.
Chile 1993P (268)
Metro. Santiago & prov.
' Although sample includes all
Chile 1997(312)
289
1
20*
1,000
2
22-55
Santiago
310
2
15-49
6,097
2 .
15-49
650
3
1,064
3
15+
18
21’
6’
58“
30“
12‘
17*
26/11'
23
women, rate of abuse is shown for
Colombia 1995 (337)
National
ever-married/partnered wo- men
Mexico 1996 (363)
Metro. Guadalajara
(N not given).
Mexico 1996P (191)
Monterrey
‘Perpetrator could be family
Nicaragua 1995 (130)
Nicaragua 1995 (163. 312)
LeOn
360
3
15-49
27/20'
member or close friend.
Managua
378
3
15-49
33/28'
'Severe abuse
69
Nicaragua 1998 (386)
National
8,507
3
15-49
12/tf
28/21'
'Any physical abuse/severe physical
Paraguay 1995-96 (105)....
Nat'l, except Chaco reg.
5,940
3
15-49
abuse only
Peru 1997 (188)
Metro. Lima (middle and
low income)
359
2
17-55
Puerto Rico 199&-96 (105)...
National
4.755
3
Uruguay 1997 (440)
Montevideo & Canelones
’Physical or sexual assault
‘In past 3 months
545
19
27
15
17’
10
31
13“
15-49
22-55
52/37'
10’
NEAR EAS & NORTH AFIRCA
34’
Egypt 1995-96(132)
National
7,121
3
15-49
16“
Israel 1994 (197)
West Bank & Gaza Strip
(Palestinians)
2,410
2
17-65
52/37'
Israel 1997P (196) ................
Arab, except Bedouin
1,826
2
19-67
32
3M
Compiled by the Center for
NORTH AMERICA
Health and Gender Equity
Canada 1993 (378)
National
12.300
1
18+
(CHANGE) for Population Reports
Canada 1991-92 (367)
Toronto
National
420
1
18-64
United States 1995- 96 (436)...
8,000
18+
29M
zr
13*
22'
POPULATION REPORTS
Box 1
Enhancing Comparability of Data on Domestic Violence
Mary Ellsberg
Most international prevalence figures on violence are not comparable due to inconsistencies in
the way that violence is conceptualized and measured. Prevalence is defined as the proportion of
abused women in a given study population during a specific period of time. Therefore, how to
define and measure "abuse" and how to determine the study population are two important
methodological challenges facing researchers on violence.
These issues have been addressed in a great variety of ways, with little consensus as to the most
appropriate method. A further complication is presented by the recognition that what we are
measuring is not the actual number of women who have been abused, but rather, the number of
women who are willing to disclose abuse. Therefore, there is always the potential for bias from
either over-reporting or under-reporting.
Finally, many researchers have pointed out that research on violence involves a number of
inherent risks to both respondents and interviewers, and that addressing these concerns is
essential, both for ethical reasons, as well as for ensuring data quality . However, the degree to
which these issues have been incorporated into study design and implementation varies a great
deal. Following are some of the greatest challenges to comparability between studies.
Selection of study participants. There is great variation in the study populations used for
domestic violence research. Many studies include all women within a specific age range
(frequently 15-49 or over 18), while other studies interview only women who are currently
married or have been married at some point in their lives (Table 1). Because both age and marital
status are associated with a woman’s risk of suffering partner abuse, the selection of eligible
participants can have a great impact on the estimates of prevalence of abuse in a population.
Table 1. Study populations from recent surveys on violence against women.
Country______
Cambodia (44)
Canada (179)
Chile (45)
Colombia (180)
Egypt (26)
Philippines (181)
Uganda (182)
Zimbabwe (183)
Study Population_________________________
Women and men aged 15-49
Women aged 18 or older
Women aged 22-55 married or partnered for more
than 2 years
Currently married women aged 15-49
Ever married women aged 15-49
Women aged 15-49 with a pregnancy outcome
Women 20-44 and their spouses/partners
Women 18 years and older
27
Definitions of violence. A further complication in the comparison of violence prevalence is the
use of inconsistent definitions of abuse. For example, some studies present only figures for
violent acts occurring in the last 12 months, whereas others measure lifetime experiences of
violence. In addition, not all studies separate different kinds of violence, so that it is not possible
to distinguish between acts of physical, sexual and emotional violence or between violence
committed by different perpetrators (184).
Enhancing disclosure. All studies on sensitive subjects, such as violence, face the challenge of
how to get people to talk openly about intimate aspects of their lives. The degree to which this is
achieved depends partly on methodological issues such as whether questions are clearly worded
and easy to understand, and how many times women are asked about violence. Another major
issue influencing disclosure is how comfortable women are made to feel during the interview.
This may be affected by many factors including the sex of the interviewer, the length of the
interview, whether others are present, and whether the interviewer appears to be genuinely
interested in her story and willing to listen without making judgments.
Over-reporting, or the fabrication of acts of violence that have not actually occurred, is generally
felt to be rare in violence research (184-187). Under reporting of violence, on the other hand, is
widely considered to be a much more common threat to validity.
Researchers on violence, and particularly feminist researchers, have proposed a series of
strategies to enhance disclosure. Two important strategies are giving several opportunities to
disclose violence within the interview, and using behaviorally specific questions, rather than
asking women more general and subjective questions, such as, "have you ever been abused?" By
focusing on acts rather than subjective interpretations women are not forced to identify with
stigmatized categories such as "battered woman" or "rape victim". Providing multiple
opportunities to disclose allows women more time to think about their answers, to recall events
that may have happened long ago or in different contexts, and to build up enough trust to talk
about violence (185). Another key strategy lies in the selection and training of interviewers who
are skilled in developing rapport with respondents (188).
Strategies to improve the quality of data on violence must take into account concerns for the
safety of both respondents and interviewers throughout the research process (189-192).
Disclosing violence may expose a respondent to the risk of retaliation by an abusive partner or
family members. It may also be emotionally distressing for her to recall past events without
adequate support. The World Health Organization recently published guidelines for addressing
ethical and safety issues in violence research. The recommendations urge researchers only to
undertake studies on violence if they are able to ensure minimal safety standards, such as
ensuring complete privacy during the interview, providing information and referrals for
respondents, and special training and support for interviewers. The WHO guidelines argue that
not only are these considerations essential for ethical reasons, but also that they are critical to
ensuring data quality, primarily because of their impact on women’s disclosure (192).
As more international data on violence against women become available, two distinct research
trends have emerged. Several national studies have produced prevalence estimates on violence.
Most of these studies, with a few important exceptions, such as the Canadian and US National
Surveys on Violence against Women (5, 179), have been primarily designed for other purposes.
28
3
For example, recent Demographic and Health Surveys' and Reproductive Health Surveys have
included a limited number of questions on violence in national surveys in Egypt (26), South
Africa (193), Puerto Rico (194), Paraguay (195), Moldova (46), Philippines (181) and Colombia
(180). These surveys typically use aggregate "gateway" questions, such as "have you ever been
beaten by anyone since you were 15/were married? By whom?"
The other trend is represented by smaller, in-depth studies providing more detailed information
on women’s experiences of violence (12, 44, 80, 187, 196, 197). Although these studies cover a
limited geographical region, they tend to place more emphasis on the interaction between
interviewers and respondents, and to be more cognizant of safety issues.
There are many potential advantages to including violence questions in national surveys designed
primarily for other purposes. In many cases, national statistical bureaus conduct the studies, thus
the results are given the legitimacy of "official statistics." This can be very useful for the
purposes of advocacy. Nationally representative data are useful for local program planning, and
also permit in-depth analysis of variation between regions. Finally, the large data sets generated
by these studies, including many other reproductive and child health outcomes, can be used to
deepen understanding of risk factors and health consequences of violence.
However, there are also potential drawbacks to this strategy. In general, prevalence estimates
have been higher in the smaller, more focused studies than in the national surveys designed
primarily for other purposes. One explanation for this may be that the focused studies are able to
produce more accurate prevalence estimates due to the use of methods for enhancing disclosure.
Therefore, one tradeoff of using multi-faceted surveys for producing prevalence estimates on
violence is the risk of significant under-reporting. Under reporting of violence will dilute
associations between potential risk factors and health outcomes, leading to falsely negative
results. Underestimating the dimensions of violence may also result in violence intervention
programs not receiving the priority they deserve in the allocation of resources. Finally, if safety
concerns are not systematically addressed, women may be placed at risk of retaliation or other
harm as a result of their participation in the study.
29
Box 2
”La Ruta Critica:"
An Evaluation of Institutional Responses to Domestic Violence
Pan American Health Organization
In 1995, PAHO began the implementation of a community diagnostic study to document what
happens when a woman affected by family violence decides to break the silence and seek
assistance in ending the abuse. The study was undertaken in ten Latin American countries1. The
Spanish term coined for this process is La Ruta Critica, which refers to "the sequence of
decisions made and actions taken by a woman in order to confront the violent situation she faces
(or has faced) and the responses she encounters in her search for help.” It is an iterative process,
where internal and external motivating factors influence the actions undertaken by women
affected by violence. These actions provoke responses from various social actors, including
service providers and community members; these responses, whether positive or negative,
intended or unanticipated, in turn affect the motivating factors for women.
In effect the study asks, "What happens when a woman decides to seek help? Who does she go
to? What factors motivate her to or inhibit her from action? What kinds of attitudes and
responses does she encounter from institutional actors? Do service provider attitudes and
responses reflect prevailing community norms?"
Figure 1: “Ruta Critica” Conceptual Framework
Motivating Factors
Information & knowledge
Perceptions & attitudes
Previous experiences
Support from others
Decisions made
Responses Factors
Access, availability
and quality of
services
Social
representations of
service providers
Results obtained
Actions
Undertaken
Search for
solutions
Decisions carried
out
1 Participating countries include Belize, Bolivia, Costa Rica, Ecuador, El Salvador, Guatemala,
Honduras, Nicaragua, Panama and Peru.
Qualitative methods were chosen to provide in-depth understanding of women's motivations and
service providers' perceptions and attitudes; four sources were chosen to allow triangulation of
information and maximize credibility and trustworthiness of the findings
All together, researchers conducted over 500 in-depth interviews with battered women,
interviewed over 1000 service providers and completed approximately 50 focus group sessions.
Results show that there are many factors, internal and external, that impact on a woman’s
decision to take action to stop the violence. This is often a long process. In some cases it takes
many years and several attempts at seeking help from several sources. Rarely is there a single
event that precipitates action. Findings suggest that many battered women are resourceful in
seeking help and finding ways of mitigating the violence. There are formidable obstacles to
ending domestic abuse in the home, and some of these are summarized below.
What factors drive battered women to search for help?
Battered women identified several factors that act as catalysts for action. An increase in the
severity or frequency of the violence may trigger a recognition that the abuser is not going to
change. An event may make it clear to her that she cannot modify the situation with her own
internal resources. A primary motivating factor is the realization that her own life, or those of
her children, are in danger.
“He mistreated the children badly. He only knew how to shout orders. There was a
period when he would beat them. The children, particularly my oldest son, had become
very disobedient, rebellious and had lost all motivation to study. He didn 7 go to
school... ”
(tIfinally decided to leave when he burned all my clothes and also burned me. ”
“The moment cam when I said to myself that I had to find someone to help me because it
was not possible to keep going on in this way. I had become hysterical, problematic,
unhappy, mainly because I could see my beaten face eveiy week in the mirror. ”
What factors inhibit the process of seeking help?
As with the precipitating factors, the factors that inhibit help-seeking are multiple and
intertwined.
“One learns to live with the person even though he is an abuser. I don’t know, for me he
was my companion because Ifelt alone, without the support ofa family. He was my
family... ”
“I used to excuse him for that and I believed that though the love I hadfor him he was
going to get better, and that this was not going to keep on.
However, economic factors appear to weigh more heavily than do emotional considerations:
“The children were very young and I didn 't think I could support them on my own. And I
didn’t want to burden my mother. ”
These barriers are reinforced by battered women’s feelings of guilt, self-blame, or abnormality.
“I tried to reflect on my own actions. What did I do to provoke him? I considered my
personality... "
“My mother would tell me that I was crazy and that is why I was seeing a psychologist
and my brothers and sisters said the same thing. ”
“There came a moment in which I really thought ‘Am I crazy?9 Then I sought help to
make sure that what was happening to me was true. ”
But across the board the greatest inhibiting factor was fear.
“He 'd threaten me... He’d say ‘Ifyou tell anyone, I’ll kill you.
“Because one of the things one has to put up with in this type ofrelationship is that he
will threaten that ifyou tell, what will happen is worse than you could ever imagine. So
that is why I never told anybody. ”
“He would tell me that he was going to burn the house and he’d grab knives...Just a
short while ago he told me that he was going to poison myfood when I wasn’t looking. n
“He took out a knife that he carried around and he told me. ‘Ifyou don’t come back
home I will kill you becazise you are going to be mine or nobody else’s. I became very
frightened and tried to calm him and I told him that ofcourse Td come back and we will
still be together. ”
“In front ofme he killed my little cat that I loved, and he told me that ifI betrayed him
that is what he would do to me.,9
Women who initiated the “Ruta Critica” rarely began with formal health or police services.
They initially relied instead on support from other women in the community, including female
family members, neighbors, and health promoters.
“Ifinally told a friend that I trusted. I went to tell her because she is an older lady and
she told me that he was wrong, that he was a sadist. She told me that I should get out of
the house. ”
“ ...Almost every time he abused me I would go to my friend s house and she would give
me a place to sleep. I would even sleep on the floor, because she was poor. She was the
one who finally said, This is too much. I am going to help you find help because that
man is abusing you too much!”
Many women brought up the rampant corruption and gender-based stereotypes that exist within
the judicial and police systems.
“They made fun of me. [The police] would laugh in my face and [my husband] would
say, ‘Go ahead, accuse me. They are all my friends, the judge is myfriend, the police are
myfriends... and see? What have you got by making all this fuss? Nothing. I am still
here. What have you gained? Nothing. ’ Everything is a joke. ”
Box 3
Examples of Innovative Prevention Programs
In 1996, the Parenting for Peace and Justice Network and 18 other U.S.-based NGOs launched
the “Family Pledge of Nonviolence,” a long-term campaign to offer families and communities
alternatives to violence. Members are asked to sign a pledge that says in part: “Making peace
must start within ourselves and in our family. Each of us, members of the
family,
commit ourselves as best we can to become non-violent and peaceable people. (Magazine of the
Women International League for Peace and Freedom, May/June 1998, p. 19)
Education Wife Assault in Toronto, Ontario works with immigrant and refugee women to help
them develop culturally appropriate violence prevention campaigns for their community. EWA
holds ’’Skill Shops" that give women leaders the background and skills they need to develop their
own culturally specific programs against domestic violence. The immigrant women then carry out
the campaigns with the technical support of EWA. EWA staff also lends emotional support to
women organizers to help them overcome the isolation and "backlash" often directed at women
working against domestic violence who are frequently perceived as threatening to community and
cultural cohesiveness (Center for Women's Global Leadership, 1992).
In South Africa the Planned Parenthood Association of South Africa (PPASA), an affiliate of
the International Planned Parenthood Federation (IPPF), together with AVSC International,
developed the Men as Partners Program to address gender violence. The program integrates
participatory exercises on gender, sexual power, and intimate relationships into PPSA’s “life
skills” workshops. The program began after a survey of 2,000 South African men found that
58% believed that the concept of rape did not apply to a husband forcing his wife to have sex
against her will, 48% thought the way a woman dressed caused her to be raped, and 22%
approved of a man hitting his partner compared with 5% who approved of a woman hitting her
partner) [AVSC International, 1999 #2049].
In Mexico, the Instituto Mexicano de Investigacion de Familia y Poblacion A.C. (IMIFAP),
a nongovernmental organization, has developed an experiential workshop for adolescents to help
prevent violence in dating and friendship relationships. The workshop, entitled, “Rostros y
Mascaras de la Violencia (Faces and Masks of Violence),” uses participatory techniques to help
youth explore expectations and feelings about love, sex, and romance, to distinguish between
romantic and controlling behaviors, and to understand how traditional gender roles inhibit both
male and female behavior [Fawcett, 1999 #2092].
In Brooklyn, New York, the Anti-Violence Education Project uses self defense training as an entre
to discuss violence prevention with children in the public schools. The project holds weekly
sessions to teach children self defense and non-violent ways to resolve conflict. It draws analogies
between relationship strategies and the philosophy of karate (The project teaches that the marshal
arts do not condone violence; rather, the true master is the one who can use the least force to achieve
his/her ends). It also teaches children to look critically at how the media misrepresents the marshal
arts through its depiction of pseudo-heros such as Bruce Lee (Ellman, 1993).
In Trinidad and Tobago, the NGO SERVOL (Service Volunteered for All), conducts 14 week
adolescent development workshops aimed at helping youth develop healthy relationships and
parenting skills. The first half of the program is devoted to helping adolescents understand how
they became who they are through the way they were parented and the second half helps them
discover how not to repeat the mistakes while forming their own families. In effect, this course is
on “emotional intelligence,” learning to distinguish emotions and how to handle them. The
students leam about the psychic impact of trauma and how it can set up destructive behaviors
later in life.
Box 4
How Women Experience Violence:
Women’s Voices from North and South
Expressions of violence against women may vary from one country to the next, and there are
specific culture bound expressions that are unique to certain regions, such as bride burning or
acid throwing in South Asia. However, the words which women use to describe the feelings of
shame, fear and powerlessness that they experience are strikingly similar and indicate that to a
large degree these feelings transcend cultural boundaries.
"He used to tell me, "you ’re an animal, an idiot, you are worthless. ” That made me feel even
more stupid. I couldn’t raise my head. I think I still have scars from this, and I have always
been insecure...! accepted it, because after a point he had destroyed me by blows and
psychologically. "
young woman from Nicaragua
Always at night I remember. When I turn over my back hurts, my arms hurt, and then I say to
myself, 11How could I have thought that my husband would be my companion, my support, my
protection. He has been my executioner. During the eleven years I have lived with him, I have
not had a single happy day."
young woman from Cuzco, Peru
(,I told my sisters-in-law that my husband had forced me into sex, but they told me that this is
part oflife."
22-year-old woman from Zimbabwe
“I was afraid people were going to treat me bad for staying there as long as I did. I took it
personal; I thought I deserved it for letting this go on. ” Middle-aged housewife, United States
"Sometimes, he would want to do it, even though I didn 7 feel like it. I would tell him sometimes
that I did not want it, and that he came near to me only to have sex. Then he would get very
angry and beat me and say that I did not like him because I was having an affair. "
42 year old married woman from Bombay, India
"I went to get pills from the health center, without [my husband’s] knowledge. He knew nothing,
but found where I had hidden the pills, and burnt them. He said that I was taking them because I
had a lover. But I was doing it to prevent hardship. I could not continue taking the pills because
he beat me. He beat me more because he said I had a lover."
Woman from peri-urban center of Guatemala City
“My husband is a great drunkard and whenever he is drunk, I used to say no to him when he
asked me for sex but her used to beat me up andforced me to have sex with him. Whether I like it
or not, I have to bow down to him and have sex. " 34-year-old woman from Papua New Guinea
References
1.
Crowell N , Burgess AW. Understanding Violence Against Women. Washington D.C.,
National Academy Press. 1996.
2.
Heise L, Pitanguy J , Germain A. Violence against women: The hidden health burden .
World Bank Discussion Paper #255: Washington D.C., 1994,. p. 72.
3.
Koss MP et al. No safe haven: Male violence against women at home, at work, and in the
community . American Psychological Association: Washington D.C., 1994.
4.
Butchart A , Brown D. Non-fatal injuries due to interpersonal violence in JohannesburgSoweto: Incidence, determinants and consequences. Forensic Science International,
1991,52:35-51.
5.
Tjaden P , Thoennes N. Prevalence, incidence and consequences of violence against
women: Findings from the national violence against women survey. In Research in Brief.
National Institute of Justice, Centers for Disease Control and Prevention: Washington,
D.C., 1998,. p. 16.
6.
Council of Scientific Affairs. Violence Toward Men: Fact or Fiction . American Medical
Association, 1994.
7.
Yoshihama M , Sorenson SB. Physical, sexual, and emotional abuse by male intimates:
Experiences of women in Japan Violence and Victims, 1994,9:63-77.
8.
Ellsberg MC et al. Candies in hell: Women's experience of violence in Nicaragua Social
Science and Medicine, In press.
9.
Leibrich J, Paulin J , Ransom R. Hitting home: Men speak about domestic abuse of
women partners. New Zealand Department of Justice in association with AGB McNair:
Wellington, New Zealand, 1995,. p. 243.
10.
Granados Shiroma M. Salud reproductiva y violencia contra la mujer: Un analisis desde la
perspectiva de genero. [SPA] [Reproductive health and violence against women: A
gender perspective] . Asociacion Mexicana de Poblacion, Consejo Estatal de Poblacion,
Nuevo Leon, El Colegio de Mexico, 1996,. p. 42.
11.
Ellsberg MC et al. Wife abuse among women of childbearing age in Nicaragua American
Journal ofPublic Health, 1999,89:241-4.
12.
Mooney J. The Hidden Figure: Domestic Violence in North London. London, Middlesex
University. 1993.
13.
Ellsberg M, Heise L , Shrader E. Researching violence against women: A practical guide
for researchers and advocates . Center for Health and Gender Equity: Washington D.C.,
1999. p. 154.
30
14.
Rodgers K. Wife assault: The findings of a national survey. Juristat service bulletin of the
Canadian Centre for Justice Statistics, 1994,14:1-22.
15.
Cabaraban M , Morales B. Social and economic consequences for family planning use in
southern Philippines. Research Institute for Mindanao Culture, Xavier University:
Cagayan e Oro City, Philippines, 1998. p. 206.
16.
Cabrejos MEB et al. Los caminos de las mujeres que rompieron el silencio: Un estudio
cualitativo sobre la ruta critica que siguen las mujeres afectadas por la violencia
intrafamiliar. [SPA] [Paths of women who broke the silence: A qualitative study of help
seeking by women affected by family violence] in Proyecto Violencia Contra las Mujeres
y las Ninas. OPS, World Health Organization: Lima, Peru, 1998,. p. 24.
17.
Schuler SR et al. Credit programs, patriarchy and men's violence against women in rural
Bangladesh Social Science and Medicine, 1996,43:1729-42.
18.
Zimmerman K. Plates in a basket will rattle: Domestic violence in Cambodia: A
summary. Project Against Domestic Violence: Phnom Penh, Cambodia, 1995. p. 263.
19.
Michau L. Community-based research for social change in Mwanza, Tanzania, in Third
Annual Meeting of the International Research Network on Violence Against Women.
1998. Washington D.C.: CHANGE.
20.
Armstrong A. Culture and choice: Lessons from survivors of gender violence in
Zimbabwe . Violence Against Women in Zimbabwe Research Project: Harare,
Zimbabwe, 1998,. p. 149.
21.
Gonzalez Montes S. Domestic violence in Cuetzalan, Mexico: Some research questions
and results, in Third Annual Meeting of the International Research Network on Violence
Against Women. 1998: CHANGE.
22.
Osakue G , Hilber AM. Women's sexuality and fertility in Nigeria. In: Petchesky R, Judd
K, eds. Negotiating Reproductive Rights. London, Zed Books, Ltd, 1998: p. 180-216.
23.
Hassan Y. The haven becomes hell: A study of domestic violence in Pakistan. In: Cantt L,
ed. Pakistan: Women Living Under Muslim Laws 1995: p. 72.
24.
Bradley CS. Attitudes and practices relating to marital violence among the Tolai of East
New Britain. Domestic Violence in Papua New Guinea. Boroko, Papua New Guinea,
Papua New Guinea Law Reform Commission, 1985: p. 32-71.
25.
Jejeebhoy SJ. Wife-beating in rural India: A husband’s right? Economic and Political
Weekly (India), 1998,23:588-862.
26.
El-Zanaty F et al. Egypt demographic and health survey 1995. Macro International:
Calverton, Maryland, 1996. p. 348.
31
'Ll.
Rosales J et al. Encuesta Nicaraguense de demografia y salud, 1998 [SPA] [1998
Nicaraguan demographic and health survey] . Institute National de Estadisticas y Censos:
Managua, Nicaragua, 1999. p. 319.
28.
David F, Chin F. Economic and psychosocial influences of family planning on the lives of
women in Western Visayas. Central Philippines University and Family Health
International: Iloilo City, Philippines, 1998. p. 85.
29.
Bawah AA et al. Women's fears and men's anxieties: The impact of family planning on
gender relations in northern Ghana Studies in Family Planning, 1999,30:54-66.
30.
Wood K, Jewkes R. Violence, rape, and sexual coercion: Everyday love in South Africa
Gender and Development, 1997,5:23-30.
31.
Khan ME et al. Sexual violence within marriage. Seminar. New Dehli, Population
Council, 1996: p. 32-5.
32.
National Sex and Reproduction Research Team , Jenkins C. National study of sexual and
reproductive knowledge and behaviour in Papua New Guinea . PNG Institute of Medical
Research: Goroka, Papua New Guinea (PNG), 1994. p. 147.
33.
Heise L. Violence against women: An integrated, ecological framework Violence Against
Women, 1998,4:262-90.
34.
Rao V. Wife-beating in rural South India: A qualitative and econometric analysis. Social
Science and Medicine, 1997,44:1169-79.
35.
Johnson H. Dangerous Domains: Violence Against Women in Canada. Crime in Canada.
Ontario, Canada, International Thomson Publishing. 1996.
36.
Romero M. Violencia sexual y domestica: Informe de la fase cuantitativa realizada en el
centro de atencion a adolescentes de San Miguel de Allende. [SPA] [Sexual and domestic
violence: Report from the qualitative phase from an adolescent center in San Miguel de
Allende]. Population Council: Mexico City, 1994.
37.
Campbell J et al. Voices of strength and resistance: A contextual and longitudinal analysis
of women's responses to battering. Journal ofInterpersonal Violence, 1999,13:743-62.
38.
Dutton MA. Battered women's strategic response to violence: The role of context. In:
Edelson JL, Eisikovits ZC, Editors. Future Interventions with Battered Women and Their
Families. London, Sage, 1996: p. 105-24.
39.
Hartigan P. Summary offindings: The critical path women take to find a solution to
domestic violence, in Role of the Health Sector in Violence Against Women. 1997.
Copenhagen, Denmark.
32
40.
O'Conner M. Making the links: Towards an integrated strategy for the elimination of
violence against women in intimate relationships with men . Women's AID: Dublin,
Ireland, 1995,. p. 153.
41.
Short L. Survivor's identification ofprotective factors and early warning signs in intimate
partner violence. In: Third Annual Meeting of the International Research Network on
Violence Against Women. 1998. Washington D.C.: CHANGE.
42.
George A. Differential perspectives of men and women in Mumbai, India on sexual
relations and negotiations within marriage. Reproductive Health Matters, 1998,6:87-95.
43.
Koenig M et al. Individual and community-level determinants of domestic violence in
rural Bangladesh in Hopkins Population Center Paper on Population. Johns Hopkins
School Public Health, Department of Population and Family Health Sciences: Baltimore,
1999. p. 32.
44.
Nelson E , Zimmerman C. Household survey on domestic violence in Cambodia .
Ministry of Women's Affairs and Project Against Domestic Violence: Phnom Penh,
Cambodia, 1996. p. 82.
45.
Larrain SH. Violencia Puertas Adentro: La Mujer Golpeada. [SPA] [Violence Behind
Closed Doors: The Battered Women]. Santiago, Chile, Editorial Universitaria. 1994.
46.
Serbanescu F et al. Reproductive health survey, Moldova, 1997 . Centers for Disease
Control, U.S. Department of Health and Human Services; Atlanta, 1998. p. 338.
47.
Ellsberg M et al. Women's strategic responses to violence in Nicaragua. Journal of
Epidemiology and Community Health, In Press.
48.
Bunge VP , Levett A. Family violence in Canada: A statistical profile . Statistics Canada:
Ottawa, Canada, 1998. p. 38.
49.
Campbell JC , Soeken KL. Women’s responses to battering: A test of the model. Research
in Nursing and Health, 1999,22:49-58.
50.
Campbell JC. Abuse during pregnancy: Progress, policy, and potential. American Journal
ofPublic Health, 1998,88:185-7.
51.
Landenburger KM. The dynamics of leaving and recovering from an abusive relationship.
Journal of Obstetric, Gynecologic, and Neonatal Nursing, 1998,27:700-6.
52.
Jacobson NS et al. Psychological factors in the longitudinal course of battering: When do
the couples split up? When does the abuse decrease? Violence and Victims, 1996,11:37192.
53.
Campbell J. Assessing Dangerousness: Violence by Sexual Offenders, Batterers, and
Child Abusers. Thousand Oaks, California, Sage Publications. 1995.
33
54.
Counts DA, Brown J, Campbell J. Sanctions and Sanctuary: Cultural Perspectives on the
Beating of Wives. Boulder, Colorado, Westview Press. 1992.
55.
Levinson D. Violence in Cross Cultural Perspective. Newbury Park, California, Sage
Publishers. 1989.
56.
Narayana G. Family violence, sex and reproductive health behavior among men in Uttar
Pradesh, India. Unpublished, 1996. p. 11.
57.
Martin SL et al. Domestic violence in northern India American Journal of Epidemiology,
1999,150:417-26.
58.
Hoffman KL, Demo DH , Edwards JN. Physical wife abuse in a non-westem society: An
integrated theoretical approach Journal ofMarriage and the Family, 1994:131 -46.
59.
Gonzales de Olarte E , Gavilano Llosa P. Does poverty cause domestic violence? Some
answers from Lima. In: Morrison A, R., Biehl ML. Editors. Too Close to Home:
Domestic Violence in the Americas. Washington, D.C., Inter-American Development
Bank, 1999: p. 35-49.
60.
Straus M , et al. Societal change and change in family violence from 1975 to 1985 as
revealed by two national surveys. Journal ofMarriage and the Family, 1986:465-79.
61.
Dutton DG. The Domestic Assault of Women: Psychological and Criminal Justice
Perspectives. Vancouver, British Colombia, University of British Colombia Press. 1995.
62.
Hoteling GT , Sugarman DB. An analysis of risk markers in husband to wife violence:
The current state of knowledge. Violence and Victims, 1986,1:101-24.
63.
Moreno Martin F. La violencia en la pareja. [SPA] [Intimate partner violence]. Revista
Panamericana de Salud Publica, 1999,5:245-58.
64.
Oropesa RS. Development and marital power in Mexico. Social Forces, 1997,75:1291317.
65.
Bertrand JT, Ward V , Pauc F. Sexual practices among the Quiche-speaking Mayan
population of Guatemala International Quarterly Communication Health Education,
1992,12:265-82.
66.
Parry C et al. Alcohol attributable fractions for trauma in South Africa Curationis,
1996,19:2-5.
67.
Kyriacou DN et al. Emergency department-based study of risk factors for acute injury
from domestic violence against women Annals of Emergency Medicine, 1998,31:502-6.
68.
McCauley J et al. The "battering syndrome": Prevalence and clinical characteristics of
domestic violence in primary health care internal medicine practices. Annals of Internal
Medicine, 1995,123:737-46.
34
69.
Hotaling GT , Sugarman DB. A risk marker analysis of assaulted wives. Journal of
Family Violence, 1990,5:1-13.
70.
Fournier M et al. Nada en espanol aqui. [SPA] [Multicenter study on cultural attitudes and
norms towards violence (ACTFVA project): Methodology]. Revista Panamericana de
Salud Publica, 1999,5:222-31.
71.
Sanday PR. The socio-cultural context of rape: A cross cultural study. Journal of Social
Issues, 1981,37:5-27.
72.
Orpinas P. Who is violent? Factors associated with aggressive behaviors in Latin America
and Spain. Pan American Journal ofPublic Health, 1999,5:232-43.
73.
Golding JM. Sexual assault history and limitations in physical functioning in two general
population samples. Research in Nursing and Health, 1996,19:33-44.
74.
Leserman J et al. Sexual and physical abuse history in gastroenterology practice: How
types of abuse impact health status. Psychosomatic Medicine, 1996,58:4-15.
75.
Koss MP, Koss PG , Woodruff WJ. Deleterious effects of criminal victimization on
women's health and medical utilization. Archives ofInternal Medicine, 1991,151:342-7.
76.
Walker E et al. Adult health status of women HMO members with histories of childhood
abuse and neglect American Journal ofMedicine, 1999,107:332-9.
77.
McCauley J et al. Clinical characteristics of women with a history of childhood abuse:
Unhealed wounds. Journal of the American Medical Association, 1997,277:1362-8.
78.
Dickinson LM et al. Health-related quality of life and symptom profiles of female
survivors of sexual abuse. Archives ofFamily Medicine, 1999,8:35-43.
79.
Felitti VJ et al. Relationship of childhood abuse and household dysfunction to many of
the leading causes of death in adults. The Adverse Childhood Experiences (ACE) Study.
American Journal ofPreventive Medicine, 1998,14:245-58.
80.
Heise LL, Ellsberg M , Gottemoeller M. Ending Violence Against Women in Population
Reports, Series L Robey B Editor. Johns Hopkins University School of Public Health
Center for Communications Programs: Baltimore, MD, 1999,. p. 44.
81.
Heise L, Moore K, Toubia N. Sexual coercion and women’s reproductive health: A focus
on research . Population Council: New York, 1995. p. 59.
82.
Najera TP, Gutierrez M , Bailey P. Bolivia: Follow-up to the 1994 Demographic and
Health Survey, and women's economic activities, fertility and contraceptive use . Family
Health International: Research Triangle Park, North Carolina, 1998. p. 8.
83.
Ballard TJ et al. Violence during pregnancy: Measurement issues. American Journal of
Public Health, 1998,88:274-6.
35
84.
Campbell JC. Addressing battering during pregnancy: Reducing low birth weight and
ongoing abuse. Seminars in Perinatology, 1995,19:301-6.
85.
Curry MA, Perrin N , Wall E. Effects of abuse on maternal complications and birth
weight in adult and adolescent women. Obstetrics and Gynecology, 1998,92:530-4.
86.
Gazmararian JA et al. Prevalence of violence against pregnant women Journal of the
American Medical Association, 1996,275:1915-20.
87.
Newberger EH et al. Abuse of pregnant women and adverse birth outcome. Current
knowledge and implications for practice. Journal of the American Medical Association,
1992,267:2370-2.
88.
Bullock LF , McFarlane J. The birth-weight/battering connection American Journal oj
Nursing, 1989,89:1153-5.
89.
Parker B, McFarlane J , Soeken K. Abuse during pregnancy: Effects on maternal
complications and birth weight in adult and teenage women. Obstetrics and Gynecology,
1994,84:323-8.
90.
Valdez-Santiago R , Sanin-Aguirre LH. Domestic violence during pregnancy and its
relationship with birth weight SaludPublica Mexicana, 1996,38:352-62.
91.
Valladares E et al. Physical abuse during pregnancy: A risk factor for low birth weight,
submitted, p. 19.
92.
Ganatra BR, Coyaji KJ , Rao VN. Too far, too little, too late: A community-based case
control study of maternal mortality in rural west Maharashtra, India. Bulletin of the World
Health Organization, 1998,76:591-8.
93.
Felitti VJ. Long-term medical consequences of incest, rape, and molestation Southern
Medical Journal, 1991,84:328-31.
94.
Koss M. The impact of crime victimization of women's medical use. Journal of Women's
Health, 1993,2:67-72.
95.
Morrison AR , Orlando MB. Social and economic costs of domestic violence: Chile and
Nicaragua. In: Morrison AR, Biehl ML, Editors. Too Close to Home: Domestic Violence
in the Americas. Washington, D.C., Inter-American Development Bank, 1999: p. 51-80.
96.
Sansone RA, Wiederman MW , Sansone LA. Health care utilization and history of trauma
among women in a primary care setting Violence and Victims, 1997,12:165-72.
97.
IndiaSAFE Steering Committee. IndiaSAFE Final Report . International Center for
Research on Women: Washington, D.C., 1999.
98.
Browne A, Salomon A , Bassuk SS. The impact of recent partner violence on poor
women's capacity to maintain work Violence Against Women, 1999,5:393-426.
36
99.
Lloyd S , Taluc N. The effects of male violence on female employment Violence Against
Women, 1999,5:370-92.
100.
McCloskey LA, Figueredo AJ , Koss MP. The effects of systemic family violence on
children's mental health. Child Development, 1995,66:1239-61.
101.
Edleson JL. Children witnessing of adult domestic violence. Journal of Interpersonal
Violence, 1999,14.
102.
Jouriles EN, Murphy CM , O'Leary KD. Interspousal aggression, marital discord, and
child problems. Journal of Consulting and Clinical Psychology, 1989,57:453-5.
103.
Spaccarelli S, Coatsworth JD , Bowden BS. Exposure to serious family violence among
incarcerated boys: Its association with violent offending and potential mediating
variables. Violence and Victims, 1995,10:163-82.
104.
Song LY, Singer MI , Anglin TM. Violence exposure and emotional trauma as
contributors to adolescents' violent behaviors. Archives of Pediatrics and Adolescent
Medicine, 1998,152:531-6.
105.
Jaffe PG, Wolfe DA , Wilson SK. Children of Battered Women. Newbury Park,
California, Sage Publications. 1990.
106.
Jejeebhoy SJ. Associations between wife-beating and fetal and infant death: Impressions
from a survey in rural India Studies in Family Planning, 1998,29:300-8.
107.
Asling-Monemi K et al. Violence against women increases the risk of infant and child
mortality: A case-referent study in Nicaragua, submitted, p. 35.
108.
Chalk R , King PA, eds. Violence in Families: Assessing Prevention and Treatment
Programs.. 1998, National Academy Press: Washington D.C. 392.
109.
Plitcha SB. Domestic violence: Building paths for women to travel to freedom and safety.
in Symposium on domestic violence and women's health: Broadening the Conversation.
1995. New York: Commonwealth Fund.
110.
Ramos-Jimenez P. Philippine strategies to combat domestic violence against women .
Task Force on Social Science and Reproductive Health, Social Development Research
Center, and De La Salle University: Manila, 1996.
111.
World Health Organization (WHO). Violence against women legislation . WHO: Geneva,
1999.
112.
Mitra N. Best practices among response to domestic violence: A study of government and
non-govemment response in Madhya Pradesh and Maharashtra [draft] . International
Center for Research on Women: Washington D.C., 1998,. p. 252.
37
113.
Sherman LW , Berk RA. The specific deterrent effects of arrest for domestic assault
American Sociological Review, 1984,49:261-72.
114.
Gamer J, Fagan J , Maxwell C. Published findings from the spouse asault replication
program: A critical review. Journal of Quantitative Criminology, 1995,11:3-28.
115.
Fagen J , Browne A. Violence between spouses and intimates: Physical aggression
between women and men in intimate relationships. In: Council NR, Editor, ed.
Understanding and Preventing Violence. Washington D.C., National Academy Press,
1994: p. 115-292.
116.
Marciniak E. Community policing of domestic violence: Neighborhood differences in the
effect of arrest, in Institute of Criminal Justice and Criminology. University of Maryland:
College Park, 1994.
117.
Sherman LW. The influence of criminology on criminal law: Evaluating arrests for
misdemeanor domestic violence. Journal of Criminal Law and Criminology, 1992,83:145.
118.
Stark E. In defense of mandatory arrest: A reply to its critics. American Behavioral
Scientist, 1993,36:651-80.
119.
Larrain S. Curbing domestic violence: Two decades of activism. In Morrison AR, Biehl
ML, Editors. Too Close to Home: Domestic Violence in the Americas. Washington, D.C.,
Inter-American Development Bank, 1999: p. 105-30.
120.
Poonacha V , Pandey D. Response to domestic violence in Karnataka and Gujurat. In:
Duwury N, Editor, ed. Domestic Violence in India. Washington D.C., International
Center for Research on Women, 1999: p. 28-41.
121.
Estremadoyro J. Violencia en la Pareja: Comisarias de Mujeres en el Peru. [SPA]
[Partner Violence: Women's Police Stations in Peru]. Cuademos de Trabajo. Lima,
Ediciones Flora Tristan. 1993.
122.
Hautzinger S. Machos and policewomen, battered women and anti-victims: Combating
violence against women in Brazil in Department of Anthropology. Johns Hopkins
University: Baltimore, 1998. p. 455.
123.
Mesquita da Rocha M. Women's police stations: Rio de Janeiro, Brazil, ed. Domestic
Violence in Latin America and the Caribbean. Washington D.C., Inter-American
Development Bank, 1997:.
124.
Thomas DQ. In search of solutions: Women’s police stations in Brazil. , ed. Women and
Violence: Realities and Responses Worldwide. London, Zed Books Ltd., 1994: p. 32-43.
125.
Sugg NK et al. Domestic violence and primary care. Attitudes, practices, and beliefs.
Arch Fam Med, 1999,8:301-6.
38
126.
Caralis PV , Musialowski R. Women's experiences with domestic violence and their
attitudes and expectations regarding medical care of abuse victims. Southern Medical
Journal, 1997,90:1075-80.
127.
Friedman LS et al. Inquiry about victimization experiences. A survey of patient
preferences and physician practices. Archives ofInternal Medicine, 1992,152:1186-90.
128.
Pan-American Health Orgamzation. Ruta critica que siguen las mujeres victimas de
violencia intrafamiliar: Analisis y resultados de investigacion. [SPA] [Help-seeking by
victims of family violence: Analysis and research results] . Organizacion Panamericana
de la Salud en Panama: Panama City, Panama, 1998,. p. 94.
129.
Cohen S, De Vos E , Newberger E. Barriers to physician identification and treatment of
family violence: Lessons from five communities. Academic Medicine, 1997,72:S19-S25.
130.
Fawcett G et al. Deteccion y manejo de mujeres victimas de violencia domestica:
Desarrollo y evaluacion de un programa dirigido al personal de salud. [SPA] [Detection
and Management of female victims of violence: Development and evaluation of a
program for health care personnel]. Population Council: Mexico City, 1998. p. 30.
131.
Kim J , Motsei M. "Women enjoy punishment:" Attitudes and experience of gender
violence among PHC nurses in South Africa, 1999,. p. 10.
132.
Watts C , Ndlovu M. Addressing violence in Zimbabwe: Strengthening the health sector
response. In: Violence Against Women in Zimbabwe: Strategies for Action. Harare,
Musasa Project, 1997: p. 31-5.
133.
d'Oliveria AFPL , Schraiber LB. Violence against women: A physician's concern? in 15th
FIGO World Congress of Gynecology and Obstetrics. 1997. Copenhagen, Denmark.
134.
Leye E, Githaniga A , Temmerman M. Health care strategies for combating violence
against women in developing countries . International Center for Reproductive Health:
Ghent, Belgium, 1999. p. 125.
135.
Red de Mujeres contra la Violencia. Como atender a las mujeres que viven situationes de
violencia domestica? Orientaciones basicas para el personal de salud. [SPA] [Care of
women living with domestic violence: Orientation for health care personnel]. Red de
Mujeres Contra la Violencia: Managua, Nicaragua, 1999. p. 58.
136.
Pan American Health Organization (PAHO). Achievements of project "Toward a
comprehensive model approach to domestic violence: Expansion and consolidation of
interventions coordinated by the state and civil society" . Women, Health and
Development Program, PAHO: Washington, D.C., 1999. p. 5.
137.
National Center for Injury Prevention and Control. National Center for Injury Prevention
and Control. Centers for Disease Control and Prevention, 2000.
39
138.
Olson L et al. Increasing emergency physician recognition of domestic violence. Annals
ofEmergency Medicine, 1996,27:741 -6.
139.
Freund KM, Bak SM , Blackball L. Identifying domestic violence in primary care
practice. Journal of Gen Internal Medicine, 1996,11:44-6.
140.
Kim J. Health sector initiatives to address domestic violence against women in Africa, in
Health Care Strategies for Combatting Violence Against Women in Developing
Countries. 1999. Ghent, Belgium: International Centre for Reproductive Health,
University of Ghent.
141.
Davison L et al. Reducing domestic violence...What Works? Health Services . Policing
and Reducing Crime Unit, United Kingdom, 2000.
142.
Kelly L , Humphreys C. Reducing domestic violence...what works? Outreach and
advocacy approaches . Policing and Crime Reduction Unit, United Kingdom, 2000.
143.
Grau J, Fagan J , Wexler S. Restraining orders for battered women: Issues of access and
efficacy. Women and Politics, 1984,4:13-28.
144.
Harrell A , Smith BN, L. Court processing and the effects of restraining orders for
domestic violence victims. The Urban Institute: Washington D.C., 1993.
145.
Buzawa ES , Buzawa CG. Domestic violence : the criminal justice response. 2nd ed.
Thousand Oaks, Calif., SAGE Publications. 1990.
146.
Keilitz S et al. Civil Protection Orders: Victims' Views on Effectiveness . National
Institute of Justice: Washington D.C., 1998,. p. 5 p.
147.
Littel K, Malefyt JD , Walker AH. Assessing the justice system response to violence
against women: A tool for communities to develop coordinated responses, 1998.
148.
Ghez M , Marin L. United States: Prevention at a crossroads. In: Marin L, Zia H, Soler E,
Editors, ed. Ending Domestic Violence: Reports from the Global Frontline. San
Francisco, Family Violence Prevention Fund, 1999:.
149.
Gamache DJ, Edleson JS , Schock MD. Coordinated police, judicial, and social service
response to woman battering: A multiple baseline evaluation across three communities.
In: Hotaling GT, Finkelhor D, Kirkpatrick JT, Straus MA, Editors. Coping with Family
Violence: Reserach and Policy Perspectives. Newbury Park, Calif:, Sage, 1988: p. 193209.
150.
Hague G. Reducing domestic violence...What Works? Multi-agency Fora . Policing and
Crime Reduction Unit, United Kingdom, 2000.
151.
Corsi J. Treatment for men who batter women in Latin America American Psychologist,
1999,54:64.
40
152.
Cervantes Islas F. Helping men overcome violent behavior toward women. In: In:
Morrison AR, Biehl ML, Editors. Too Close to Home: Domestic Violence in the
Americas. Washington D.C., Inter-American Development Bank, 1999: p. 143-7.
153.
Axelson BL. Violence against women- A male issue. Choices, 1997,26:9-14.
154.
Edelson J. Controversy and change in batterer's programs. In: /Edleson JL, Eisikovits ZC,
Editors, ed. Future Interventions With Battered Women and Their Families. Thousand
Oaks, California, Sage Publications, 1995: p. 154-69.
155.
Mullender A , Burton S. Reducing domestic violence...what works? Perpetrator
Programmes . Policing and Crime Reduction Unit, Home Office, London, 2000.
156.
Edleson J, L. Intervention for men who batter: a review of research. In: Stith SR, Straus
MA, Editors. Understanding partner violence: Prevalence, causes, consequences and
solutions. Minneapolis, MN, Naitonal Council on Family Relations, 1995: p. 262-73.
157.
Gondolf E. A 30-month follow-up of court mandated batterers in four cities in
http://www.iup.edu/maati/publications/30month.htmlx, 1999.
158.
Tolman RM , Edleson JL. Intervention for men who batter: A review of research. In:
Smith SR, Straus MA, Editors. Understanding Partner Violence: Prevalence, Causes,
Consequences and Solutions. Minneapolis, Minnesota, Nation Council on Family
Relations, 1995: p. 262-73.
159.
Ellsberg M, Lilj estrand J, Winkvist A. The Nicaraguan Network of Women Against
Violence: Using research and action for change. Reproductive Health Matters,
1997,10:82-92.
160.
Mehrotra A et al. A Life Free of Violence: It's Our Right. UNIFEM, 2000.
161.
Jaffe PG et al. An evaluation of a secondary school primary prevnetion program on
violence in intimate relationships. Violence and Victims, 1992,7:129-46.
162.
Krajewski SS et al. Results of a curriculum intervention with seventh graders regarding
violene in relationships. Journal ofFamily Violence, 1996,11:93-112.
163.
Lavoie F et al. Evaluation of a prevention program for violence in teen dating
relationships. Journal of interpersonal violence, 1005,10:516-24.
164.
Heise L. Violence against women: Global organizing for change. In: Edleson JL,
Eisikovits ZC, Editors. Future Interventions with Battered Women and Their Families.
Thousand Oaks, California, Sage Publications, 1996: p. 7-33.
165.
International Center for Research on Women. Domestic Violence in India . International
Center for Research on Women: Washington D.C., 1999,. p. 52.
41
166.
American College of Obstetricians and Gynecologists (ACOG). ACOG committee
opinion. Mandatory reporting of domestic violence. Number 200, March 1998.
Committee on Health Care for Underserved Women. International Journal of
Gynaecology and Obstetrics, 1998,62:93-5.
167.
Hyman A, Schillinger D , Lo B. Laws mandating reporting of domestic violence: Do they
promote patient well-being? Journal of the American Medical Association,
1995,273:1781-7.
168.
Jezierski MB, Eickholt T , McGee J. Disadvantages to mandatory reporting of domestic
violence [letter]. Journal ofEmergency Nursing, 1999,25:79-80.
169.
Washington College of Law, Pan American Health Organization , Health and
Development Policy Project. Conclusions and recommendations: Pan American
consultation oflegal and health experts, Unpublished. 1996.
170.
Bradley J et al. Whole-site training: A new approach to the organization of training in
AVSC Working Paper. AVSC International: New York, 1998,. p. 16.
171.
Cole TB. Case management for domestic violence. Journal of the American Medical
Association, 1999,282:513-4.
172.
McLeer SV et al. Education is not enough: A systems failure in protecting battered
women Annals ofEmergency Medicine, 1989,18:651 -3.
173.
Tilden VP , Shepherd P. Increasing the rate of identification of battered women in an
emergency department: Use of a nursing protocol. Research in Nursing Health,
1987,10:209-15.
174.
Harwell TS et al. Results of a domestic violence training program offered to the staff of
urban community health centers. Evaluation Committee of the Philadelphia Family
Violence Working Group. American Journal ofPreventive Medicine, 1998,15:235-42.
175.
Kelly L. Tensions and possibilities: Enhancing informal responses to domestic violence.
In: Edelson JL, Eisidovits ZC, Editors. Future interventions with battered women and
theirfamilies. Thousand Oaks, CA, Sage, 1996: p. 67-86.
176.
Shrader E. La ruta critica: An evaluation of institutional responses to domictic violence,
in American Public Health Association. 1998. Washington D.C.
177.
Fawcett GM et al. Changing community responses to wife abuse: A research and
demonstration project Iztacalco, Mexico. American Psychologist, 1999,54:41-9.
178.
Carter J. Domestic violence, child abuse, and youth violence: Strategies for prevention
and early intervention . Family Violence Prevention Fund, 2000.
179.
Johnson J , Sacco V. Researching violence against women: Statistics Canada’s national
survey. Canadien Journal of Criminology, 1995:281-304.
42
180.
PROFAMILIA. Encuesta Nacional de Demografia y Salud 1995 in Demographic and
Health Surveys. PROFAMILIA and Macro International Inc.: Bogota, Colombia, 1995.
181.
Macro International Inc. , National Statistics Office M, Philippines,. National Safe
Motherhood Survey, 1993 . National Statistics Office, Manila, Philippines, Macro
International Inc., 1994,. p. 91-6.
182.
Blanc AK et al. Negotiating reproductive outcomes in Uganda . Institute of Statistics and
Applied Economics and Macro International, 1996,. p. 215.
183.
Watts C. Unpublished data on violence . London School of Hygiene and Tropical
Medicine, In press.
184.
Smith MD. Enhancing the Quality of Survey Data on Violence Against Women Gender
& Society, 1994,8:109-27.
185.
Koss MP. Detecting the scope of rape: a review of prevalence research methods. Journal
ofInterpersonal Violence, 1993,8:198-222.
186.
Hamby SL, Poindexter VC , Gray-Little B. Four measures of partner violence: Construct
similarity and classification differences. Journal of Marriage and the Family,
1996,58:127-39.
187.
Haj-Yahia MM. The incidence of wife-abuse and battering and some socio-demographic
correlates as revealed in two national surveys in Palestinian society . Besir Center for
Research and Development: Ramallah, West Bank, 1998,. p. 108.
188.
Brush LD. Violent acts and injurious outcomes in married couples: Methodological
issues in the National Survey of Families and Households. Gender and Society,
1990,4:56-67.
189.
Fontes LA. Ethics in Family Violence Research: Cross Cultural Issues. Family Relations,
1998,47:53-61.
190.
Finkelhor D, Hotaling GT , Yllo K. Special Ethical Concerns in Family Violence
Research. In: /Finkelhor D, Hotaling GT, Yllo K, Editors, ed. Stopping family violence:
Research priorities for the coming decade. London, Sage, 1988: .
191.
Liss M , Solomon S. Ethical considerations in violence-related research . National
Institutes of Health: Bethesda, 1996,. p. 31.
192.
World Health Organization. Putting women's safety first: Ethical and safety
recommendations for research on domestic violence against women . Global Programme
on Evidence for Health Policy, World Health Organization; Geneva, 1999. p. 13.
193.
Macro International (MI), South Africa Department of Health. South Africa demographic
and health survey 1998: Preliminary report. MI: Calverton, Maryland, 1998,. p. 41.
43
194.
Davila AL, Ramos G , Mattei H. Encuesta de salud reproductiva: Puerto Rico, 1995-1996.
[SPA] [Reproductive health survey: Puerto Rico, 1995-1996] . Centers for Disease
Control and Prevention: San Juan, Puerto Rico, 1998. p. 82.
195.
Centro Paraguayo de Estudios Poblacion, US Centers for Disease Control and Prevention
(CDC) , US Agency for International Development. Paraguay: Encuesta nacional de
demografia y de salud reproductiva 1995-1996 ENDSR-95-96. [SPA] [Paraguay national
demographic and reproductive health survey 1995-1996] . CDC: Asuncion, Paraguay,
1997. p. 278.
196.
Jewkes R et al. He must give me money, he mustn't beat me: Violence against women in
three South African Provinces. Medical Research Council: Pretoria, South Africa, 1999.
p. 29.
197.
Deyessa N, Kassaye M , Demeke B. Magnitude, type, and outcomes of physical violence
against married women in Butajira, Southern Ethiopia. Ethiopian Medical Journal, 1998,
36: 83-92.
44
Chapter 5
Elder Abuse
Status:
Draft: 3rd
Peer Reviewed: Yes
Date of Current Draft: 25 August 2000
Table of Contents
I.
n.
m.
IV.
V.
VI.
VII.
Introduction
Definition of Elder Abuse
A. Developed Countries
B. Developing Countries
Scope of the Problem
A. Elder Abuse in Domestic Settings
B. Elder Abuse in Institutional Settings
Elder Abuse in Domestic Settings
A. Risk Factors
B. Consequences of Mistreatment
Elder Abuse in Institutional Settings
Responses to the Problem
A. National Responses
B. Intervention Strategies
1. Social Services
2. Health Care
3. Legal Actions
4. Education, Training and Public Awareness
Summary and Recommendatons
A. The Need for Knowledge
B. The Call for Public Policies
C. The Importance of Prevention Strategies
2
Introduction
Abuse of older persons by family members has been a consistent theme in world literature since
ancient times but remained a private matter, hidden from public scrutiny, until the advent of the
battered child and women's movements in the last part of the 20th century. Initially framed as a
social welfare and then ageing issue, elder abuse, like other forms of family violence, has
become a public health and criminal justice concern and thus open to the conceptual and
methodological perspectives, instruments, and modalities of intervention employed in those
domains. The focus of the chapter is on abuse of elders by family members or persons known to
them either in their homes or institutional/residential settings. It does not deal with the broader
concept of violence that includes crime by strangers, street crime, gang warfare, or military
conflict.
Mistreatment of older persons, which has come to be known as "elder abuse," was first described
in British scientific journals in 1975 as "granny battering." (1,2). However, it was the US
Congress that first seized upon the problem as a socio-political issue, joined later by researchers
and practitioners.
By the 1980s reports of scientific investigations and/or governmental
activities were coming from the US, Canada, Norway, Sweden, Hong Kong, and Australia and in
the 1990s from the UK, other European countries, Israel, India, South Africa, Latin America and
Japan. Although elder abuse was first identified in the developed countries, which have
produced the extant research on the topic, reports and anecdotal evidence from some developing
countries reveal its occurrence in the South. The emergence of the problem reflects a growing
world wide concern about human rights, gender equality, domestic violence, and population
ageing.
Numbers and Trends in Ageing
Concern about the mistreatment of older persons is particularly disturbing given the expected
population explosion in the older age categories in both developed and the developing nations.
Presentation of data on numbers and trends in ageing is complicated by the fact that ageing is
defined differently in these nations. In western society, old age has become associated with
chronological time; generally, set at the age of retirement, 60 or 65 years of age. However,
among the developing nations, the socially constructed concept based on school age, work age,
or retirement age has little significance. A more meaningful interpretation is related to the roles
that are assigned to persons during their lifetime. Thus old age is viewed as that time of life
when persons, because of physical decline, can no longer carry out their role in the family or the
field. As the definition of "elder abuse" has evolved, the chronological standard has been
adopted by both developing and developed countries. Elders in this context are persons 60 or 65
years, although in practice some consideration is given to including younger persons 50 or 55
years who may have shortened life expectancy, such as the developmentally disabled. Studies
may not use a numerical age but ask older persons their experience with abuse since the "age of
retirement."
It is expected that by the year 2025, the global population of persons 60 years and older will
double, from 590 million to 1.2 billion. Throughout the world, 1 million people turn 60 every
month, 80 percent of whom are in the developing world. In 2050, the percentage of people over
3
60 will exceed those under 15, and the gap will continue to widen for the next 100 years.
Although the proportion of elders to the total population will remain higher in the developed
nations, the percentage of increase of the elderly population will be greater in the developing
countries. Aged populations in countries such as Germany, France, or Sweden are undergoing 30
to 60 percent increases, while developing countries such as Thailand, Kenya, and Columbia are
expected to experience a more than 300 percent and Indonesia, a more than 400 percent increase
in their older population through 2025 (3). The number of older people in the developing
countries will more than double, reaching 12 percent of their total population. For example,
because of lower fertility and mortality rates in Latin American and the Caribbean, the proportion
of older persons is expected to reach 10 percent of the population. By 2020, Cuba, Argentina,
Thailand and Sri Lanka will have a higher proportion of persons over 65 years than the United
States (4). The over 65 population of China will grow from 63 million in 1990 to 400 million in
2050. Ghana's 60 years and older cohort will increase from 1.0 to 2.2 million; South Africa,
from 2.7 to 6.3 million, and Tanzania from 1.6 to 3.1 million in the next 25 years.
Women are the majority of the older population in all nations although it is reported that more
men are represented in most older groups in India, Bangladesh and Egypt (4). Due to improved
medical science and preventive medicine, life expectancy of women will continue to exceed that
of men. Today 58% of older women live in the developing world; by 2025, this percentage will
increase to 75%.
However, the gender gap is much smaller in developing countries, due
primarily to higher rates of maternal mortality and lately to the HJV/AIDS epidemic.
This demographic revolution is taking place in developing countries alongside increases in
mobility, emigration, economic recession, and changing family characteristics. The process of
industrialisation has eroded long-standing patterns of interdependence between the generations
producing material and emotional hardships for elders (5/ Family and community networks in
many developing countries that formally provided support tp their older generation have been
undermined by social and economic changes. The AIDS pandemic is changing the situation of
older persons both quantitatively and qualitatively. About one in three adults in Botswana are
already infected which means that two-thirds of today's 15 year olds will die of AIDS. In South
Africa where about 20% of the population is infected with HIV, researchers forecast that the
gross national product will be reduced 17% by 2010 (6). Children are being orphaned at an
alarming rate as parents die from the disease. Older persons who anticipated support from their
children in old age find themselves to be the primary caregivers today and without family to help
them in the future.
Only 30 percent of the world's aged are covered by pension schemes. In the conversion from a
planned to a market economy, many elders in Eastern Europe and the states of the former Soviet
Union have been left without retirement income and health and welfare services that were part of
the previous communistic regimes.. Structural inequalities in both the developed and developing
countries, which have resulted in low wages, high unemployment, poor health services, gender
discrimination, and lack of educational opportunities, have contributed to the vulnerability and
impoverishment of older persons. For elders in the developing world, the risk of communicable
diseases still exists. At the same time, as life expectancy increases, they will be subject to the
long-term, incurable, and often disabling diseases associated with old age. Environmental
hazards and social violence present further threats to well-being. Nevertheless, medical and
4
social welfare advances promise that for many elders, disability-free old age will increase in
length, diseases will be prevented or ameliorated with good health promotion and prevention
strategies, and the large number of older people will constitute a huge reservoir of experience,
wisdom, and common sense. Recognition of the potential challenges (poverty, industrialisation,
family migration, etc.) may not prevent elder abuse but with the involvement of young and old in
model strategies, changes may come about in the conditions that have allowed it to develop
unchecked.
This chapter begins with the definition of elder abuse, examples from both a developed and
developing country, and then gives information on the prevalence, risk factors, and consequences
of elder abuse in domestic and institutional settings. Examples of national responses to the
problem and intervention strategies follow. A list of recommendations completes the chapter.
Definition of Elder Abuse
Developed Countries
Elder abuse is a multidimensional construct. It can be used as an all-inclusive term representing
all types of abusive behaviour against the elderly or it can refer to a specific physical act. Most
experts agree that elder abuse can be an act of commission (abuse) or omission (neglect),
intentional or unintentional, and of one of more types: physical, psychological ( emotional,
verbal aggression), and financial abuse and neglect that results in unnecessary suffering, injury,
pain, loss and/or violation of human rights and decreased quality of life (7). Whether the
behaviour is labelled as abusive, neglectful, or exploitative may depend on how frequently the
mistreatment occurs, the duration, intensity, severity, consequences, and, most significantly, the
cultural context. For example, among the Navajo people what appeared to the outsider
(researcher) as economic exploitation by family members was in part defined by the elders as
their cultural privilege and duty to share with their families (8). Other Indian tribes interpreted
elder abuse not as a problem of the individual but of the community (9).
A definition developed by the UK’s Action on Elder Abuse (10) and adopted by the International
Network for the Prevention of Elder Abuse (77) states that: ’’Elder abuse is a single, or repeated
act, or lack of appropriate action, occurring within any relationship where there is an expectation
of trust which causes harm or distress to an older person." It is usually categorised in the
following way.
Physical Abuse: the infliction of pain or injury, physical coercion, physical/chemical
restraint
• Psychological/Emotional Abuse: the infliction of mental anguish
• Financial/Material Abuse: the illegal or improper exploitation and/or use of funds
or resources
• Sexual Abuse: non-consensual contact of any kind with an older person
• Neglect: the refusal or failure to fulfil a caretaking obligation including/excluding a
conscious and intentional attempt to inflict physical or emotional distress on the older
persons
•
5
Case of mistreatment often include more than one type.
Even though the definition of elder abuse has been heavily influenced by the work done in North
America and Great Britain, reports coming from Finland, Greece, Hong Kong, Israel, India,
Ireland, Poland, and South Africa showed some distinctive differences (72). Using a
phenomenological approach, Norway researchers identified abuse with the 'triangle of violence'
that includes a victim, perpetrator, and others, who, directly or indirectly, observe the actors. For
those countries (Hong Kong/China) with an emphasis on harmony and respect, circumventing or
ignoring the care or treatment of an elder is called an act of elder abuse. The failure of family
members to fulfil kinship obligations in providing food or housing could also constitute neglect.
For the practitioner, cases of elder abuse present a myriad of problems. The ageing process and
diseases of old age compound the impact of physical and psychological violence on the health
status of the elder victims. Cognitive and functional deficits diminish the elders' ability to leave
a dangerous relationship or to make good decisions. Cultural values regarding kinship obligations
and use of the extended family network to resolve difficulties rather than outsiders may also
weaken the ability of elders, particularly women, to leave abusive situations. Often, the abuser
may be the victim's only source of companionship. Isolation and frailty of the victims may
induce unscrupulous family members and acquaintances to acquire, through deception and undue
influence, their financial and material property, leaving them homeless and penniless. With
regard to intervention, the most difficult question in cases of elder abuse is how to balance the
elder's right to self-determination and the clinician's belief that something should be done.
Developing Countries
In many nations, older people were to be respected and provided comfort and leisure in old age
(13). This reverence for the aged was reinforced through the major philosophical traditions and
public policy. In Chinese society it was imbedded in a value system that stressed "filial piety."
Mistreatment of older persons was unrecognised and certainly unreported. A study of KoreanAmericans attitudes toward elder abuse revealed their belief in primacy of family harmony over
individual well being in determining whether behaviour was perceived as abusive or not (14).
Similarly, Japanese-Americans considered the "group" to be important and paramount with the
individual's well-being sacrificed for the group (15).
Displacement of elders as heads of the household and deprivation of their autonomy in the name
of love are cultural norms even in countries where the family is the central institution and the
sense of filial obligation is strong (16). Nevertheless, this "overprotection" leaves the older
person feeling isolated and demoralised. Such infantalisation may be considered a form of abuse.
In some traditional societies, older widows are subject to abandonment and "property grabbing."
Mourning rites of passage for widows in most of Africa and India include cruel practices, sexual
violence, forced leviterate marriages, and evacuation from their homes (17). Accusations of
witchcraft are directed at isolated, older women, often connected with unexplained events in the
local community, such as a death or crop failure (5,18). In sub-Sahara Africa, the practice of
witchcraft has driven many older women from their homes and communities to live in poverty in
the urban areas. In Tanzania an estimated 500 women are murdered every year after being
6
accused of witchcraft (19). These acts of violence are customs that have been firmly entrenched
in the social structure and may not be readily identifiable as "elder abuse." (BOX-Witchcraft).
An early workshop (1992) in South Africa differentiated between mistreatment (verbal abuse,
passive and active neglect, financial exploitation and over medication) and abuse(physical,
psychological and sexual; and theft) (12). More recently, focus groups were held with older
persons recruited in three historically "black" townships in South Africa to determine the level of
knowledge and understanding of elder abuse. In addition to the typical western schema of
physical, verbal, financial, sexual and neglect, the participants included (1) loss of respect for
elders, which was paired with neglect, (2) accusations of witchcraft and consequences of being a
witch, and (3) systemic abuse. They produced the following definitions (20):
Physical Abuse: beatings, pushing, shoving
Emotional/Verbal:
age discrimination, hurtful words, insults, denigration,
intimidation, false accusations, psychological pain, and distress
• Financial Abuse: extortion and control of pension money, exploitation to render
care services to grandchildren, theft of property
• Sexual Abuse: incest rape, criminal rape, other types of deviant sexual behaviour
• Neglect: loss of respect for elders, withholding affection, a lack of interest in the
elder’s well-being
• Accusations of witchcraft: brandishment, stigmatisation, ostracism, physical danger
• Systemic Abuse: the dehumanising treatment given older persons at health clinics,
and pension offices, marginalisation by the government
•
•
These lay definitions (classified by the researchers) were the result of an initial effort in South
Africa to elicit information directly from older persons about elder abuse and represent a first
attempt at classification of elder abuse in a developing country. They build on the Western
schema but include aspects that are relevant to the indigenous population.
Scope of the Problem
Elder Abuse in Domestic Settings
With most developing nations just becoming aware of the problem, information on the rate of
elder abuse has relied on five community surveys conducted in the past decade in five developed
countries (21-26). The results show a 4% to 6% rate if physical, psychological abuse (verbal
aggression), and financial abuse and neglect are included (Table 1). A difficulty in making
comparisons is the variation in the time frame among the studies. The Boston, Canada, and
Amsterdam survey findings refer to the "preceding year." The Finnish town study results were
based on abuse since the "age of retirement," and the British survey, "in the past few years." The
Boston, Canada, and Amsterdam research found no difference in prevalence rates by age or
gender; the Finnish authors report a higher proportion of female victims (7.0%) compared to men
(2.5%). (No breakdown by age or gender given in the British survey). Because of the
differences in the methodology used by the 5 surveys and relatively small numbers of victims,
further comparative analysis is not warranted.
7
Since the estimates of all five surveys were based on self-reports and, except for the Boston
project, excluded individuals who could not respond to a survey question, the percentages are
considered to be an underestimation of the problem. However, the percentages are still lower
than those that have been associated with child abuse and domestic violence. The US National
Family Violence Surveys (1985) found a rate of 161 per 1,000 women had been subject to a
violent act by an intimate partner during the preceding year and 34 per 1000 for "severe
violence." Using the same criteria, the rate for child abuse was 110 incidents per 1000 children
and 6.9 if "hitting with an object" was omitted from the list of physically abusive acts (20).
Studies of violence against women in a group of developed and developing countries showed
percentages ranging from 16% of women in 62 Belgium municipalities who experienced "serious
to moderately serious violence" to 60% of women in Santiago, Chile who experienced "abuse by
male intimate"(27). National random sample surveys conducted in low income countries also
uncovered high proportions: 39% of peninsular Malaysian women reported being physically
beaten by a partner; 20% of Columbian women were physically abused; 33%, psychologically
abused; and 10%, raped by husbands .
A newly released Canadian survey of family violence found 7% of people (aged 15 years and
older) who were married or living in a common-law relationship were the victims of some type
of violence by a partner during the previous 5 years. The 5-years rate of violence (threatened,
threw something, pushed, slapped, kicked, etc.) was similar for women (8%) and men(7%).
However, women tended to report more severe violence and more negative emotional
consequences than men. They were more likely to be injured and to report repeated
victimizations to the authorities. The survey reported 7% of older adults experiencing some
form of emotional abuse, 1%, financial abuse, and 1% physical abuse or sexual assault in the
previous 5 years by children, caregivers (paid and unpaid), and partners. Men (9%) were more
likely than women (6 %) to report being victims of emotional or financial abuse. Because of
differences in the survey questions and time frame, these findings cannot be compared to the
earlier Canadian study which found a much smaller proportion of emotional abuse (1.4%) and a
larger proportion of financial abuse (2.5%) (28..
Elder Abuse in Institutional Settings
In developed countries where the proportion of elders in institutions had reached a high of 9%
(29), a shift in emphasis has occurred in recent years towards care by the community and use of
less restrictive residential settings. Current rates for nursing home utilisation are in the range of
4-7 % They include Canada (6.8%), the United States (4%) as well as countries like Israel
(4.40/0)
SoUth Africa (4.5%). Older people, in most African countries, can be found in longstay hospital wards, homes for the destitute and disabled and in sub-Saharan countries, in
witches camps. Social, economic, and cultural changes that are taking place in some of the
developing societies will leave families less able to care for their frail relatives and thus portend
an increasing demand for institutional care (5). The demand for institutional care for older
persons is becoming the normal expectation of the general public in China. In fact, institutional
care has rapidly bypassed family care in Taiwan (13).
Utilisation figures for the Latin American countries range from 1-4%. Institutional care is no
longer viewed as an unacceptable place for an old person but considered as an alternative for
8
families. The government sponsored 'asilos' in Latin America, large institutions resembling the
early English workhouses, have been converted to smaller facilities with professional staff
representing many disciplines.
Other homes are operated by religious commumties of
immigrant origin. Utilisation figures are not available in the countries of the former eastern bloc
in Europe, because the authorities did not allow publication of scientific studies about
institutions.
Despite the fact that a vast literature exists on quality of care issues in institutional settings, and
abuses have been well-documented in reports of governmental inquiries, ethnographic studies,
and personal histories, no prevalence or incidence data are available for any country. A survey of
nursing home personnel in one US state disclosed that 36% of the nursing and aide staff reported
having seen at least one incident of physical abuse in the preceding year by other staff members,
and 10% admitted having committed at least one act of physical abuse themselves. At least one
incident of psychological abuse against a resident had been observed by 81% of the sample in the
preceding year, and 40% admitted to having also committed such an act (30).. The findings
suggest that mistreatment of older residents may be even more extensive than generally believed.
The rates of elder abuse in the community and the extent in institutional settings are also greater
than the general statistics on violent acts collected by countries would indicate. Some of the
disparity is due to the fact that elder abuse has gone unrecognised until the last few decades.
Some deaths of older persons both in institutional settings and the community have been
attributed to natural, accidental, or undetermined circumstances when in fact they were the
consequences of abusive or neglectful behaviour.
Although the relative numbers of older persons who are victims of different forms of violence
may be less or more than younger populations, the repercussions can be very serious. Compared
to younger adults, older people are physically weaker and more vulnerable, their bones more
brittle, and their convalescence slower. Even a relatively minor injury can cause serious,
permanent damage. Many have limited incomes so that the loss of only a small amount of money
may have a significant impact. If isolated, lonely, or troubled by illness, they become likely
targets for fraudulent schemes. To the degree that concern about personal safety limits choices,
reduces independence, and diminishes autonomy, the quality of life is threatened.
9
Elder Abuse in Domestic Settings
Risk Factors
Most of the early work on elder abuse was limited to domestic settings and carried out in the
developed countries. In searching for explanations for elder abuse, researchers drew from the
psychological, sociological, gerontological and family violence literature.
As a way of
accommodating the complexity and multiplicity of factors associated with elder abuse,
researchers are turning to the ecological model, first applied to child abuse (37,32) and more
recently to intimate partner violence (33-35). The ecological model is selected because it can
account for interactions that take place across multiple systems, initially conceived as a nested
arrangement of four levels of the environment. Mistreatment of elders in this conceptualization
is viewed as the interplay of individual, interpersonal, social contextual, and societal factors.
Individual Determinants. Early on, researchers renounced individual personality disturbances as
causal agents of family violence in favour of socio-psychological and socio-cultural factors (36).
However, more recent research in family violence has shown that abusers who are the most
physically aggressive are more likely to have personality disorders and alcohol problems than the
general population (37). Likewise, elder abuse studies have found that perpetrators are more apt
to have mental health and substance abuse problems than family members or caregivers who are
not abusive. (38-40).
Cognitive and physical impairments of the abused elder were also strongly identified in the initial
studies as risk factors. However, later work with abuse and non-abuse cases from a social
service agency revealed that these elders were not more debilitated than non-abused elders and
might even be less so, particularly in cases of physical and verbal abuse (41). Similarly, a
comparison of abuse and non-abuse cases from samples of Alzheimer patients showed degree of
impairment was not a factor (42,43) in abuse by the caregiver. Among cases reported to
authorities, however, a greater proportion are the very old and the most impaired .
Historically, culturally sanctioned beliefs about the rights and privileges of husbands have led to
the domination of men over women throughout the world. Feminist theorists contend that
power and gender are key elements leading to intimate partner violence but the relationships may
be more complex than these factors can explain. Gender has also been proposed as a defining
issue in elder abuse because older women especially have been subject to oppression and
economic disadvantage all of their lives (44).. Unlike intimate violence involving younger
people, in which the victim is almost always the women, older men are also at risk of abuse by
their spouses, adult children, and other relatives in about equal proportions to women as evidence
by the community based prevalence studies (21,22). It has been noted that the "actions and
writings of most older women give little indication that they wish to emphasise the issue of
gender... they may not wish to promote the understanding of women's oppression through elder
abuse at the expense of abused older men.... It also may reflect the internalisation of societal
sexism and their belief that they are of less value than men and may need men's power to
succeed" (45, p.71).
10
Interpersonal Context. Derived from early theoretical models, caregiver stress as a risk factor
became the centre of the framework used to link elder abuse with the caring of an elderly
relative. (46,47). Although the popular image of abuse depicted a dependent victim and stressed
caregiver, evidence has accumulated that neither of these factors differentiate between abuse and
non-abuse cases . While not denying the stressful component, researchers have tended to
incorporate it into a wider context examining the quality of relationships as a causal factor
(39,43,48). Some of the studies on caregiver stress, Alzheimer's disease, and elder abuse suggest
that the nature of the pre-morbid caregiver-carer relationship may be the important predictive
factor (43,49,50). Today, the conclusion is that stress may be a contributing factor in cases of
abuse but does not explain the phenomenon.
It is important to add that work with dementia patients has shown that violent acts perpetrated by
a care recipient can act as “triggers” for reciprocal violence by the carer (5/). It may be that the
violence is a result of the interplay of several factors “..stress, the relationship between the carer
and the care recipient, the existence of disruptive behaviour and aggression by the care recipient,
and the depression in the caregiver "(52).
Living arrangements, particularly overcrowded living conditions and lack of privacy have been
associated with intrafamily conflict. Although abuse can occur when the abuser and victim live
apart, the older person is more at risk when living with others. Dependency is another risk factor
associated with the early theories. First interpreted as the dependency of the victim on the
caregiver or abuser, later work on cases of physical abuse identified abusers who were dependent
on the victims, usually adult children dependent on elderly parents for housing and financial
assistance (41). In some of these cases a "web of interdependency" was evident, a strong
emotional attachment between abused and abuser that often hindered intervention efforts.
Social Context/Institutional Influences. In almost all risk factor analyses, social isolation
emerges as a significant variable in elder mistreatment (22,38,54,55). Similar to the work done
with battered women, isolation can be both a cause and consequence of abuse. In the case of
older persons, many are isolated because of physical and/or mental infirmities. Further, the loss
of friends and family members reduces the opportunities for social interaction.
Although income of the elder was not a significant factor in the prevalence study earned out in
the USA, financial difficulties of the abuser did surface as an important risk factor. Sometimes,
it was in connection with a substance abuse problem in which the adult child extorted money
from the elder, usually a pension check or it could be resentment on the part of family members
having to spend money on the care of the older parent.
While the emphasis in the developed world has been on individual or interpersonal attributes as
potential causal factors, cultural norms and traditions such as ageism, sexism, and violence, are
recognised as the context in which elders are viewed. The depiction of older persons as frail,
weak, and dependent has made them appear to be less worthy of governmental investment than
other groups and even of family care and ready targets for exploitation. The unequal status of
women may prevent some older women from seeking help. As noted earlier with respect to
witchcraft, some cultures condone violence.
11
Societal Factors. Societal factors are given great weight as risk factors for elder abuse in the
developing countries although there is no empirical evidence to date. With respect to Africa,
they include the patrilineal and matrilineal inheritance systems and land rights that affect the
political economy of relationships and the distribution of power inherent in them; the social
construction of gender that place older women at risk; rural-urban migration and formal
education that reduce the interdependency of generations within the family unit; and the loss of
the traditional domestic, ritual, and arbitration roles of elders within the family through
modernization (17).. According to the members in the South African focus group study, much of
the abuse occurred in the context of social disorganization, specifically domestic violence,
exacerbated by crime, alcohol, and drugs (20). Similar conclusions were found in an exercise
conducted by 7 male community leaders of the Tamaho squatter camp in Katlehong, South
Africa. To show the linkage between poverty and violence, they described how dysfunctional
family life, lack of money for essentials needs, and lack of education and job opportunities for
youth have contributed to a life of crime, drug peddling, and prostitution by young people.
Elders are viewed as targets for abuse and exploitation; their vulnerability a result of poverty
marked by a lack of pension support, job opportunities, poor dietary and hygiene practices,
disease, and malnutrition (17).
The period of political transformation affecting the eastern European post-communist nations
have also been suggested as producing conditions that have increased the risk of elder abuse.
The pauperization of significant parts of society, the lack of stability and social security, the
release of aggressive behaviour especially among young people, and unemployment are factors
that have effected the psycho-social and health status of ordinary people and, in particular,
elderly people, increasing their vulnerability to mistreatment (56). In Chinese societies a
multitude of factors have been suggested for mistreatment of older people including a lack of
respect by the younger generation, a state of tension between traditional and new family
structures, a restructuring of the basic support networks for the elderly, and migration of young
couples to new towns leaving elderly people in deteriorating residential areas in the town centres
(57).
The integrative model that encompasses individual, interpersonal, contextual, and societal
perspectives overcomes some of the bias evident in the elder abuse field at the present time
which has focused on interpersonal and family issues. It recognises the difficulties that older
persons face, especially older women, who live in poverty, without the basic necessities of life
and without family support and that this deprivation may increase their risk of abuse, neglect, and
exploitation. The ecological framework offers an approach that may better capture the common
elements of all forms of family violence issues and help lead to common solutions.
Consequences of Mistreatment
Very few empirical studies have been conducted to determine the consequences of mistreatment
even though clinical and case study reports about the severe emotional distress experienced by
older persons as a result of mistreatment are plentiful. Some evidence, from developed
countries, is available to show that there is a higher proportion of persons with depression or
psychological distress in an abuse sample than a non-abuse sample. (40,58,59). Since these
studies were cross-sectional in design, there is no way of knowing whether the condition was an
12
antecedent or consequence of the mistreatment. Other suggested symptomatology associated
with these cases include feelings of learned helplessness, alienation, guilt, shame, fear, anxiety,
denial, and post traumatic stress syndrome. (60,61). Emotional effects were also cited by the
South African focus group participants along with health problems and in the words of one
member, "illness of the heart" (20).
In a seminal study, data from an annual comprehensive health and welfare study of a
representative sample (n — 2,812) of elders in one US city was merged with the data base of the
local adult abuse agency for each year over a 9-year period (56). Information for the health
survey was recorded by nurses who saw the elders at the hospital for the first year's data
collection and every third year thereafter. In between years, the data were updated by telephone.
Information about abuse and neglect was obtained by case workers using existing protocols after
investigating the claim, usually involving a home visit. The merged database allowed the
researchers to identify who in the sample of 2,812 had been reported during the 9 years as a
substantiated case of physical abuse or neglect. The mortality rates were then calculated for the
non-abused, the physically abused, and neglected individuals beginning with the first year of the
survey and extending for 12 years thereafter. When the mortality rates were compared, by the
13th year following the initiation of the study, 40% of the non-reported (non-abused, non
neglected) group compared to 9% of the physically abused or neglected elders were still alive.
After controlling for all possible factors that might affect mortality (e.g., age, gender, income,
functional status, cognitive status, diagnosis, social supports, etc.) and finding no significant
relationships, the researchers speculate that mistreatment causes extreme interpersonal stress that
may confer an additional death risk.
Elder Abuse in Institutional Settings
Elder mistreatment has been identified in continuing care facilities (nursing home, residential
care, hospitals, day care facilities, etc.) in almost all countries in which they are located. Abuse
may occur at a number of different levels, for example, an older person may be abused or
neglected by a paid member of thestaff, another resident, a voluntary visitor or relatives and
friends. An abusive or neglectful relationship between the older person and their carer may not
necessarily cease on admission to institutional care; sometimes, the abuse may continue although
the setting has changed.
There is, equally, a distinction to be made between individual acts of abuse or neglect inflicted
upon individuals and instititutionalised abuse, in which the regime of the institution itself maybe
abusive . However, in reality, within an abusive or neglectful institution it is often difficult to
define whether the reasons for abuse are caused by individual acts or omissions or are due to
intrinsic managerial failings; both are often found in the same institution.
The spectrum of abuse and neglect within various types of facilities spans a remarkable range
(64) related to provision of 1) care (e.g., resistance to changes in geriatric medicine, erosion of
individuality in the care, inadequate nutrition, deficient nursing care (pressure sores, etc.), lack of
care in the terminal stages of life); 2) staffing (e.g., work related stress and staff burnout, poor
13
physical working conditions, insufficient training, personal psychopathology); 3) staff-resident
interaction (e.g., poor communication, aggressive residents, cultural differences); 4) environment
(e.g., lack of basic standards of privacy, use of various types of restraints, little sensory
stimulation; accident history; and 5) organisational policies (e.g., a "closed" institution, staff
shortages, run down establishment, lack of choice, fraud involving residents' possessions or
money, poor attitudes or culture; history of deficiencies; high staff turnover, lack of
resident/family council). As expressed in a report from India, institutional abuse is perpetuated
by staff through unquestioning regimentation (in the name of discipline or imposed protective
care) and exploitation of the elder's dependence, exacerbated by the lack of professionally trained
management.
With the present state of knowledge, it is impossible to know at this time how pervasive these
conditions are. The top ten deficiencies (broad categories) cited by the US government in their
1997 survey of 15,000 nursing homes were food preparation (21.8%), comprehensive
assessment (17.3%), comprehensive care plans (17.1%), accidents (16.6%), pressure sores
(16.1%), quality of care (14.4%), physical restraints (13.3%), housekeeping (13.3%), dignity
(13.2%), and accident prevention (11.9%) (65).
An assumption that an abusive or neglectful situation arises only in poor quality institutions is
probably not accurate. The differences in quality of care between different types of institution
may not be very great. A key finding from an examination of reports of inquiries into scandals
into residential care suggested that the change required to alter an acceptable or good care
practice into an abusive one was not very large and could occur rapidly with only a slight,
perhaps barely detectable, change in the situation (66/
Responses to the Problem
National Responses
Among the nations of the world, efforts to mobilise social action against elder abuse at a national
level and to develop legislation and other policy initiatives are at varying stages of development.
Several authors (67,68) have used Blumer's (69) model of social problems to describe the
process; emergence of a problem, legitimisation of the problem, mobilisation of action,
formulation of an official plan, and implementation of the plan. The US is furthest advanced
with a fully developed system for reporting and treating elder abuse cases that is state-based. The
federal government's involvement is limited to supporting a National Center on Elder Abuse
which offers technical assistance and provides a small amount of funds to the states for elder
abuse preventive services. A national focus is also provided by the National Committee for the
Prevention of Elder Abuse, a non-profit organisation formed in 1988 and the National
Association of State Adult Protective Services Administrators, organised in 1989.
In Canada and Australia, some provinces/states have systems to deal with elder abuse cases in
place but no official federal policy has been pronounced. New Zealand has established a series
of pilot projects throughout the country. All three have formed national groups. The New
14
or SXteres. in responding to abuse; and in 1W. the Canadian Network for the Pre.erd.on
of Elder Abuse, to promote ways of working together for the development of policies, programs
and services to eliminate elder abuse.
;
;(AEA)
— ), a national charity. AEA has
Action was mobilised in Britain by Action on Elder Abuse
th(T abuse of older people in the community and
helped to focus government attention on C.
institutions with resultant policy documents from the Department of Health and the Social
the Scandinavian countries with parliamentary
Service Inspectorate. Norway leads among
support for a service project in Oslo and a resource centre for information and studies on
violence, the latter primarily the result of action by elder abuse advocates Other European
countries (Germany, France, Italy, Poland), are at the "legitimisation stage; elder abuse
activities mainly limited to individual researchers and local service programs.
The Latin American Committee for the Prevention of Elder Abuse has been active: in ^ngmg
the problem of elder abuse to the attention of South Amencan and Caribbean countries by
offering training at regional and national meetings. For some cotmtnes, awareness
he
problem is still emerging (Peru, Uruguay, Venezuela, Cuba) with profession^ meetings
topic and research studies while others have moved on to legitimisation and action (Argen ,
Brazil and Chile). In Buenos Aires, "Proteger," which deals exclusively with elder abuse cases
was established two years ago as one of the programs of the Promotion of Social Welfare and
Old Age. Professionals and other workers in the program receive a six-month training penocI on
eerontology mainly focusing on prevention of violence and intervention on elder abuse. It als
maintains^ free helpline. Official sanction and support for training on elder abuse has been
m“d by tire Brazil Ministry of Justice, Health, and Welfare. In Chile, as the result o the
work of a 1994 interministerial commission for the prevention of violence an official bulletin
was issued about intrafamiliar violence that included the mistreatment of old persons. Studies
from researchers in Asian countries (Hong Kong, India, Japan, Korea) have. dlmnmec the
problem but no official recognition in terms of policies or program development has followed
far.
Reports about elder abuse in South Africa first appeared in 1981 and a P^ntive progjam on
institutional abuse was established in 1994 by the State and private sector (70). Elder abuse
activists have been responsible for initiating a National Strategy on Elder Abuse imder
consideration by the government and for inclusion of elderly in the final declaration of the
Southern African Development Countries Conference on the Prevention of Violence Agains
woman (7J) The Nigeria Coalition on Prevention of Elder Abuse involves all agencies and
groups that work with Ld for the elderly. For many other African nations, efforts to deal with
elder abuse are overshadowed by other seemingly more pressing concerns including debt,
poverty, and conflict.
The expansion of activities to recognise and prevent elder abuse world-wide led to ±e formation
of the International Network for the Prevention of Elder Abuse (INPEA)in 1997.
W
15
representatives from the six continents, INPEA is primarily devoted to increasing public
awareness, promoting education and training, furthering advocacy on behalf of abuse and
neglected elders, and stimulating research into the causes, consequences, treatment, and
prevention. During this initial stage in the organisation's development, workshops have been the
primary training medium, conducted at professional meetings in Brazil, Cuba, Canada, Australia,
the USA, and the UK. A quarterly newsletter and website have aided communication and
dissemination of information. INPEA was also an inspiration behind the formation of the
Canadian Network for the Prevention of Elder Abuse and the Australian Network for the
Prevention of Elder Abuse. Some preliminary planning for cross-cultural research is underway
but awaits further discussion and funding opportunities.
Intervention Strategies
Most of the programs organised to handle cases of abuse, neglect, and exploitation are located in
the high income countries. They are conducted under the auspices of the social service, health
care, and legal systems or in coordination with family violence programs.
Although the
presence of elder abuse in some low/middle income countries has been substantiated, generally
no specific programs have been established to deal with it. Clients under these circumstances
would be seen by the govemment/nongovemmental social service agencies whose staff may or
may not be informed about elder abuse. The case of Costa Rica (see BOX) is an exception (16).
In some countries, there is no social service or health care system to respond..
Social Services. Generally, countries delivering services to abused, neglected, and exploited
elders have done so through existing health and social service systems. Because of the
complexity of the cases that often involve medical, legal, ethical, psychological, financial, law
enforcement, and environmental issues, guidelines and protocols are used to assist the worker,
and special training made available to them. Multidisciplinary consulting teams with
representation from the various disciplines are called upon to assist in planning the care.
Coalitions, task forces, etc. are organised with representation from statutory, voluntary, private,
and charitable organisations to offer consultation, sponsor training, develop legislation, and
identify needs in the service system. Helplines to take reports of elder mistreatment are often
one of the first components in an elder abuse system (72,73). They are currently operating in
Britain and in local communities within France, Germany, and Japan. BOX. Only the US, along
with several Canadian provinces, has created a system (Adult Protective Services) solely for
handling reports of adult mistreatment, with case managers investigating cases, making
assessments, developing care plans, and monitoring the cases until they can be turned over to
existing ageing service agencies.
Since much of elder abuse is spouse abuse (see Table 1), there is growing interest in providing
services modelled after those developed for battered women. Emergency shelters and support
groups specifically for older victims are relatively new. Through sharing similar experiences, the
victims gain psychological resources to deal with their fear, self-doubt, stress, and anxiety. The
eventual goal is to improve self-esteem and enhance coping abilities. One example of the
adaptation of the domestic violence model to elder abuse is the program established by the
Finnish Federation of Mother and Child Homes and Shelters in cooperation with a local nursing
home and the open care system. The elder abuse project consisted of emergency shelter beds
16
located in the nursing home, a telephone assistance program that offered advice and an
opportunity for elders to talk about their situation, and a victim support group that met on a
biweekly basis. Other emergency shelters for elder victims of abuse have been located in the
US, Canada, Germany, and Japan.
In low income countries without the social service infrastructure to undertake the type of
programs listed above, local projects can be established that will enable older persons to plan
programs, develop their own services, and advocate for change at the community level. Through
these activities they gain a sense of empowerment and self esteem.
As an example, in
Guatemala, blind older people were being ejected from their homes by their families. They
formed a committee, created a safe house for themselves, and developed handicrafts and income
producing projects in the community to support it (74).
Health Care. In some European and Latin American countries and in Australia, members of the
medical profession have played a leadership role in bringing information about elder abuse to the
attention of public and government officials. In others (such as the US, Canada), physicians
are late comers many years behind the social work or nursing professions. Few intervention
programs for abused elders are housed in hospital settings; if so, they are consultation teams on
call when a suspected case of elder abuse is found. Health care providers are important
participants in detection and screening programs. Although physicians are considered to be in
the best position to identify cases of elder abuse because of the trust which most elders place in
them, many do not diagnose abuse because it is not part of their knowledge base and hence does
not enter into their list of differential diagnoses. There is also evidence little attention is paid to
the special needs of elderly people in emergency rooms; health care professionals feel less
comfortable caring for elderly people than for younger clients; social and personal concerns of
the older person are frequently ignored; psychosocial issues such as elder abuse, depression,
suicide prevention, and substance abuse are not commonly addressed; and most emergency
departments do not use protocols for detecting and dealing with elder abuse (75/
Investigation of a patient's condition for possible abuse (76,77) is warranted when the health
care provider notices:
delays between injury or illness and seeking medical attention,
implausible or vague explanations provided by either party,
differing histories from patient and abuser existence of a "closed institution,"
frequent emergency department visits for chronic disease exacerbations despite a care
plan and available resources,
• functionally impaired elders who arrive without the main carer present, and
• laboratory findings that are inconsistent with .the history provided.
•
•
•
•
17
When conducting an examination (78), the physician or health care provider should
•
•
•
•
•
interview the patient alone with direct enquiries about physical violence, restraints or
neglect;
interview the alleged abuser alone
observe the relationship and behaviour of the patient and alleged abuser,
conduct a comprehensive geriatric assessment that includes medical, functional,
cognitive, and social factors, and
document the social networks (formal and informal) available to assist the patient.
Table 2 presents a list of indicators that can serve as a guide when mistreatment is suspected;
however, the presence or absence of any of the indicators does not serve as a proof that abuse has
or has not taken place. As noted earlier, health care providers who are ill-prepared, overworked,
and unsympathetic to the needs of older people may be at risk of abusing the patient.
Legal Actions. Despite the growing interest in the problem, most countries have not passed
specific elder abuse legislation. Aspects of the problem can be covered under civil rights, family
violence, mental health, property rights, and criminal acts. However, specific laws on elder
abuse that cover definitions, investigation, assessment, and treatment procedures at the same time
underscoring the rights of the abused and abuser imply a stronger commitment to the eradicating
the problem. Even in those countries that have these laws, cases of elder abuse have rarely been
prosecuted because of elders' reluctance or inability to press charges against family members, the
perception that elders make poor witnesses, or the hidden nature of elder abuse. As long as elder
abuse is viewed as a caregiver issue, legal action is not a likely option. However, adopting the
family violence paradigm does imply involvement with the criminal justice system.
Only the US states, the Canadian Atlantic Canadian provinces, and Israel have mandatory
reporting legislation. Forty-three of the 50 states require professionals and others working with
elderly persons, if they have "reason to believe" abuse, neglect, or exploitation has occurred, to
report it to a state designated agency. The first state passed legislation in 1976; the most recent,
in 1999. Newfoundland passed its adult protection law in 1973; the last of the four provinces
(Prince Edward Island) in 1988 and Israel in 1989. Like the child abuse reporting laws, elder
abuse legislation was designed to prevent evidence of abuse from going unnoticed. Mandatory
reporting was considered to be a valuable means of case-finding, particularly in situations where
victims are unable to report and professionals are reluctant to refer to the service system.
Although research on the impact of mandatory reporting does not provide a clear answer as yet,
the conclusion at this time is that the decision to report a case appears to have less to do with the
legal mandate than with other organisational, ethical, cultural, and professional issues (79).
Education, Training, and Public Awareness. Education, training, and public awareness have
been critical elements in disseminating information about elder abuse in the developed nations.
Because education is not just learning new information but about changing attitudes, behaviours,
and values, it is a fundamental preventive strategy. Training sessions, seminars, continuing
education programs, workshops, educational and scientific meetings, and conferences on
gerontology and related disciplines are ideal venues for raising awareness, increasing knowledge,
presenting and evaluating researcher results, and sharing information. Audiences include not
18
only practitioners in disciplines such as medicine, nursing, social work, law, criminal justice,
ageing, mental health, religion, but also researchers, educators, policy makers, and
administrators. A basic generic curriculum suitable for most disciplines includes learning about
the background of elder abuse, the signs and symptoms, and local agencies to contact for
assistance. Intermediate training enhances skills in interviewing, assessment, and care planning.
Dealing with ethical and legal issues, which are often very difficult to resolve, requires more
advanced training from specialists in those areas. Since it is often necessary to consult with
various professionals when handling elder abuse cases, courses in how to work with other
professionals and in multidisciplinary teams have also become part of an advanced training
curriculum. The above information should also be available to health and welfare workers in
developing countries through print or electronic means.
Public education is an important aspect of preventing abuse and neglect. The goal is to inform
the general public about the types of abuse, how to identify the signs, and where to turn for help
in a similar way that has occurred with child abuse and domestic violence. Persons who come in
contact with elders and may be in a good position to identify those needing assistance, such as
family members, neighbours, friends, public service providers (e.g., postal workers, bank tellers,
utility meter readers) are a target group for education. Programs for older persons are usually
more successful if the information about abuse is incorporated into sessions dealing with a
broader topics such as successful ageing or health care.
The media is a powerful resource for raising public awareness about elder abuse. It can be very
influential in calling attention to the issue and in disseminating information. Positive imaging
and a greater role for older persons in the media can also help to reduce stereotyping and change
attitudes. The participants in the South African focus group study also called attention to the
importance of the media in increasing public awareness. 'It must be in the news. Everybody must
hear about if (20).
They suggested that awareness of the problem be promoted through
community workshops with the involvement of the government. In other developing countries
with little resources, local level associations can provide basic health care and education. Older
people’s organizations, community centers, and day care programs can serve an educational
function, providing instruction on safety and physical security. Self-help and support groups can
also be a means of disseminating information.
Very few intervention programs have been evaluated so it is not possible to identify which
approach is most successful . Efforts to assess the effectiveness of the projects have been
hampered by the absence of common definitions, varied theoretical explanations, disinterest by
the scientific community, and lack of funding necessary to support rigorous studies. Two
methods of measuring intervention success now under development include a risk assessment
tool to gauge the risk of future abuse and a system to evaluate program performance in terms of
client outcomes. A literature review of elder abuse intervention studies located 117 that had
been published (in English) between 1989 and 1998 (80) but not one had a comparison group or
met standard criteria for a valid evaluation study. Based on the findings, the authors state that
there was insufficient evidence to support any specific intervention for elder abuse and neglect.
Six studies were singled out as most closely meeting the criteria but they too had serious
methodological weaknesses and were descriptive in design. Case resolution among the six
ranged from 22% to 75% of the cases following intervention.
19
Summary and Recommendations
Although violent acts against the elderly by family members, caregivers, and others in whom
they placed their trust is better understood today than 25 years ago when elder abuse was first
recognised as a problem, a more solid knowledge base is needed for policy, planning and
practice purposes. Many aspects of the phenomenon remain unknown: the theoretical
underpinnings, actual prevalence, causes, and consequences. Almost no valid and reliable results
have been generated by research on the effectiveness of interventions. More attention must be
given to primary prevention beginning with a commitment to help bring about a world in which
older persons are allowed to live out their lives in dignity with adequate food, shelter, health care,
and opportunities for self-fulfilment.
For some countries that are facing increasing
impoverishment, the challenge is enormous. Perhaps, the most insidious form of abuse against
elders, however, is the negative attitudes that prevail about older persons and the ageing process,
whether expressed as myths, stereotypes, intergenerational conflict, or the glorification of youth.
As long as older people are devalued and marginalised by society, they risk being subjected to
discrimination by others and robbed of their personhood and self esteem.
The Need for Knowledge
Knowledge about elder abuse is a priority in creating a world-wide response. Since different
countries are at different stages in the development of intervention and prevention programs,
learning more about the status of the problem is a necessary first step. In 1990, the Council of
Europe convened representatives from the European countries to gather information on
definitions, available statistics, legal provisions, supportive and preventive measures, treatment,
general programmes and policies of prevention, and relevant publications and other materials
(81). Using that model, a study group on elder abuse should be formed to gather information
from the countries of the world. This material can be utilized in the creation of a common
definition that applies to both developed and developing countries. It can serve as the foundation
for the development of a minimum data reporting form and the compilation of country/world
data that will confirm the scope and magnitude of the problem, both in domestic and institutional
settings.
Cross-national research should be conducted not only to help explain the role of culture but also
to offer insight into the nature of elder abuse and to suggest methods of successful intervention.
Studies should be undertaken to find out how older persons can take a greater role in defining the
problem and designing programs to prevent it. That effort is underway in Canada but might
have particular relevance in developing countries whose specific traditions regarding ageing and
the family must be honoured and where resources are in short supply.
Another necessary step for increasing knowledge are research studies that meet rigorous
standards. Too much of what has been done in the past has involved small samples and weak
methodology, sometimes resulting in conflicting results.
The abuser's mental state and
substance abuse have been shown to be risk factors in some studies but how these lead to abuse
or neglect in particular situations and not in others has not been investigated. Further work is
needed to resolve the contradictory data about cognitive and physical impairments as risk factors
for elder abuse. Similarly, more research is needed to clarify the role of caregiver stress as a
20
factor in elder abuse which early on was viewed as a primary cause of elder abuse. With the
increasing prevalence of Alzheimer’s disease in the world and the higher prevalence of abusive
behaviour in these families, greater attention should be given to the origins of the relationships
between caregiver and care recipient and how abusive/aggressive acts can be prevented. The
manner in which social isolation or lack of a support system contributes to abuse or neglect may
be obvious. Nevertheless, victims in such situations generally are unwilling to take advantage of
programs that foster social interaction, such as senior centres or day care activities. Studies into
the nature and circumstances of the victims might offer more acceptable solutions.
Perhaps, the facet of elder abuse that has received the least attention is the impact on the victim.
One ingenious study has shown a decidedly negative effect on mortality but how much was due
to the abuse and how much to the "intervention” (adult protective services investigation, etc.) is
not known. Longitudinal studies that track both abused and non-abused persons should be part
of a research agenda. Few studies have looked at the psychological impact. Except for
depression which occurs more often in abuse cases than non-abuse cases, little is known about
the emotional state of the victim.
The role of ageism in leading to abuse of elders has yet to be the subject of research
investigations although some theorists suggest that the marginalisation of elders is a contributory
factor to abuse. Cross-cultural studies might be particularly useful in understanding this effect.
Other societal factors need to be examined. Although they are viewed as major determinants of
elder abuse in the low income countries, there is no research to support the claim. The process
that shows the linkage between these societal factors and elder abuse needs to be elucidated.
Some of the actions in the developing countries are directly abusive such as witchcraft or
abandonment of widows. Other factors such as poverty, modernization, inheritance systems are
indirect causal agents. Using the ecological model to explain elder abuse is a new approach. Its
usefulness as a theoretical construct requires more study.
A variety of intervention models have been developed (e.g., mandatory reporting, protective
service units, social service protocols, emergency shelters, support groups, self-help groups,
consultation teams) but few have been evaluated using an experimental or quasi-experimental
research design. Because questions still remain about the effectiveness of the various
approaches, evaluative research of high standards is a critical need. Generally speaking, the topic
of elder abuse has not attracted the attention of established researchers; yet, their expertise is
needed. Greater investment of funds in elder abuse studies would help to encourage this level of
research.
The Call for Public Policies
A basic requirement across the globe is the establishment of laws to guarantee the human rights
of older persons. Elders as a group should be included within existing laws on domestic or
intrafamily violence. Efforts should be made to ensure that current criminal and civil laws refer
to abuse, neglect, and exploitation of older persons or other violent acts affecting older persons.
A special law on protection of older persons would indicate the nation's commitment to the
problem.
21
As has been described earlier, many existing traditions are abusive towards older women (e.g.,
witchcraft, abandoment of widows, etc.). Working toward the passage of laws to outlaw these
customs will require collaboration among many groups and a long-time commitment. Advocacy
groups consisting of older persons and younger people should be formed at the local, provincial,
and national levels to work toward change. Governmental health and welfare programs will be
necessary to offset the negative impact on old people that has taken place with modernization and
the resulting changes in family life. As life expectancy increases even in the developing
countries, the creation of an adequate pension system is critical.
The Importance of Prevention Strategies
Prevention begins with public awareness. Education, training, and the media are valuable tools
in increasing both professional and public awareness. Human service providers at all levels,
working in the community and in institutional settings, should receive some basic training on
elder abuse detection. Working with the media can be particularly useful as a means of
informing the public and authorities. Although the media tend to concentrate on the most
grievous situations, they have the ability to mobilize governments into action.
A number of
programs can be instrumental in preventing domestic elder abuse or stopping it from occurring in
which elders themselves play a leading role. These include recruiting and training elders to serve
as visitors (companions) to isolated elders; forming support groups for victims; initiating
community programs to stimulate social interaction and community participation of older people;
creating social networks of older persons to promote solidarity and social support among their
peers within villages, neighborhoods, or housing units; and working with older people to create
"self-help" program that will provide them with the chance to be productive.
Helping abusers, particularly adult children, is a more difficult task. Mental health and
substance abuse services should be offered and job and educational opportunities made available.
New methods of conflict resolution may have to be developed to replace the traditional role of
community elders.
Finally, the problem of elder abuse cannot be addressed without at the same time ensuring that
the basic needs of older persons for food, shelter, economic security and access to health care are
met. A world wide climate should be promoted "in which ageing is accepted as a natural part of
the life cycle, the potential richness of late is stressed, anti-ageing attitudes are discouraged, and
an atmosphere amenable to reflection and achieving inner peace is permitted to flourish" (82).
22
Box 1
Permanent Forum Against Abuse of the Older Person
To indicate the kinds of actions presently being taken to combat and prevent elder abuse we focus on the activities of
a Costa Rican organisation, the Permanent Forum Against Abuse of the Older Person..
This organisation was formed in 1991 as a spin-off of a course on holistic care of the elderly offered by the
University of Costa Rica and the training arm of the Social Security Administration (CENDEISSS) for the
Association of Nursing Homes. The 40 participants ranged from administrators to direct service providers and
volunteers. Inspired by the course to take action against the problem of elder abuse, they decided to form this
organisation. Membership in the "Permanent Forum Against Abuse of the Older Person," which was still meeting
regularly in 1996, includes representatives of governmental and non-governmental organisations that have
jurisdiction over or provide services to older persons: the Ministries of Public Health, Justice, and Labor; the welfare
bureau; the Social Security Administration; the geriatric hospital; public and private universities; a large nursing
home; and two social agencies.
The Forum's stated objective (mission statement) is to take actions to prevent the causes of "aggression," whether it
be "domestic, institutional, or social." Given the multiple perceived causes of elder abuse, the Forum’s efforts
encompass a wide range of issues that affect the quality of life of older Costa Ricans and are not focused narrowly on
abuse. In 1995, for example, the Forum wrote to the President of Costa Rica asking that the government enforce
payment of Social Security taxes by employers so that on their retirement workers will not find themselves without
pensions. This letter was also sent to the media where it was aired and published.
Other activities of the Forum over the period 1991-1996 have included disseminating information on elder abuse
through public service announcements on radio and television, organising conferences, publishing pamphlets on care
of elders for family members, protesting negative media portrayals of older people, and, occasionally, receiving and
responding to reports of abuse. On the few occasions when the Forum has tried to get action on reported cases of
abuse, however, they have been unsuccessful. One member of the Forum said that "Intervention doesnt work
because the law is too broad and no one is responsible for enforcing it." The Forum has concluded that the best
course is to place its main emphasis on public education. Under discussion was the possibility of developing and
disseminating an age sensitivity curriculum for the elementary schools. Simultaneously with this focus on education,
however, one of the Forum’s 1996 projects was to urge the president of Costa Rica to designate one branch or office
of the government to be responsible for all issues related to ageing. The feeling was that services for the elderly
were too fragmented, allowing cases of abuse to fall through the cracks.
Source: Gilliland N, Picado LE. Elder abuse in Costa Rica. Journal ofElder Abuse & Neglect, 2000, 12(1)_______
23
Box 2
Emotional Abuse, Deprivation of Freedom and Financial Exploitation
Julia, an Argentinian single woman is 76 years old and has no known relatives, to recover from a bad case of flu, at
the suggestion of Maria, her house porter, she was admitted from her own flat to a private nursing home for a two
week period. As soon as she felt better, she began to ask to return home but was always given "no" as an answer.
Occasionally she would have visitors. She was never left alone and was prevented from leaving the nursing home
even at Christmas time. Almost a year passed, and she remained in the nursing home without any medical reason.
Martha, the nursing home manager, who is a close friend of Maria, took all of Julia’s pension ($600 US) as payment
leaving Julia with no cash.
On behalf of Julia, Rosa, an old neighbour, appealed to the government claiming illegal deprivation of freedom and
emotional abuse. A couple of days later, Julia was visited by a social worker from the Buenos Aires Government
Program on Elder Abuse called "Proteger" who tried to remove her from the nursing home. She found a lot of
opposition from the nursing home owners who also asked that their own physician examine Julia and provide a
written report that she had no signs of physical mistreatment. After a couple of hours had passed and with great
perseverance on the part of the social worker, Julia was able to go home.
Besides the emotional shock and the traumatic situation that she had to endure, Julia found upon return home that her
jewelry and money that she kept in a safety box had been removed. It appears that they had been taken by Maria
who had a key to the flat. The following day, Julia and the social worker went to the police station to report the theft
and the situation. The case is in the court.
Julia, who always was a fairly happy person, independent, polite, educated, and self-confident is now, after being
held against her will, distressed, frightened, and insecure. Currently, she has a home help to assist her with daily
activities, psychological assistance, and a weekly social worker follow-up session.
"Proteger" (to protect), one of the current Programs on Promotion of Social Welfare and Old Age, deals exclusively
with elder abuse situations in the Buenos Aires D.C. area. Workers are given a six month training course on
gerontology with a focus on the prevention of violence and elder abuse. In addition, the program operates a helpline.
24
Box 3
Japan Elder Abuse Prevention Center Help Line Service
In 1993, The Society for the Study of Elder Abuse (SSEA), an independent group that consisted of social workers,
university professors, and a nursing home director, conducted a nation-wide survey of community care support
centres to find out about elder abuse cases. The findings confirmed the existence of elder abuse in Japan and the
need to have the problem addressed by the Health and Welfare Ministry. With information from the survey, which
showed the necessity of taking definitive measures to prevent elder abuse, SSEA concluded that a telephone
counselling service, similar to the one established by Action on Elder Abuse in the UK, was the best way to attack
the elder abuse problem in Japan.
The group organised the Japan Elder Abuse Prevention Centre, a non-profit organisation, and initiated the volunteer
telephone counselling service known as "Help Line" on March 3, 1996. A grant-in-aid from a non government
public organisation provided start-up funds for the telephone installation and office supplies.
Through the
generosity of one of the SSEA members, who directs a nursing home ("Ryokujyu-en"), a room was made available as
an office for the program. In the initial stages, the nursing home also paid for the telephone bill although currently it
is maintained by grants-in-aid from non-govemment public organisations. A notice of the opening of the telephone
counselling service was sent to home services support centres and other consultation agencies in the community. An
article with the telephone number and purpose also appeared in the family section of two major newspapers.
The Help Line Service is provided every Monday from 13:00 to 16:00. It offers information on nursing care, social
services, and since 1998 legal counselling and is available to anyone who has an elder abuse problem including
health, medical, and welfare professionals. In emergency situations, assistance is sought from a public temporary
assistance agency in the neighbourhood. Decisions are based on the mutual consent of all counsellors, and office
work is shared among them.
Initially, the counsellors were restricted to members of SSEA but after the first year, three new volunteers were
added to the staff. Qualifications for serving as a counsellor include professional status in social work, aptitude and
experience as a counsellor, interest and willingness to study elder abuse, and availability as a volunteer. One or two
counsellors are assigned to a particular day. An telephone answering service gives the available service hours during
off hours.
The guidelines for the program set by SSEA state that it is a telephone service only. If a caller wants face-to-face
counselling, the case will be transferred to a home service support centre in that person's neighbourhood.
Counsellors are not assigned to particular cases but take calls as they come in. If the caller asks help line support
frequently, he or she is informed that the responding counsellor might vary each session. The counselling service
call is limited to one hour. An emphasis is placed on maintaining the privacy and anonymity of the callers and the
confidentiality of the files.
Training for new counsellors include basic policies and procedures for counselling, information about social welfare
services, and knowledge about elder abuse, especially for ways to cope with the problem, as well as how to record a
case and how to deal with emergency situations. Usually a trainee will "sit" with a counsellor and observe for a few
months; then take calls with assistance from the more established counsellors, especially in situations where the
caller seemed to be emotionally upset. All counsellors attend a monthly SSEA meeting in which information about
elder abuse is exchanged, technical papers and case studies are reviewed, and supervision is provided. . Bi-monthly
meetings are held for discussion of cases with one of the SSEA members serves as a specialist in the case. Other
professionals may be called upon with regard to the case studies.
Source: Yamada Y. A telephone counseling program for elder abuse in Japan. Journal of Elder Abuse & Neglect,
1999, 1 l(l):105-l 12.
25
Box 4
Influencing The Future:
An Intergeneratonal Curriculum On Elder Abuse
Sensitising children to old age and providing opportunities for intergenerational relationships may be critical factors
in the prevention of elder abuse and neglect. By including lessons on aging in school curricula, children may
develop an increased respect for seniors, may be less inclined to mistreat older people and may even attempt to
prevent others from doing so.
The recognition of the need to raise awareness among children and adolescents resulted in a two-phase
intergenerational education project funded by Health Canada. The first initiative is an early childhood education
interactive storytelling kit designed to bring old and young together with games and stories for children three to
seven. Although elder abuse is not addressed directly, the book fosters positive imaging of old age and is replete
with the wisdom of older people. This book has also been proven to be effective when used with older children who
have limited English literacy skills. Seniors, youth and members of the community at large link with each other to
discuss the subject of elder abuse and how they can collectively, as a community, can reduce or eliminate abuse in
general and elder abuse in particular.
The primary goal of the formal curriculum, the second initiative, is to change the deep-seated negative societal
attitudes and beliefs about aging and to decrease the incidents of elder abuse. The content was based on learning
theories and is developmentally appropriate for youth in their transition years. Before the creative strategy was
developed, a needs assessment was completed with numerous consultation and focus groups consisting of: seniors,
teachers, youth, faith communities, health care providers, multi-cultural health care providers, academics, volunteers,
police force who work with seniors and deal with seniors’ issues, intergenerational programmers and recreationalists.
Pre-testing of the program was done to ensure language appropriateness for use by teens, adolescents, and seniors
from a wide range of multi-cultural backgrounds. Curriculum content on aging and elder abuse was incorporated
into relevant class studies, such as health, social sciences, family-life education and personal safety. Ontario schools
have included conflict resolution into the school curriculum and teachers find that elder abuse concepts fit in with
this material. The educational program provides background and planning information as well as direction on how
to link seniors and youth.
The curriculum model contains 15 lessons/sessions: A Study of Our Canadian Society; Your World, My World, Our
World; Power and How Power Can Be Used to Influence The Lives of Others; The Link Between Power and
Potential Abuse; The Impact of the Media on Our Society; The Link Between Media Messages and How
Stereotyping Can Lead to Possible Abuse of Older Persons (Elder Abuse); Types of Abuse of Older Persons, The
Factors That May Lead to Each and the Prevalence of Abuse in Our Society; Abuse of Older Persons: The Hidden
Crime; Steps That Can Be Taken to Lessen Abuse of Older Persons; Community, Planned Community Action; The
Community In Action; Bringing Closure to the Program; and, Reflection.
The curriculum includes a drama kit consisting of a video and two plays, which can be performed for both stage and
radio. The plays address the misconceptions that different generations may have about each other as well as the
various forms of elder abuse. The 30-minute video (ages 12 - 16) is a story of betrayal and love as seen through the
eyes of a 15-year-old girl. The video portrays the healing of generational estrangement and ultimately the timeless
and priceless value of grandmothers. Through the young girls’ narrative, the causes of - and solutions to - elder abuse
are explored.
These educational tools have been used successfully outside of the school: in nursing homes, at seniors’ events, at
summer camps and youth conferences. The outcomes of the curriculum, which was initiated in 1999 are being
evaluated and documented. The evaluation focuses on determining how well the objectives are being met, analysing
the problems and identifying alternative changes and revisions. Before initiating elder abuse concepts it is critical to
26
prepare teachers through curriculum development workshops to increase their knowledge and confidence in
addressing elder abuse content.
Educating children about elder abuse and neglect must be recognised by the larger community as a necessary
prevention strategy. Educational curricula for children can not only foster positive attitudes and concern for the
safety and well-being of older people, they can also contribute to increased intergenerational interaction.
Prepared by Elizabeth Podnieks, PhD, Ryerson University, Toronto, Ontario, Canada
27
Box 5
Witchcraft: A Violent Threat
In Tanzania, an estimated 500 women are murdered very year after being accused of witchcraft. The problem is
particularly acute in the Sukumaland area of northern Tanzania. In Shinyanga region, 178 people were killed as a
result of being accused of witchcraft in the 18 months from January 1996 to June 1997, according to a study by the
Tanzania Media Women’s Association, citing data from the Ministry of Home Affairs. The murders represented
nearly 40 percent of all murders reported in the region in that period.
The killings are only part of the story. Many more women are driven from their homes and communities for fear of
being accused of witchcraft, living in destitution in urban areas, nearly a third of those interviewed in the study were
afraid of getting old because of the fear of being attacked as a witch.
Although belief in witchcraft goes back centuries, the violence has risen steeply in recent years. Research by
HelpAge International points to social and economic problems including poverty, pressure on land, inadequate or
inaccessible health services and poor education as the underlying causes, people need explanations for misfortunes
such as illnesses and death, crop failures and dried-up wells, and they look for a scape-goat.
Traditional healers often make accusations of witchcraft at the suggestion of the accused person's family or
neighbours. They may blame witchcraft to explain events that they cannot understand or control. For example, one
young boy killed his mother after a traditional healer told him that she was the cause of his problems.
Land disputes are a common underlying cause of violence against widows. According to inheritance laws, widows
may remain on their husband's land, but they do not own the property. When they die, the land becomes the property
of their husband's sons. Although this system is intended to protect women, it is sometime used to get rid of
troublesome "tenants." And it can be a lucrative income for hired killers.
Women’s low status
Although men may be accused of witchcraft, women's circumstances and their low status in society make them more
vulnerable to attack. Women usually live longer than men and marry men older than themselves, many are therefore
alone in old age, often living in poor housing, making them an easy target, male domination of policing and
counselling systems also mean that older women receive little protection.
Myths about the physical appearance of witches increase women's vulnerability. For example, many older women
are believed to be witches because they have red eyes-people do not know that red eyes can result from a lifetime of
cooking over a smoky stove or from conditions such as conjunctivitis.
Call for action
Community leaders are calling on the government to take strong measures. A former party secretary said: 'It is a
question of educating the people. In other areas of the country where people are better educated, we don't face this
problem.'
Until recently, the government of Tanzania was reluctant to acknowledge that witchcraft beliefs still existed.
However, witchcraft is now being widely discussed and official condemned. In 1999, the Tanzania government
made witchcraft the theme for International Women's Day.
Steps to improve the security of older women are being taken by a Tanzanian NGO, Magu Poverty Focus on Older
People Rehabilitation Centre (MAPERECE) and HelpAge International programme in Tanzania, through a project
that combines both practical and advocacy work. The project will involve older people's organisations, churches.
28
schools and other groups to explore and change attitudes to witchcraft-related violence, and address practical
problems such as poor housing and use of inefficient smoky stoves.
Source: Reprint of article in HelpAge International. Aging and Development New, No. 6, July 2000.
29
REFERENCES
1. Baker AA. Granny-battering. Modern Geriatrics, 1975, 5(8):20-24
2. Burston GR. 'Granny battering'. British Medical Journal, 1975, 6 September:592
3. WHO, 1995.
4. HelpAge International. The ageing & development report: A summary. London: HelpAge
International, 1999.
5. Apt N. Ageing in Africa. Report prepared for the World Health Organization, Geneva, 1997.
6. UNAIDS Program report. The nation's health, August 2000.
7. Hudson MF. Elder mistreatment: A taxonomy with definitions by Delphi. Journal of Elder
Abuse &Neglect,199, 3(2):l-20.
8. Brown AS. A survey on elder abuse in one Native American tribe. Journal of Elder Abuse
& Neglect, 1989, 1(2): 17-37.
9. Maxwell EK, Maxwell RJ. Insults to the body civil: Mistreatment of elderly in two Plains
Indian tribes. Journal of Cross-Cultural Gerontology, 1992, 7:3-22.
10. Action on elder abuse bulletin, May-June 1995, No. 11
11. INPEA. Newsletter, No.2, 1999.
12. Kosberg JI, Garcia JL. Common andunique themes on elder abuse from a world-wide
perspective. In Kosberg JI, Garcia JL (eds.) Elder abuse: International and cross-cultural
perspectives. Binhamton, NY: Haworth Press, 1995.
13. Kwan AY. The gray wave, unpublished manuscript, 2000.
14. Moon A, Williams O. Perceptions of elder abuse and help-seeing patterns among African
American, Caucasian American, and Korean American elderly women. The Gerontologist,
1993, 33(3): 386-395
15. Tomita SK. Exploration of elder mistreatment among the Japanese. In Tatara T (ed.)
Understanding elder abuse in minority populations. Philadelphia: Francis & Taylor, 1999:
119-139.
16. Gilliland N, Picado LE. Elder abuse in Costa Rica. Journal of Elder Abuse & Neglect,
2000, 12(l):73-87.
17. Agyarko R. Personal communication, 2000.
30
18. Gorman M, Petersen T. Violence against older people and its health consequences:
Experience from Africa and Asia. London: HelpAge International, 1999.
19. HelpAge International. Ageing and development news, July 2000, 6.
20. Keikelame J, Ferreira M. Mpathekombi, ya bantu abadala: Elder abuse in black townships
on the Cape Flats. HSRC/UCT Centre for Gerontology, March 2000.
21. Pillemer K, Finkelhor D. Prevalence of elder abuse:
Gerontologist, 1988, 28(l):51-57.
A random sample survey.
The
22. Podnieks E. National survey on abuse of the elderly in Canada. Journal of Elder Abuse &
Neglect, 1992,4(1/2): 5-58.
23. Kivela SL, Kongas-Saviaro P, Kesti E, Pahkala K, Ijas ML. Abuse in old age:
Epidemiological data from Finland. Journal ofElder Abuse &. Neglect, 1992, 4(3): 1-18.
24. Ogg J, Bennett GCJ. Elder abuse in Britain. British Medical Journal, 1992, 305:998-999.
25. Comijs HC, Pot AM, Smit JH, Bouter LM, Jouter C. Elder abuse in the community:
Prevalence and consequences. Journal of the American Geriatrics Society, 1998, 46:885888.
26. Straus MA, Gelles RJ. Physical violence in American families. 2nd e New Brunswick, NJ:
Transaction, 1990.
27. Crowell, N.A. & Burgess, A.W. (1996).
Washington, DC: National Academy Press.
Understanding violence against women.
28. Canadian Centre for Justice Statistics. Family violence in Canada: A statistical profile 2000.
Ottawa: Health Canada.
29. Kane RL, Kane RA. Long-term care in six countries: Implications for the United States.
Washington, DC: United States Department of Health, Education and Welfare, 1976.
30. Pillemer KA, Moore D. Highlights from a study of abuse of patients in nursing homes.
Journal ofElder Abuse & Neglect, 1990,2( 1 /2):5-30.
31.Garbarino J, Crouter A. Defining the community context for parent-child relations: The
correlates of child maltreatment. Child Development, 1978, 49:604-616.
32. Garbarino J, Kostelny K. Child maltreatment as a community problem. Child Abuse &
Neglect, 1992, 16:455-464.
33. Schiamberg LB, Gans D. An ecological framework for contextual risk factors in elder abuse
by adult children Journal ofElder Abuse & Neglect, 1999, 11 (1 ):79-103.
31
34. Carp RM. Elder abuse in the family: An interdisciplinary model for research. New York:
Springer, 2000.
35. Chalk R, King PA. Violence in families: Assessing prevention and treatment programs.
Washington, DC: National Academy Press, 1998.
36. Gelles RJ. Through a sociological lens: Social structure and family violence. In: Gelles RJ,
Loeske DR, eds. Current controversies on family violence. Newbury Park: Sage, 1993:3146.
37. O’Leary KD. Through a psychological lens: personality traints, personality disorders, and
levels of violence. In Gelles RJ, Loeske DR, eds. Current controversies on family violence.
Newbury Park: Sage, 1993:7-30.
38. Wolf RS, Pillemer KA. Helping elderly victims: The reality of elder abuse. New York:
Columbia University Press, 1989.
39. Homer AC, Gilleard C. Abuse of elderly people by their carers. British Medical Journal,
1990,301:1359-1362.
40. Bristowe E, Collins JB. Family mediated abuse of non-institutionalised elder men and
women living in British Columbia. Journal ofElder Abuse & Neglect, 1989, l(l):45-54.
41. Pillemer KA. Risk factors in elder abuse: Results from a case control study. In: Pillemer K,
Wolf RS, eds. Elder abuse: Conflict in the family. Dover, MA: Auburn House, 1989:239264.
42. Paveza GJ, Cohen D, Eisdorfer C, et al. Severe family violence and Alzheimer’s Disease:
Prevalence and risk factors. The Gerontologist, 1992, 32(4): 493-497.
43. Cooney C, Mortimer A: Elder abuse and dementia - A pilot study. International Journal of
Social Psychiatry, 1995, 41(4):276-283.
44. Aitken L, Griffin G. Gender issues in elder abuse. London: Sage, 1996,
45. Harbison J. The changing career of 'elder abuse' as a social problem in Canada: Learning
from feminist frameworks? Journal ofElder Abuse & Neglect, 11(4):59-80.
46. Steinmetz SK. Duty Bound: Elder abuse andfamily care. Newbury, CA: Sage, 1988.
47. Eastman M. Old age abuse. London: Age Concern England/Chapman and Hall.
48. Reis M, Nahamish D. Validation of the indicators of abuse (IOA) screen. The Gerontologist,
1998, 38(4):471-480.
49. Hamel M, Gold PD, Andres D, et al. Predictors and consequences of aggressive behavior by
community-based dementia patients. The Gerontologist, 1990, 30(2):206-211.
32
50. Nolan M. Carer-dependent relationships and the prevention of elder abuse. ImDecalmer P,
Glendenning F, eds. The mistreatment of elderly people, 1997:148-158.
51. Pillemer KA, Suitor JJ. Violence and violent feelings: What causes them among family
caregivers? Journal of Gerontology, 1992, 47(4): S165-S172.
52. O’Loughlin A. Abuse, neglect and mistreatment of older people: An exploratory study.
National Council on Ageing and Older People, Report No. 52, Dublin, Ireland.
53. Kosberg JI, Nahmiash D. Characteristics of victims and perpetrators and milieus of abuse
and neglect. In Baumhover LA, Beall, SC (eds) Abuse, neglect, and exploitation of older
pesons: Strategies for assessment and intervention. Baltimore: Health Professions Press,
1996.
54. Phillips LR. Theoretical explanations of elder abuse. In: Pillemer K, Wolf RS, eds. Elder
abuse: Conflict in the family. Dover, MA: Auburn House, 1989:197-217.
55. Grafstrom M, Nordberg A, Winblad B. Abuse is in the eye of the beholder. Scandinavian
Journal ofSocial Medicine, 1994, 21(4): 247-255.
56. (anonymous reviewe's comments).
57. Kwan AY. Elder abuse in Hong Kong: A new family problem for the east? In Kosberg JI,
Garcia JL (eds) Elder abuse: International and cross-cultural perspectives. Binghamton,
NY: Haworth, 1995.
58. Phillips LR. Abuse and neglect of the frail elderly at home: An exploration of theoretical
relationships. AdvancedNursing, 1983, 8:379-382.
59. Pillemer KA, Prescott D. Psychological effects of elder abuse: A research note. Journal of
Elder Abuse & Neglect, 1989, l(l):65-74.
60. Booth BK, Bruno AA, Marin R. Psychological therapy with abused and neglected patients.
In Baumhover LA, Beall SC, eds. Abuse, neglect, and exploitation of older persons:
Strategies for assessment and intervention. Baltimore, MD: Health Professions Press, 1996.
61. Goldstein M. Elder mistreatment and PTSD. In Ruskin PE, Talbott JA, eds. Aging and
posttraumatic stress disorder. Washington, DC:
American Psychiatric Association,
1996:126-135.
62. National Institute on Aging. Established populations for epidemiologic study of the elderly:
Resource data book. Rockville, MD: NLA, 1986.
63. Lachs MS, Williams CS, O’Brien S, Pillemer KA, Charlson ME. The mortality of elder
mistreatment. Journal of the American Medical Association, 1998, 280(5):428-432.
33
64. Bennett G, Kingston P, Penhale B. The dimensions of elder abuse: Perspectives for
practitioners. London: Macmillan, 1997.
65. Harrington CH, Carrillo H, Thollaug SC, Summers PR. Nursingfacilities, staffing, residents,
andfacility deficiencies, 1991-1997. San Francisco: Dept, of Social and Behavioral Sciences,
Univesity of California, 2000.
66. Clough R. Scandals in residential care: A report to the Wagner Committee. Unpublished
report. 1988.
67. Leroux TG, Petrunik M. The construction of elder abuse as a social problem: A Canadian
Perspective. International Journal ofHealth Services, 1990, 20:651-663.
68. Bennett GCJ, Kingston P. Elder abuse:
Chapman & Hall, 1993.
Concepts, theories and interventions. London:
69. Blumer H. Social problems as collective behaviour. Social Problems, 1971,18(3):298-306.
70. Eckley SCA, Vilakas PAC. Elder abuse in South Africa. In: Kosberg JI, Garcia JL, eds.
Elder abuse: International and cross-cultural perspectives. Binghamton, NY: Haworth
Press, 1995:171-182
71. South African Development Countries. Summary Report: SADC Conference on the
prevention of violence against women.
72. Action on Elder abuse. Hearing the despair: the reality of elder abuse. London: AEA
Family Policy Studies Centre.
73. Yamada Y. A telephone counseling program for elder abuse in Japan. Journal of Elder
Abuse & Neglect, 1999, 11(1): 105-112.
74. Checkoway B. Empowering the elderly: Gerontological health promotion in Latin America.
Ageing and Society, 1994, 14:75-95.
75. Sanders AB.
Care of the elderly in emergency departments: Conclusions and
recommendations. Annals ofEmergency Medicine, 1992, 21(7):79-83.
76. Lachs MS, Pillemer KA. Abuse and neglect of elderly persons. New England Journal of
Medicine, 1995, 332(7):437-443.
77. Jones JS. Geriatric abuse and neglect. In: Bosker G, Schwarz GR, Jone JS, Sequeira M, eds.
Geriatric emergency medicine. St. Louis, MO: Mosby, 1990:
78. Mount Sinai/Victim Services Agency Elder Abuse Project. Elder mistreatment guidelines:
Detection, assessment and intervention. New York, 1988.
79. Wolf RS. Elder abuse: Mandatory reporting revisited. In: Cebik LE, Graber GC, Marsh FH,
eds. Violence, neglect, and the elderly. Greenwich, CT: JAI Press, 1996: 155-170.
34
80. Bolan G, Ploeg J, Hutchinson B. A systematic overview of the effectiveness of interventions
for elder abuse. Hamilton, Ontario, Canada: McMaster University.
81. Steering Committee on Social Policy. Violence against elderly people. Strasbourg: Council
of Europe, 1991.
82. Ageing Assembly Ageing Report submitted to the Millennium Forum. Washington, DC:
AARP., 1999.
35
Chapter 6
Sexual Violence
Status:
Draft: Outline
Peer Reviewed: No
Date of Outline: 16 July 2000
Table of Contents
I.
Introduction
IL
Concepts and Definitions
A. What is Sexual Violence?
B. How has it been Defined?
C. Working Definition for the Chapter
III.
Forms and Contexts of Sexual Violence
A. Child Sexual Abuse
B. Marital Rape
C. Non-marital Sexual Violence
1. Date/acquaintance
2. Stranger
3. Sexual harassment
4. Female genital mutilation
5. Prostitution and trafficking
6. Rape during conflict or war
D. Contexts
1. Home
2. Community
3. Work place
4. Other institutions
IV.
Magnitude of the Problem
A. Prevalence Estimates
B. Trends and Patterns of Sexual Violence
C. Impacts (Physical, Sexual, Emotional, Psychological, Reproductive)
V.
Risk and Protective Factors
A. Vulnerability
B. Resiliency
VI.
Responses to Sexual Violence
A. Prevention
B. Policy Responses
C. Effectiveness of Responses
VII.
Recommendations
A. Data
B. Research
C. Prevention and Policy
VIII.
Conclusion
2
Chapter 7
Self-Directed Violence
Status:
Draft: 2nd
Peer Reviewed: No
Date of Current Draft: 16 October 2000
Table of Contents
I.
IL
III.
IV.
V.
VI.
Introduction
Definitions
Epidemiology of Suicidal Behaviour
A.
Fatal Suicidal Behaviour
1. Cautions in the use of suicide data
B.
Non-Fatal Suicidal Behaviour and Suicidal Ideation
Risk Factors for Suicidal Behaviour
A.
Psychiatric Risk Factors
B.
Biological and Medical Markers
Relationship Factors
C.
D.
Social and Environmental Factors
Response to the Problem: Prevention, Policy Responses
A.
Psychiatric and Psychological Factors
1. Treatment of mental disorders
2. Treatment of suicidal behaviours
B.
Behavioural Interventions
Green Card
C.
Biological Factors
D.
1. Pharmocotherapy
Relationship Factors
F.
1. Psychosocial interventions
Social and Environmental Factors
G.
1. Suicide prevention centers
2. School-based interventions
3. Restriction of access to means of suicide
4. Media reporting
Intervention after a Suicide: Postvention
H.
Policy Responses
I.
Recommendations
Box 1: Suicide in China
Box 2: Suicide Among the Inuit
Box 3: Aboriginal Suicide in Australia
Box 4: Depression and Suicide
Box 5: Alcohol and Suicidal Behaviour
Box 6: Suicidal Behaviour in Children and Young People
Box 7: Access to Means of Suicide
2
Introduction
It has been estimated that in the year 2000 approximately one million people died from suicide,
worldwide (1). This represents a global mortality rate of about 16 per 100,000 (Fig. 1) or one
death every 40 seconds. Suicide is estimated to constitute 2.3% of the total burden of disease (2).
Moreover, suicide represents the 12th leading cause of death in the world overall, but among
those of 15-44 years of age self-inflicted injuries are the fourth leading cause of death worldwide
and the sixth leading cause of ill health and disability (3).
Suicide deaths represent only a portion of the magnitude of the problem. In addition to the
number of suicidal deaths, many more persons make non-fatal attempts to take their lives or harm
themselves, often seriously enough to require medical attention (4). Furthermore, for every
suicidal death there are many other people who survive them whose lives are profoundly affected
emotionally, socially, and economically. Each year, the economic costs associated with selfinflicted injuries are estimated to be in the billions of dollars (5).
This chapter describes the international epidemiology of fatal and non-fatal suicidal behaviour,
the factors that increase the risk of suicide, various preventative and policy responses, and
recommendations for future reseach and intervention. The text is enriched by a number of boxes
which illustrate themes or aspects of the suicide phenomenon of particular relevance at the
present time.
Definitions
Suicidal behavior ranges in severity from merely thinking about self-destruction (suicidal
ideation), to developing a plan and obtaining the means to commit suicide, attempting to take
one’s own life (commonly called “attempted suicide”, “parasuicide” or “deliberate self-harm”),
and finally to completing the final act (“completed suicide”).
The term ’’suicide” in itself evokes direct reference to violence and aggressiveness. Apparently,
Sir Thomas Browne was the first to coin the word "suicide" in his Religio Medici (1642). A
physician and a philosopher, Browne based the word on the Latin sui (of oneself) and caedes
(murder). The new term reflected a desire to distinguish between the homicide of oneself and the
killing of another (6). In fact, the original meaning of ’’self-murder" is still a popular choice to
designate suicide in some northern European countries, such as Germany, where selbstmord is
used more commonly than "suizide".
A frequently quoted definition of suicide is the one reported by Shneidman in the 1973 edition of
the Encyclopaedia Britannica: "The human act of self-inflicting one’s own life cessation” (7).
Based on that (the basic principle of the mors voluntaris), intention to die is a key-issue in the
attribution of a cause of death to suicide. However, the reconstruction of the intentions and
desires of the person prior to death is often difficult to accomplish unless the person made clear
statements relevant to intent prior to death or left a suicide note. Having in mind that not all the
subjects who survive a suicidal act intend to live, nor all suicidal deaths are planned, the
correlation between intent and outcome may be very problematic. In many jurisdictions, a death
3
is certified as suicide if the circumstances are consistent with suicide and if murder, accidental
death and natural causes can be ruled out.
There has been much disagreement about the terminology that is best for describing suicidal
behaviour. Recently, the outcome-based terms of “fatal suicidal behaviour” has been proposed
for suicidal acts that result in death and “non-fatal suicidal behaviour” for suicidal actions that do
not result in death (8). The latter term replaces the terms “attempted suicide” (a term which is
popular in the USA), “parasuicide” and “deliberate self-harm” (terms which are popular in
Europe).
Suicidal ideation refers to thoughts of killing one’s self, featuring varying degrees of intensity
and elaboration. The available literature on the topic also includes feelings of tiredness for life,
that life is not worth living, and the wish for not waking up from sleep (9, 10). Although these
different ideations express varying levels of severity, this does not imply any particular
continuum among them. Moreover, the intention to die is not a necessary criterion for non-fatal
suicidal behavior.
Another common form of self-directed violence is self-mutilation. This represents the direct and
deliberate destruction or alteration of parts of the body without any conscious suicidal intention.
Favazza (11) proposes three main categories: major self-mutilation (self-blinding, amputations
of fingers, hands, arms, limbs, feet, genitalia, etc); stereotypical self-mutilation (head banging,
self-biting, arm hitting, throat and eye gouging, scrathcing, hair pulling, etc); superficial-tomoderate self-mutilation ( skin cutting, scratching and burning, needle sticking, compulsive hair
pulling). Self-mutilation involves very dfferent meanings from suicidal behaviours, and will not
be discussed further. For an extensive review of self-mutilation, see Favazza (11).
Epidemiology of Suicidal Behaviour
Fatal Suicidal Behavior
Reliable information on suicide mortality data can be accessed through a number of agencies
around the world. The World Health Organisation provides mortality data on suicide starting
from 1950, by age and gender (web site: www.who.int). Other agencies that may provide
information are the World Bank, UNICRI, INTERPOL, EUROSTAT, UNICEF, ISPCAN,
UNIFEM, ICE and INCLEN.
A number of governmental agencies, national associations and voluntary organizations may also
provide information; the Centers for Disease Control, USA, the Swedish National Centre for
Suicide Research and Prevention, the Australian Institute for Suicide Research and Prevention are
some examples of these. The International Association for Suicide Prevention, the International
Academy for Suicide Research, the American Association of Suicidology, AusEinet have their
own web sites which can be accessed for information. Normally, the most recent suicide
mortality data available from these agencies refer to a period up to between 18-36 (or more)
months back, depending on the country in question.
4
National suicide rates vary considerably (Tab.l). Among countries reporting suicide, the highest
suicide rates are found in Eastern European countries (e.g. Lithuania, 48.2; Russia, 41.5) and
Asian countries (e.g. China, 16.1; Japan, 17.9). The lowest rates are found mostly in Latin
American, Arabic and some Asian countries (e.g. Argentina, 6.6; Brazil,3.5; Kuwait, 1.8;
Thailand, 4). Countries in Central and Northern European, North American, South East Asia and
Western Pacific countries fall somewhere in between these extremes (e.g. Australia, 13; Canada,
13.4; India, 9.7; New Zealand, 14.9; USA, 11.8). Unfortunately, there is little information on
suicide in most African countries (12) (Figure 2).
For the two nations with suicide rates available from the 1700s (Finland and Sweden), the trend
has been for the suicide rate to increase over time (13). For the 1900s, 7 nations had a significant
increase (Finland, Ireland, the Netherlands, Norway, Scotland, Spain and Sweden), four a
significant decrease (England/Wales, Italy, New Zealand and Switzerland) and one no change
(Australia) (13). For the period 1960-1990, 28 nations experienced a rising suicide rate (including
Bulgaria, Costa Rica, Mauritius, Singapore, and Taiwan) while eight experienced a declining
suicide rate (including Australia and England/Wales) (13).
Suicide rates in the general population are not distributed equally. One of the important
demographic markers of suicide risk is age. Cross-nationally, suicide rates tend to increase with
age, although some countries such as Canada have recently experienced a secondary peak in
youth (15-24 years of age). Globally speaking, as recently as 1995, starting from 0.9 per 100,000
in the age group 5-14 years old, suicide rates gradually increased up to 66.9 per 100,000 in the
age group 75+ years (see Fig. 3). In general, rates among those 75 years and older are
approximately three times higher than those of youth under 25 years of age. This trend is
observed for both sexes, and it is steeper for males. In females, suicide rates actually present
several distinct patterns. In some nations, female suicide rates rise with age, in other female rates
peak in middle age while, particularly in developing nations and minority groups, female suicide
rates peak in young adults (14).
Although suicide rates are largely higher among the population over 60 years of age than in any
other age group, given the demographic distribution, the actual number of cases of suicide is
higher among people less than 45 years of age than in people over 45 years of age (see Tab. 2 and
Fig. 4). This is a remarkable change in relation to just 50 years ago - when the number of cases of
suicide was proportional to age - and is not explained in terms of the overall aging of the global
population; in fact, it goes against this demographic transition. Currently, suicide rates are
already higher among younger people in approximately one third of all countries, irrespectively
of continent, level of industrialization or wealth. Examples of countries in which current suicide
rates (and absolute number of cases) are higher among those below 45 years of age include
Australia, Bahrain, Canada, Colombia, Ecuador, Guyana, Kuwait, Mauritius, New Zealand, Sri
Lanka and UK.
Gender, race, and ethnicity are also important factors in the epidemiology of suicide. In terms of
sex distribution, suicide rates are higher among men. Between 1950 and 1995 the male: female
ratio varied from 3.2:1 to 3.6:1. The only exception to this predominance is observed in rural
China, where the ratio of male:female suicide is 0.8:1 (see Box 1). However, the ratio of
male/female suicide seems rather influenced by the cultural context, going from 1.3 to 1 of India
to more than 5 to 1 in several of the former Soviet Union countries, up to 8 to 1 in Puerto Rico.
5
On average, it seems that men commit suicide about 3 times more frequently than women, with
substantial consistency throughout different periods of life, with the exception of extreme
advanced age in which men tend to present even higher rates. In general, the sex difference in
suicide rates is smaller in Asian countries than elsewhere in the world. These differences between
countries and by sex indicate the importance of each country monitoring its own epidemiological
trends in order to ascertain which groups of its citizens are most at risk for suicide.
In terms of race/ethnicity, the prevalence of suicide in Caucasians is approximately double that of
other ethnicities, although a trend towards an increase in African-American people has recently
been reported (4). In the USA about 2 out of 3 suicides are committed by Caucasians. Also in
Zimbabwe and South Africa suicide rates are higher in whites than in blacks (15). Suicide rates
are higher in many aboriginal groups, for example in some Native American and Native
Canadian tribes (16) and in aboriginal groups in Taiwan (17) and Australia (18) (See Boxes 2 and
3). Data on suicide in aboriginal groups in other countries are unavailable.
Cautions in the use of suicide data. Comparison of different countries with respect to suicides is
often performed, but it must be borne in mind that the recording of mortality procedures varies
greatly amongst countries, seriously affecting comparability. Even in those countries which adopt
standardised criteria, such as Australia, the application of these criteria may vary considerably. A
recent survey on suicide coding reliability of 20 western countries demonstrated percentages of
sensitivity of suicide certification ranging from 89.6% (Austria) to 51.7% (Greece)(19). A WHO
Working Group on Suicide Prevention Practices (20) advised on the appropriate way of making
comparisons by using national statistics only for trend analysis.
Suicide rates are expressed as number of suicidal deaths per 100,000 population. If reported rates
refer to small populations (cities, provinces or even countries with small population) their
interpretation requires extra caution since just a few deaths may greatly influence their
representation. Normally, for populations under 250,000 crude suicide number are expected to be
mentioned. Some rates may be reported as age-standardised and this can exclude suicides under
15 years of age because of small numbers usually characterising this age group [but in many
countries there is an alarming increase in suicides in this age group, including Italy (21)].
Furhermore, suicide mortality data usually carry an underestimation of their real number. Many
different issues contribute to this. Suicide data are the end result of a chain of informants which
involves those finding the body (eg, family members), doctors, police, coroners and statisticians.
Any of these individuals, for a variety of reasons, may be reluctant to call a death a suicide. This
is thought to be particularly true in those regions where religious and cultural attitudes condemn
suicide. However, Cooper and Milroy (22) have found an undercount of 40% in official records
of suicide in regions in England. In general, a suicide may be voluntarily hidden to avoid
stigmatisation, for social convenience, for political reasons, to benefit from insurances policies,
or because it was masked as an accident (eg, a road accident). Suicide can also be misclassified as
an undetermined cause of death, or as natural cause (for example when people -especially in the
elderly- neglect to take life-sustaining medicaments). Lozano (23) reported that in Chile the
number of injurious deaths classified as “undetermined” is of the order of 45% and that 1 out of 4
of these deaths actually reveals to be a case of suicide.
6
Suicide can also go officially unrecognized when people overdose as drug-abusers, as are in self
starvation situations [so called “suicidal erosions”(24)J, or when people die some time after their
suicide attempt (in these cases usually it is the clinical cause of death which is the one officially
reported), or in case of euthanasia or assisted suicide. The probability of under-recognition is also
related to the age of the person, with underreporting generally much more prevalent in elderly
people. Despite this and the other motives above mentioned, it has been argued that the relative
ranking of national suicide rates is reasonably valid. For example, Sainsbury and Barraclough
(25) found that suicide rates of immigrants from other nations to the USA were in roughly the
same order as the suicide rates in their home nations.
Non-Fatal Suicidal Behaviour and Suicidal Ideation
Relatively few countries have reliable data on non-fatal suicidal behaviour. The main reason for
that resides in the difficulty of collecting information. Only a minority of attempters present to
health facilities for medical attention. It has been calculated that, on average, only about 25% of
subjects with suicidal acts make contact with public hospitals (which may represent the easiest
source for data collection) (26) and they do not necessarily represent the most serious ones. This
aspect is known as "tip of the iceberg phenomenon", underlining that the large majority of
suicidal people remain unnoticed (27). Several institutions, like national centres for injury control
and prevention or department of statistics (and of justice, in several countries) keep records of
these non-fatal events that are registered at health services. They represent useful data for
research and preventative efforts, since suicide attempters constitute a high risk group for
subsequent suicidal behaviour, both fatal and non-fatal. Public health officials also rely on
reviews of hospital records, population surveys, and special studies. These sources often include
data that is lacking in mortality data systems.
Available data indicate that, both in absolute and relative numbers, non-fatal suicidal behavior is
higher in younger people than among older people. It is estimated that the ratio of fatal versus
non-fatal suicidal behaviour in old age (i.e., those over the age of 65 years) may be of the order of
1 to 2-3, while in the young people (those less than 25 years of age) this may reach the level of 1
to 100-200 (28, 29/ Although suicidal behaviour is less frequent in the elderly, the probability of
a fatal outcome is much higher (24, 30). On average, suicide attempts in old age are, in
psychological and medical terms, more serious and the “failure” of a suicidal action is often due
to unpredictable and fortuitous circumstances. As a general trend, non-fatal suicidal behaviour
also tends to be 2-3 times higher in women than in men. Finland, however, represents a
remarkable exception to that (31).
Data from an ongoing, cross-national study of non-fatal suicidal behavior in 13 countries, reveal
that for the period 1989-1992 the highest average male age-standardised rate of suicide attempts
was found for Helsinki, Finland (314/100,000), and the lowest rate (45/100,000) was for
Guipuzcoa, Spain, representing a seven-fold difference (31). The highest average female agestandardised rate was found for Cergy-Pontoise, France (462/100,000), and the lowest
(69/100,000) again for Guipzucoa. With only one exception (Helsinki), the person-based suicide
attempt rates were higher among women than among men. In the majority of centres, the highest
person-based rates were found in the younger age groups. The rates amongst people aged 55
years and over were generally the lowest. The methods used were largely poisoning and then
cutting. More than 50% of the suicide attempters made more than one attempt, and nearly 20% of
7
the second attempts were made within 12 months after the first attempt.
Data from a longitudinal, nationally representative sample of nearly 10,000 adolescents aged 1220 years in Norway indicated that 8% had ever attempted suicide and 2.7% made an attempt over
the two-year study period. Logistic regression analyses showed that future attempts were
predicted by previous attempts, female gender, young age, pubertal timing, suicidal ideation,
alcohol use, not living with both parents, and poor self-worth (32).
Suicidal ideation is more common than both attempted and completed suicide (10). However, its
exact dimension is still unclear. For'example a review of studies published after 1985 on
community surveys in adolescent populations (particularly high-school students) reveal that
between 3.5% and 52.1% of adolescents report suicidal thoughts (27). Generally speaking, the
great differences existing in these percentages may be explained by the different definition used
of suicidal ideation and with the different time intervals to which the study referred (e.g., past
year, lifetime, past two weeks, etc.). There is also evidence of a higher percentage of suicidal
thoughts in female subjects which is maintained among women in old age (33). Overall, the
prevalence of suicidal ideation among older adults (in both sexes) is estimated to be between
2.3% (last two weeks) and 17% (life-time)(34). However, compared to other suicidal behaviors
(e.g., attempts), suicidal ideation may not be a useful indicator of which adolescents or adults are
most in need of preventive services.
Risk Factors for Suicidal Behavior
Suicidal behaviour is a multidetermined phenomenon. The factors that place individuals at risk
for suicide are complex, interactive, and interdependent. Identifying these factors and
understanding their role in both fatal and non-fatal suicidal behavior is a key to preventing
suicide. Epidemiologists and suicidologists have described a number of specific characteristics
that are closely associated with an elevated risk for suicidal behavior. These include
demographic (e.g., age, sex as mentioned above), psychiatric, biological, relationship, and social
and environmental factors.
Psychiatric Risk Factors
Much of what is known about suicide risk is derived from studies in which researchers interview
a surviving parent, friend, or other close proxy to identify specific life events and psychiatric
symptoms that a suicide victim experienced before death. These types of studies are referred to
as psychological autoposy studies. Research using this type of approach has shown that many
adults who complete suicide had evidence of a psychiatric condition that could be diagnosed at
the time of death and retrospectively months or even years earlier (35, 36).
Major depression, other mood disorders (e.g., bipolar disorder), schizophrenia, conduct,
personality, and anxiety disorders, impulsivity, and sense of hopelessness are some of the major
psychiatric and psychological factors associated with suicide (Tab. 3 provides the prevalence of
mental disorders in 5,588 cases of suicide) (37). A diagnosable depressive disorder plays a major
8
role in suicide and is estimated to be involved in approximately 65-90% of all suicides with
psychiatric pathologies (38) Among depressive people, risk seems to be higher when patients are
not compliant with treatment or consider themselves as untreatable (or are considered as such)
(39) (See Box 4). The life-time risk of suicide in those affected by major and bipolar depression
has been estimated around 12-15% (40,41).
Schizophrenia is another psychiatric condition with high exposure to suicide, with a life-time risk
estimated to be around 10-12% (42). The risk is particularly relevant in young male patients, in
the early stages of the disease, especially for those subjects with good psychosocial functioning
before the onset of the illness, chronic relapses, and fears of “mental disintegration” (43). Other
factors, such as feelings of hopelessness and helplessness also increase the risk of committing
suicide. Beck et al (44), for example, in a ten-year longitudinal evaluation, underlined the
importance of feelings of hopelessness as a major predictor of suicidal behaviour. In their study,
lack of future expectations correctly identified 91% of subjects who subsequently died by
committing suicide.
Alcohol and drug abuse also plays an important role in suicide. In the USA it has been reported
that at least one fourth of all suicides involves alcohol abuse (45). Life-time risk of committing
suicide in alcoholics is not much lower than that of depressive disorders (45); on the other hand,
points of contact between these two pathologies are mutiple and are often difficult to distinguish.
There are many possible causal links: 1) alcohol abuse may result in depression directly or
through the downward mobility and failure that most alcoholics experience; 2) alcohol abuse may
be a way of self-medicating to alleviate depression; or 3) both depression and alcohol abuse may
be the result of similar stressors in the person’s life. However, while in depressive disorders
suicide happens relatively early in the history of the disease (especially in the fourth decade of
life), in alcoholics suicide usually occurs late in the condition, often in conjunction with other
factors such as a breakdown in important relationships, social emargination, indigence, and onset
of a somatic complication from a chronic abuse (See Box 5).
Some of the strongest risk factors for a completed suicide, however, are previous non-fatal
suicidal behaviors. A previous suicide attempt is perhaps the most powerful predictor of
subsequent fatal suicidal behaviour (4). The risk is higher in the first year, and especially in the
first 6 months after the attempt, with nearly 1% of individuals dying by suicide during that time
(46). The level of increased risk due to the history of a previous attempt varies from study to
study. Gunnell and Frankel, for example, report a 20-30 fold increase in risk in comparison to the
general population, which is consistent with several other reports (47). While the presence of
previous suicide attempt is very common in suicided people, it should be noted that the majority
who those who die by suicide do not present such an aspect (48).
Biological and Medical Markers
A family history of suicide is one of the recognized markers for increased risk of suicide. To
some researchers, this suggests that there may be a biological trait passed from generation to
generation which predisposes some people to suicidal behavior. Data from clinical, twin and
adoption studies suggest that biological factors may play a role in some suicidal behaviour. Twin
9
studies have shown that monozygotic twins, who share 100% of their genes, have a significantly
higher concordance for both suicide and attempted suicide than dizygotic twins who share 50%
of their genes (49). However, the twin studies have not yet considered monozygotic twins reared
apart, a prerequisite for methodologically sound twin studies, and none of the studies have
carefully controlled for psychiatric disorders. It could be that it is a psychiatric disorder that is
inherited, and this increase the risk of suicidal behaviour in related individuals.
Adoption studies show that significantly more biological relatives of adoptees who committed
suicide had themselves suicided in comparison with biological relatives of control adoptees (50).
As these suicides were largely independent of the presence of psychiatric disorder, it suggests
that there is a genetic predisposition for suicide independent of, or additive to, the major
psychiatric disorders associated with suicide. Other social and environmental factors probably
also interact with family history to increase risk for suicide.
Other evidence suggesting a biological basis for suicide is from studies of neurobiologic
processes that underlie many psychiatric conditions, including those that predispose to suicide. A
number of studies, for example, have shown altered levels of serotonin metabolites in the
cerebrospinal fluid of adult psychiatric patients who completed suicide (51, 52). Low levels of
serotonin and blunted neuroendrocrine responses to serotonergic challenges have been shown to
persist over time after episodes of illness (53, 54). Serotonergic trait abnormalities are thought to
lead to a lowering of the threshold for suicidal behaviour at times of stress or psychiatric illness.
Impaired prefrontal cortex serotonergic function may underlie a reduced ability to resist impulses
to act on suicidal thoughts (55, 56).
Suicide may also be the consequence of a severe and painful illness, especially one that is
disabling. The prevalence of physical illness in suicided subjects is estimated to be at least 25%
and in more than 40% of cases it is considered an important contributory factor to suicidal
behaviour and ideation, especially if concomitant to a mood disorder or depressive symptoms
(57). Understandably, the perspective of unbearable suffering and humiliating dependence may
render envisageable the “rational” hypothesis of prematurely ending life. However, several
investigations have demonstrated that only rarely does suicide occur in subjects suffering from a
physical illness in the absence of psychiatric symptoms (38).
Relationship Factors
Certain life events may serve as precipitating factors in the etiology of suicide. Those that pertain
to personal loss, interpersonal conflict, the disruption of a relationship, and pending legal or
work-related problems have been linked to suicidal risk in a number of studies (58-61).
For some individuals a loss of a loved one either through divorce, separation, or death may
trigger intense depressive feelings, especially if the loss involves a partner or a very near and dear
person. For others, conflict associated with interpersonal relationships in the home, school, and
workplace can also trigger feelings of hopelessness and depression (See Box 6). For example, in
a study of over 16,000 adolescents in Finland, the researchers found an increased prevalence of
depression and severe suicidal ideation among both those who were bullied in school and among
those who were perpetrators of bullying (62). Controlling for age, sex, and mental disorder,
10
researchers in South East Scotland found adverse interpersonal events to be independently
associated with suicides in a retrospective case-control study (63). In a review of all suicides
over a two year period in Ballarat, Victoria, Australia, the researchers found social and personal
difficulties to be associated with suicide in over a third of the cases (64). Previous research also
shows an elevated risk of depression and suicide attempts among victims of intimate partner
violence (65-68).
A history of physical or sexual abuse in childhood also contributes to suicide risk in adolescence
and adulthood (69-71). Humiliation and shame are typically present in sexual abuse victims (4).
Some of the consequences of abuse during childhood and adolescence include generalized
feelings of mistrust in interpersonal relationships, difficulty in maintaining such relationships,
persistent sexual difficulties, and intense feelings of inadequacy and inferiority. For example, a
study in the Netherlands comparing adolescent functioning and sexual abuse in 1490 students,
found that the abused sample had significantly more suicidal behaviour and emotional and
behavioural problems, than the non-abused adolescents (72). In an Australian study, 68 sexually
abused children, 5 years after initial clinical presentation, displayed more disturbed behaviour,
had lower self-esteem and were more depressed and anxious than age and gender-matched
controls (73). An ongoing 17-year longitudinal study of 375 subjects in the USA, found 11%
reported physical or sexual abuse before the age of 18. Abused participants at the ages of 15 and
21 reported more suicidal behaviour, depression, anxiety, psychiatric disorders and emotionalbehavioural problems than those not abused (74).
Though data are lacking, there is also a belief that sexual orientation may be related to an
increased risk for suicide among adolescent and young adult populations. Estimates of suicide
among gay and lesbian youth, for example, range from 2.5% to 30% (75, 76). Factors such as
discrimination, intrapersonal stressors, drugs and alcohol, HIV/AIDS, and limited support
structures contribute to suicide and suicide attempts (77, 78).
Being in a stable marital relationship, on the other hand, seems generally to be a ‘‘protective”
factor against suicide. Childrearing responsibilities confer an additional protective element (79).
Research examining the relationship between marital status and suicide reveals high rates of
suicide among single or never-married persons in western cultures, even higher rates among
widows, with some of the highest rates found among those who are separated or divorced (80,
81). This phenomenon is particularly evident in male subjects, especially with regard to the first
months from the loss/separation (82).
In contrast with the generally protective effect of the marriage are early marriages (<20 years of
age). For these teenage marriages, a rate constantly higher than that of unmarried peers has been
reported in several studies (83, 84). It is also important to point out that marriage is not protective
in all cultures. Higher rates of both fatal and non-fatal suicidal behaviour have been reported
among married women in Pakistan compared to married men and single women (85, 86). Factors
such as legal, social, and economic discrimination may predispose these women to psychological
stress and subsequent suicidal behavior (85). Higher rates of suicide have also been reported
among married women over the age of 60 in Hong Kong compared to the widowed and divorced
in this age group (83).
11
While problems in interpersonal relationships increase the risk of suicidal behaviour for some
individuals, social isolation is also a marker for suicidal behaviour. Social isolation underlies
Durkheim’s concepts of ‘egoistic’ and ‘anomic’ suicide (87), both of which incorporate the
notion of insufficient social connectedness. It is readily accepted that social isolation can be a
precipitating or triggering factor for suicide. For example, following the death of a loved one, a
person may complete suicide if they are insufficiently supported by those around them during the
grieving period. In addition, social isolation can be a sign or symptom of potential suicidal
behaviour. A large body of literature has reported that individuals who experience isolation in
their lives are more vulnerable to suicide than those who have strong social ties with others (8891).
In a comparison study of social behaviour between groups of suicide attempters, suicide
completers and people dying of natural causes, Maris (92) found that the group of suicide
completers had participated in less social organisation, were often without friends and had shown
a progressive decline of interpersonal relationships leading up to a state of total social isolation.
Psychological autopsy studies indicate that social withdrawal frequently precedes the suicidal act
(92). This was also highlighted in a study by Negron et al (93) who found that suicide attempters
were more likely to isolate themselves in an acute suicidal phase than suicide ideators. Wenz
(94) identified anomie, actual and expected social isolation as etiological factors in widow
suicide. Additionally, social isolation has been frequently identified as a contributing factor in
suicidal ideation among the elderly (95, 96). In a study on suicide attempt among adolescents
under 16 years of age who had been referred to a general hospital, Hawton (97) found that the
most frequent problems underlying this behaviour were relationship difficulties with parents,
problems with friends and social isolation.
Social and Environmental Factors
The social and environmental context is also important for understanding fatal and non-fatal
suicidal behaviour. Previous research has identified a number of important social and
environmental factors related to suicidal behavior, including the availability of means, place of
residence, immigration, employment, economic stability, social integration, and religion.
12
A major factor determining whether a suicidal behaviour will be fatal or non-fatal is the method
chosen. (See Box 7) In the United States, firearms are used in approximately two-thirds of all
suicides (4). In other parts of the world, hanging is more common, followed by firearm, jumping
from a height, and drowning. In China, however, intoxication by pesticides is the most
commonly used method.
In the last two decades, in several western countries and particularly in Australia there has been a
remarkable increase in hanging, especially by younger people, accompamed by a nearly parallel
decrease in firearm use (98). In general, elderly people tend to adopt methods implying less use
of physical strength, such in the case of drowning and especially of jumping from heights, as it is
the case of the elderly in Singapore or Hong Kong (99). Nearly everywhere, women tend to
utilize more “soft” methods (for example, overdosing with medicines), both in fatal and in nonfatal suicidal behaviours (31). A well-known exception is self-burning in India.
Apart from age and gender, the choice of method in suicide is influenced by several other factors.
For example, tradition influences the perpetuation of the practice of hara-kiri in Japan. Imitation,
especially in young people often in relation to a media event (100) or to the suicide of a celebrity
(101), has been seen to strongly influence the choice of the method (102). The degree of intention
is generally related to the lethality of the method: elderly people normally express a greater
determination to die and they tend to chose more violent methods (eg, firearm, hanging, jumping
from a high place) in the context of circumstances that aim not to offer possibilities of being
rescued (103). The place of residence is also strongly related to the choice of method. For
example, in rural communities of Eastern European countries, the easy availability of herbicides
and pesticides renders these means as very frequently adopted for suicidal purposes, but the same
holds true for the Pacific islands of Samoa, where -as confirmation of this hypothesis- the control
of the sale of paraquat, a herbicide, led to an actual decrease in the number of deaths due to
suicide (104). In rural communities of Australia, where possession of firearms is very common,
their use as sucidal method is far more frequent than hanging (105).
Suicide risk appears to be related to place of residence also for different motives than access to
means. In fact, although the number of suicides is far greater in urban areas, rates of suicide are
often higher in rural and remote areas. For example, in 1997 New York (Manhattan) recorded
1,372 suicides, a number three times higher than that of the state of Nevada (411), but the latter
has more than three times the rate of New York (24.5 -the highest in die USA- versus 7.6) (106).
Urban/rural differences in suicide rates are not unique to USA. Similar distributions have been
reported for Australia (105), but also in European countries such as Scotland and England and
Wales, where farmers have been found to have high rates of suicide (107). Higher rates of suicide
among women living in rural areas of China have also been reported (see box on China). Social
isolation, difficult detection of warning signs by proxies or other community members, limited
access to health facilities or general practitioners, lower level of education, all are factors that
may contribute to explain the higher rate of suicide in rural and remote areas.
The impact of immigration on suicide rates has been been studied particularly in countries such
as USA, Canada and Australia. In these countries, where the population is constituted by
different ethnic groups, suicidal behaviour in a given group appears to be analogous to that of the
country of origin, with rates tendentially slightly increased. In Australia, for example, immigrants
from Greece, Italy, Pakistan evidence suicide rates remarkably lower than those of immigrants
13
coming from Eastern European countries, or from Scotland or Ireland, countries with
traditionally higher suicide rates (108)(Tab.4). This observation strongly emphasises the role of
cultural factors in suicidal behaviour.
Several studies have also revealed increased rates of suicide during periods of economic
recession and high unemployment rates (109). The reverse has been demonstrated during
“booming” periods. In a study examining the impact of economic factors on the frequency of
suicide in Germany, Weyerer (110) investigated the effect of four economic variables and their
relationship to suicide rates between the year 1881-1989 (pre-unification). The strongest
correlation was found during times of obvious social disintegration, high unemployment with
diminished state safeguards and increased frequency of bankruptcy. A preliminary investigation
into the above average suicide rate in Kuzbass (Russia) from 1980 to 1995 gave economic
instability, the disintegration of the USSR as well as specific historical factors as possible
contributions (111). A qualitative account by Berk (112) of his visits in Bosnia reported a higher
than expected rate of suicide as well as alcoholism among children: while they had survived the
most immediate threat posed by war, the young had succumbed to the long-term stress.
At the individual level, suicidal behaviour is more frequent in unemployed than in employed
people. Indigence and a socially deteriorated role —both a consequence of lack of work- appear
to be variables often associated with increased suicidal behaviour, especially in case of sudden
loss of a previous occupation. However, research into unemployment impact has generally
suffered from a number of confounding factors, such as the commistion of subjects waiting for
first employment with others who have lost their occupation, length of the unemployment period,
“under the table” work, concomitance with psychiatric conditions and personality disorders, etc.
Religion has long been thought of as an important factor in fatal and non-fatal suicidal behavior.
Previous research indicates that suicide rates are highest in countries where religious practices are
either strongly discouraged or prohibited (such as was the case in former communist countries),
followed by countries in which Asian religions predominate, and in countries where Protestant
Christianism is stronger. Countries that are predominantly Moslem have some of the lowest
suicide rates in the world, immediately preceded by countries that are largely Roman Catholic.
Unfortunately no data are available in relation to the majority of countries following Animistic
religions, mostly found in Africa. This obviously does not capture the importance nor the degree
of individuals’ adherence to and observation of the precepts of a given religious denomination
(113).
Durkheim believed that suicide stemmed from a lack of identification with a unified group and
postulated that the incident of suicide would be reduced in countries with a high degree of
religious integration. Accordingly, Durkheim argued that shared religious practices and beliefs,
such as those associated with Catholicism, are protective factors against suicide (87). Several
studies that have investigated Durkheim’s hypotheses have found support for his argument (114,
115). A study by Simpson and Conlin (116) which examined the impact of religion found that
belief in Islam reduced suicide rates more than a belief in Christianity. Several other studies have
found no association between the percentage of Catholics in a population and a reduced suicide
rate (117, 118).
Amidst the difficulties of research looking specifically at religious denomination, further studies
14
have investigated the influence of church attendance and networks as a measure of religious faith.
The findings of these studies have suggested that church attendance and a network interpretation
have a strong preventative influence (119). The degree of commitment and involvement in
religion was found to be an inhibitor of suicide (120). Similarly, a study by Kok (121)
investigated the suicide rate among the three ethnic groups of Singapore and found that the
Malays (a Muslim group who are strongly opposed to suicide) had by far the lowest suicide rate.
Furthermore, the Hindu groups who believe in reincarnation do not strictly forbid suicide, had the
highest rate of suicide. Another study examining differences between Afro-American and white
populations of the US, found that the lower rate of suicide among Afro-American could be
attributed to orthodox religious beliefs and personal devotion (122).
In sum, risk factors for suicidal behaviour are numerous and interactive. Psychiatric disorders,
especially depression and schizophrenia, are found to have been present in many people who
have suicided. Alcohol and drug abuse and previous suicide attempts are also strong indicators
of completed suicide. Biological studies propose a genetic link to suicide and lower levels of
serotonin and its metabolite in the brain. Severe and disabling physical illnesses contribute to
suicidal behaviour, although are usually coexisting with a psychiatric disorder. Other risk factors
include: loss of a loved one, physical or sexual abuse, separation or divorce, social isolation,
living in rural areas, migration and unemployment. Although not one “cause” of suicide can be
given, a predisposition coupled with a combination of risk factors, may help to predict those most
in need for prevention strategies.
Response to the Problem: Prevention, Policy Responses
Considering the increase in the occurrence of suicidal behaviour, in particular among youngsters,
there is a need for effective prevention and intervention programs. Multiple factors have been
identified as risk factors for suicidal behaviour. Knowledge of these risk factors is the foundation
on which prevention and intervention programs are based. Although many of these programs
have been developed over the years, very few of them have demonstrated some influence in
reducing suicidal behaviour.
Psychiatric and Psychological Factors
Treatment of mental disorders. Since the majority of published studies - and clinical
experience, as well - indicate that a few mental disorders are significantly associated with suicide,
the early identification and appropriate treatment of those disorders emerge as an important
strategy for the prevention of suicide. Mood disorders, alcoholism and other substances abuse,
schizophrenia and some types of personality disorders are particularly relevant in this respect.
There is evidence that the education of primary health care personnel in the identification and
treatment of people with mood disorders may effectively result in a reduction of suicide rates
among those at risk. Also, there are indications that new generation medication for both mood
and schizophrenic disorders, with less side effects and a more specific therapeutic profile,
increases adherence to treatment and better outcomes, thus reducing suicide rates among clinical
populations. The reduction of the stigma still attached to people with mental disorders in many
15
communities favours this people coming forward to receive treatment at early stages of the
disease, when treatment is more efficient, thus contributing also to the reduction of suicide.
Treatment of suicidal behaviours. In clinical populations two basic strategies of treatment are
employed. According to the first strategy, suicidal behaviours are secondary to the mental
disorder. Treatment is primarily focussed on the mental disorder under the assumption that its
improvement will result in reductions in suicidal behaviours. The second strategy engages
treatment, which directly targets suicidal behaviours. Reduction of suicidal behaviours is the
primary goal of treatment (123). In alignment with the latter strategy different treatments, or
interventions, have been developed, two of which will be discussed below.
Behavioural Interventions
Behavioural interventions employ a certain behaviour and problem solving focus. In the therapy
sessions the client has to discuss current and past suicidal behaviours, including ideation, threats
and communications with a mental 'health worker, exploring connections to the possible
underlying or controlling factors (123). Results on the efficacy of these treatments are promising,
although there is not a conclusive answer on the efficacy yet.
A study by Salkovskis et al. (124) included multiple high-risk suicide attempters with a history of
suicidal behaviour, aged 16-65 years, who were admitted to the Emergency Ward for an
antidepressant overdose. Patients received either treatment as usual (TAU) or both TAU and a
brief problem-oriented intervention. Salkovskis found a significant difference in parasuicide
repeat rates in favour of the experimental group six months after treatment. Unfortunately, this
difference was not significant anymore when the difference was assessed after 18 months.
A study by Linehan (125) was a one-year intervention aimed at patients with multiple
parasuicides with borderline personality disorders, multiple behavioural dysfunctions and
significant mental disorders. During the first year after treatment, patients who received
dialectical behaviour therapy (DBT) had fewer parasuicidal episodes than patients who had
received TAU did.
Another research study (126), which adopted a behavioural therapy approach examined whether
high risk parasuicide patients showed a deficit in positive future thinking and whether such a
deficit could be remedied by a brief, manual assisted psychological intervention (manual assisted
cognitive behaviour therapy: MACT). Patients were randomly assigned to either MACT or
treatment as usual (TAU) and assessed again at 6 months follow-up. They found that parasuicidal
patients showed reduced positive future thinking compared to a sample of controls. Patients who
received MACT showed a significant improvement in their positive thinking in the follow-up
period, whereas patients who received TAU did not. However, the control group did also show a
significant improvement in positive future thinking, a finding that could not be explained. In
terms of the effectiveness of the intervention on the parasuicide repetition rate, the median rate of
repetition per month was halved in the MACT group, although this was not statistically
significant.
16
Green Card
The green card is a relatively simple non-demanding intervention. The client receives a card,
which gives him/her immediate access to different sources of crisis intervention, such as an oncall psychiatrist and/or immediate hospitalisation. The green card has not proven to be
particularly effective, but does exert some beneficial effect on first-time attempters.
Morgan’s study (127) involved first-time suicide attempters. The green card gave patients easy
access to the accident and emergency department and the availability to contact an on-call trainee
psychiatrist. The experimental group received TAU and the green card; the control group only
received TAU. At 12 months follow-up the experimental group showed reductions in rates of
repeated suicide attempts, although not significant. However, when suicide attempts and threats
were combined there was a significant reduction in the experimental group.
Cotgrove et al (128) examined a group of adolescents. In this study the green card gave them the
possibility of re-admission to the hospital on demand. The green card did not include a special
telephone support service. After one year, following the suicide attempt, differences in repeated
suicide attempts were not significant, but the results suggested lower rates in the experimental
group, even if the card was not used.
A recent study by Evans et al (129) used the green card in a mixed group of first-time attempters
and patients with a history of suicide attempts. Patients were randomly allocated to control
groups, which received only TAU, or experimental groups, which received TAU and the green
card. The green card offered a 24-hour crisis telephone consultation with an on-call psychiatrist.
The green card had a different effect for first-timers and patients who had made previous suicide
attempts. Among the latter group, the odds of repeating suicide attempts were higher in the
treatment group, while the green card appeared to exert a protective effect, although non
significant, on those who had not previously attempted suicide.
In conclusion, the green card seems beneficial to a certain degree to first-timers, but as indicated
in Evans' study (129) it does not seem to be beneficial to patients who have made previous
attempts. As indicated in the literature, patients who have made multiple suicide attempts are a
vulnerable group. It might be that telephone support alone, as given in Evans’ study, was not
enough and the green card should have offered easy access to more or different sources of crisis
intervention to this population.
Another intervention, which is based on the principle of connectedness and easy access and
availability of help, is a Tele-Help/Tele-Check service for the elderly operating in Italy (130).
Tele-Help is an alarm system that the client can activate to call for help. The Tele-Check service
contacts the client twice a week for assessment of needs and for emotional support. In this study
12,135 individuals aged 65 years and over were connected to the Tele-Help/ Tele-Check service
for 4 years. During this period only one suicide was found as compared with the expected
number of 7.44 (130).
17
Biological Factors
Pharmacotherapy. Pharmacotherapy has been examined for its efficacy in affecting
neurobiologic processes, which underlie psychiatric conditions, including those that are related to
suicidal behaviour.
Verkes et al (131) indicated that paroxetine might be effective in reducing suicidal behaviour.
This study was conducted on the basis that suicidal behaviour has been associated with reduced
serotonergic function. Paroxetine is a selective serotonin reuptake inhibitor (SSRI), which
enhances the serotonergic function. In a 1 -year double blind study paroxetine and placebo were
compared in 91 patients with a history of suicide attempts and who had recently attempted
suicide. These patients had not suffered major depression, but the majority had a cluster B
personality disorder. The results showed that enhancing serotonergic function with an SSRI, in
this case paroxetine, might reduce suicidal behaviour in a subgroup of patients with a history of
suicide attempts, but who do not suffer from a major depression.
Relationship Factors
Research has indicated that susceptibility to suicide is related to the social relationships of a
person; the greater the degree of social relationships, the less the susceptibility to suicide (132).
A number of interventions have focussed on the enhancement of social relationships in order to
reduce repeated suicidal behaviour. The general approach is to explore problems in different
areas of the client’s social life and to target them in collaboration with the therapist. Although the
main goal of the intervention is to prevent recurrent suicidal behaviour, the improvement of
social relationships is also considered important.
Research has shown that the interventions are ineffective regarding prevention of recurring
suicidal behaviour. However, intermediate goals related to enhancement of social relationships
were achieved.
Psychosocial interventions. Litman and Wold (132) investigated the efficacy of a reaching-out
service, called continuing relationship maintenance (CRM). A total of 400 subjects, who were
evaluated as high-risk by lethality rating scales, were assigned to this program for an average of
18 months. They either entered the experimental (CRM) group or the control group, the latter
receiving on-going counselling with the clients taking initiative for contact themselves. The
intervention did manage to reduce suicidal ideation, attempts, and completed suicide. However,
the intermediate goals were achieved, with the CRM group showing significant improvement
compared with the control group. The CRM group showed reduced loneliness, improvement in
love relationships, better use of professional help, less depression, and more confidence in using
community resources.
Gibbons et al (133) examined different outcome measures for patients who received routine
treatment and those who received so-called task-centred casework. There was no difference in the
18
number of repeated attempts between both groups, but they showed differences on measures of
social problems and satisfaction with the service they had received. The experimental group
showed greater improvement in social problems and was more satisfied with the service than the
control group.
In a study by Hawton et al (134) 80 overdose patients either received out-patient counselling
(OP) or were returned to their general practitioners (GP) with recommendations for further care.
Again, there was no statistical difference in the rates of repeated suicide attempts, but there were
indications of some degree of increased benefit for the OP group at a 4-months assessment. A
greater proportion of the OP group had resolved or improved their target problems and on the
Social Adjustment Scale (SAS) they showed particular improvements in social adjustment,
relationship with extended family, marital adjustment, and family relations. Counselling seemed
most beneficial for women and patients with dyadic problems.
Social and Environmental Factors
Suicide prevention centres. Besides the specific interventions, as described above, there are
community mental health services available for persons exhibiting suicidal behaviour. A suicide
prevention centre is supposed to serve as a crisis centre offering immediate help mostly via
telephone contact and crisis-oriented programs.
Dew et al (135) performed a quantitative literature review of the effectiveness of suicide
prevention centres. Results suggested that suicide prevention centres have no specific effect,
either positive or negative, on the population rates. However, methodological limitations of this
study make it difficult to firmly conclude that centres do not prevent suicide. Furthermore, they
found that the proportion of suicides among prevention centre clients are greater than the
proportion of suicides in the general population, and that individuals who committed suicide are
more likely to have been centre clients. Both these results indicate that the suicide prevention
centres are at least successful in attracting the population they are supposed to help: the high-risk
population.
Lester (136) reviewed 14 studies examining the effectiveness of suicide prevention centres on
suicide rates. Of these, seven studies were found to provide some evidence for a preventive
effect. A study by Riehl et al (137) actually reported an increase in suicide rates in three of the 25
German cities with a suicide prevention centre he examined. Thus, Lester found some support for
thq preventive effect of suicide prevention centres, but the results also indicated that this effect
might not be found for all subgroups in a population, or for all methods of suicide.
School-based interventions. These programs train school staff, community members, and health
care providers to identify those at risk for suicide and make referrals to mental health services.
The extent of training varies from program to program, but all emphasise a strong link to local
mental health services. The interventions have demonstrated improvements in knowledge of
suicide and willingness to refer.
A note of caution though is made by Lester (138) who suggested that with the education of
school staff members, students might be referred to mental health professionals on fewer
occasions, which might resulted in increased mortality. He based this suggestion on the finding
that states in the US with school programs witnessed an increase in their youth suicide rates
19
(138). Although education of school staff members, parents and others involved in school
programs is important, they can not replace the role of the mental health professional. It is
important that vulnerable students are referred on to those with the skills to intervene and treat,
because only then programs and treatments are beneficial. Nevertheless, there are certainly good
reasons for schools to act as a medium for suicide prevention. The primary reason is that health
care facilities alone can not meet all the needs of youths.
A school program that has been created by Dade County Public Schools (DCPS) in Miami,
Florida shows the different types of prevention as previously described. This program is the
Suicide Prevention and School Crisis Management Program (SPSCMP), which was implemented
in 1989 (139). The results of this program were rather promising. Overall, the program seemed to
be effective in reducing the rate of suicide attempts and completed suicides, although the rate of
student suicidal thoughts remained relatively stable. However, two important limitations of this
study should be mentioned. First of all the number of suicides is very small. Within the period of
1980-1994 the number of suicides fluctuated between 3 per year in 1993 and 18 per year in 1989.
This brings us immediately to the second point of criticism namely the sudden increase in 1988,
just before the program started. The number of suicides was the highest recorded since 1980. The
decline in suicides after 1989 seems thus enormous because of this high initial number of
suicides. The reason for this sudden increase in 1988 has not been explained and the decline
might be a natural response to this, regardless of the implementation of the program.
Restriction of access to means of suicide. Restriction is particularly relevant when access to
preferred means of committing suicide is amenable to control. The first evidence of this approach
was demonstrated in 1972, by Oliver and Hetzel (140), in Australia, who demonstrated a
reduction in suicide rates when access to sedatives (mainly barbiturates, lethal in high doses) was
reduced.
In addition to the study of Oliver and Hetzel, on the reduction of availability of sedatives, there is
evidence of the impact of the reduction of availability of other toxic substances, such as
pesticides, widely disseminated in rural areas of many developing countries. The case of Samoa
is perhaps one of the best-studied examples (104): until 1975, when "paraquat” was introduced in
the country, total suicide rates were below 5 per 100,000. In 1976 suicide rates started to climb to
reach nearly 50 per 100,000 in 1982, when access to "paraquat” was drastically curtailed. Within
two years, suicide rates had dropped to approximately 10 per 100,000. It is interesting to note that
between 1976 and 19$2 the so-called "paraquat suicides" represented between 50% and 80% of
all suicides, depending on the year, and that in spite of the mechanisms for the control of access
to "paraquat", after 19;84 more than 90% of all suicides are represented by the so-called "paraquat
suicides". (See Figured).
I
The removal of carbon monoxide from domestic gas and from car exhausts represents the best
known examples of the reduction of suicide rates through gas detoxification. The mechanisms
through which this approach is applied may vary from place to place. Suicides from poisoning
with domestic gas in England started to decline in 1955 soon after carbon monoxide began to be
removed (141). Although this impact was dramatic on suicides using that particular method, it
was also observed in relation to the overall suicide rates (141). In spite of a slight increase of the
total number of suicides after the end of the process of detoxification of domestic gas, it never
reached the rates prior to 1955, probably due also to other effective interventions (see Figure 6).
20
Similar declines in the use of domestic gas for suicide have been noted in Scotland, the
Netherlands, Japan, the United States and Switzerland (142). The introduction in 1965 in the
USA of catalytic converters for the removal of carbon monoxide from car exhausts resulted in a
clear sustained reduction of suicides using that method (143). The same phenomenon was
observed in Japan (142). In addition, data from the UK, where no such device was introduced,
shows a remarkable increase over the same period (See Figure 7).
The relationship between the possession of handguns at home and suicide rates has been welldemonstrated (144). The efficacy of firearm control depends largely on the adoption of
legislation, which regulates areas ranging from guns sales, ownership and storage, to the
incorporation of mechanisms that distance guns from bullets and others, such as trigger blocking
devices. An additional advantage of this approach resides in that it also contributes to wider
injury prevention programs, be those injuries accidental or intentional. In some countries
(Australia, Canada and the United States) restrictions on the ownership of firearms has been
associated with a reduction in their use for suicide (142).
Media reporting. The possible impact of mass media on suicide rates has also been documented
and already referred. The evidence indicates that the impact of media reporting on imitation
suicides depends largely on the way it is reported, e.g., the tone of the text, terms employed, the
location of the matter and the use of graphic and unnecessary material. There is also some
indication that the “vulgarisation” of reports about cases of suicide might create a “suicide
culture”, in which suicide is perceived as a normal and acceptable way out of some difficult
situations.
Responsible media reporting also includes the provision of adequate and reliable information
about agencies and places where help can be obtained in case of need, particularly in crisis
situations (145). Deglamourising suicide also means the provision of adequate information in the
cases of suicides committed by celebrities, such as the pain and mourning impinged upon
survivors. Media can also be used to bring health issues in the publicity. Community-based
interventions often involve media for its great potential to reach large parts of the population.
Evaluations of the effectiveness of health promotion via the mass media have suggested limited
results (146). The mass media is most effective in making people aware of certain health issues
they had little knowledge of. Although media campaings have a modest effect on general
attitudes, they hardly bring about behaviour changes (146).
Intervention after a Suicide: Postvention
The loss of a person by suicide evokes different feelings of grief in the survivors than death from
natural causes. In general, there is still a taboo attached to the discussion suicide. Those people
bereaved by suicide might therefore have less opportunity to talk with others about their grief.
Communication of feelings is an important part of the healing process. Survivor groups serve a
very important role in this process. The first self-help support groups were established in North
America and the United Kingdom in 1960. In 1970, the first support group for suicide survivors
started in North America, followed by various countries throughout the world. Self-help support
groups are described as groups of people who are directly and personally affected by a specific
21
issue. The members run the groups, but access to outside resources and assistance is made
available.
Evidence has suggested that the self-help groups have positive outcomes for their participants.
The common experience of loss by suicide bonds people and might encourage them to
communicate their feelings, which is often difficult in society in general (147).
Policy Responses
In 1996 the United Nations Department for Policy Coordination and Sustainable Development
issued a document in which the importance of a guiding policy on activities related to suicide
prevention was highlighted and developed, and which became a landmark in the subject (148). In
addition to that, WHO published a series of documents on strategies for the primary prevention of
mental, neurological and psychosocial disorders, in which a fascicle was dedicated to suicide;
later, that series was edited in a book format (149). Other reports on suicide and prevention have
also been developed (for example, CDC guide).
In 1999 WHO included the prevention of suicide among its priorities and launched SUPRE
(SUicide PREvention), a global initiative aiming at the prevention of suicidal behaviours, with
the following objectives:
1. To bring about a lasting reduction in the frequency of suicidal behaviours, with emphasis on
developing countries and countries in social and economic transitions.
2. To identify, assess and eliminate at early stages, as far as possible, factors that may result in
young people taking their own lives.
3. To raise the general awareness about suicide and provide psychosocial support to people with
suicidal thoughts or experiences of attempted suicide, and to their relatives and close friends, as
well as those of people who committed suicide.
The main strategy for the implementation of SUPRE is based on two elements, along the lines of
the Primary Health Care Strategy:
1. organisation of global, regional and national multi-sectoral activities to increase awareness
about suicidal behaviours and their effective prevention, and
2. strengthening of countries capability to develop and evaluate national policies and plans for
suicide prevention, which might include, e.g.:
- Support and treatment of populations at risk (e.g., people with depression, the elderly, youth),
- Reduction of the availability and access to means of suicide (e.g. toxic substances, handguns),
- Support/strengthening of networks of survivors of suicide, and
- Training of primary health care workers and other sectors.
SUPRE has now been complemented by a Multisite Intervention Study on Suicide (SUPREMISS) which aims at the identification of both specific risk factors and specific interventions
effective for the reduction of suicidal behaviours.
22
Recommendations
Several important implications may be drawn from the information presented in this chapter for
the development and implementation of effective prevention strategies aimed to reduce both fatal
and non-fatal suicidal behaviour rates.
Firstly, suicide and attempted suicide are multidetermined phenomena that represent the interplay
of biological, psychological-psychiatric, and social factors in very individualised expressions.
Every suicidal action has multiple causes, so that any prevention programs should be
“multimodal” by definition (24). In other words, complexity of causes necessarily involves
complexity of approaches and, consequently, complexity of strategies.
Secondly, programs for suicide prevention are doomed to be ineffective if they are not cast within
the framework of large-scale plans carried out by multidisciplinary teams, comprised of
representative of governments, health-care planners and health-care workers.
Thirdly, given the size of the phenomenon and of its many fall-downs, major investments are
needed both in the area of research and in preventative efforts. In particular, there is the neccesity
of long-term, evaluable projects. So far, those that have been established have been of short
duration and assessment, if occurred, only concerned with short-term follow-ups.
Lastly, it is very important that countries do not rely on epidemiological surveys and prevention
strategies that have been developed in other nations. Research teaches us that cultural factors play
a major role in suicidal behaviour (150) and that there are huge differences in the dimension and
characteristics of this problem around the world. As an example, the average ratio between the
highest and the lowest suicide rates is 1:102.4 for men and 1:35.8 for women (151). Also,
epidemiological trends vary from country to country and the same holds true for research results.
What has shown to have a positive effect in preventing suicide in a given nation, may
demonstrate totally ineffective or negative impacts in another cultural ambience.
Consequently, there is an urgent need for more and better information concerning the causes of
suicidal behaviour, both at national and international level, with a particular attention to minority
groups. Cross-cultural comparisons, such in the case of the WHO/EURO Multicentre Study of
Suicidal Behaviour and of the very recent WHO/SUPRE-MISS, should be encouraged. They may:
help us to better understand causative and protective factors, and consequently assist us to re-j
orientate preventative efforts. With regards to this it is necessary that:
•
Data collection on both fatal and non-fatal suicidal behaviour is stimulated. It is noteworthy1
that nearly half of the countries owing to the United Nations are not reporting mortality data:
for suicide to WHO. This is particularly true for many nations of Africa and the Middle East,
and also for countries of Latin America. General hospitals and other socio-medical services,
should be encouraged to keep records of non-fatal suicidal behaviours.
•
Data collection must be valid and up-to-date. Within this aim, certification and classification
procedures should be improved by the adoption of uniform criteria and definitions. Once
established, these have to be constantly applied and reviewed. Data collection has to be
organised in such a way as to avoid duplication of statistical records and in the meantime to
be readily usable for analytical/epidemiological investigations by researchers and
23
national/intemational agencies.
•
Data banks have to be built in network with relevant agencies (eg, general hospitals,
psychiatric and medico-legal institutions, coroners, departments of justice, education and
labour, bureau of statistics, etc) in order to permit long-term monitoring and thus a better
understanding of risk factors and pathways to suicide. This implies that governmental
agencies work together in an interdisciplinary, coordinated manner and that all health
professionals and officers of involved agencies are educated in the detection and referral of
suicidal cases and in the appropriate coding of these behaviours.
•
Data sets of major social indicators should also be available in parallel with suicidal
behaviour data. Ideally, they should contain indices of “happiness” of a given nation, quality
of life indicators, rate of divorce, of unemployment, of homicide, alcohol and drug abuse,
education, religious appartenence, sexual orientation, ratio young/elderly in the population,
percentage of women in formal work settings, etc.
Once reliable data sets are obtained, a carefully designed multidisciplinary research program,
using control groups, may be set in operation. There is an urgent need for coupling psychosocial
measures with biological parameters. This may permit a great advancement of current
knowledge. For example, the rapidly expanding research in molecular genetics, especially the one
addressing the identification of genetic subtypes and polymorphism in alleles controlling
serotonin metabolism, appears to be a particularly promising field of research. Also structural,
and especially functional, brain imaging deserves more research investments. Longitudinal
evaluations carried out in suicide attempters may clarify if abnormalities possibly encountered
have “trait” or “state” dimension and this could be particularly relevant if associated with, for
example, a reduced serotonergic metabolism (55).
Clinical research may further develop, contributing to highlight the role of co-morbid conditions
and their causative role (for example, interaction between depression/alcohol abuse), as well as
sub-grouping subjects on the basis of their age (suicide in the elderly has different characteristics
from that in youngsters) and of their personality and temperamental characteristics. In addition,
the importance of aspects such as hostility, aggression, impulsivity, tendency to dyscontrol, waits
to be elucidated.
The large role that psychiatric risk factors play in increasing the risk of suicidal behaviour
indicates that improving treatment options for those who have a psychiatric disturbance would go
far in preventing suicide. This requires several components. First, pharmaceutical companies
must be encouraged to develop newer and more effective medications for psychiatric disorders.
For example, the advent of the serotonin re-uptake inhibitors may have resulted in a decline in the
Scandinavian suicide rates (152).
Second, research funding must be targeted to devising more effective techniques of
psychotherapy and counseling for suicidal individuals. In particular, there is the need for
developing more specific techniques for those people whose personality disorders are more
frequently associated with suicidal behaviour. This can be accomplished by issuing “requests-forfunding” rather than making general grants available for whoever applies.
24
Third, the gatekeepers in the society (family physicians, social workers, and clergy, among
others) must be trained to recognize, refer and treat appropriately those with psychiatric
disorders, especially affective disorders. In particular, the early identification and the appropriate
treatment of individuals suffering not only from mental disorders, but also from drug and alcohol
abuse and dependence, should constitute an absolute priority in the agenda of governments and
their health-care planners. The program set up in Gotland (Sweden) by Rutz (153) accomplished
this for physicians and has provided a model for other countries to follow
In addition to potential developments in treatment, many changes in the environment which
reduce access to lethal methods for suicide (such as emission controls on cars which reduce the
carbon monoxide content of the exhaust) are introduced to ameliorate other social problems (such
as pollution). However, some social policies can be directed specifically toward preventing
access to methods for suicide, such as:
25
•
fencing in high bridges and access to the tops of high buildings,
•
passing laws to force automobile manifacturers to change the shape of tail pipes and have
automatic engine turn-off after specified periods of engine idle,
•
restricting the availability of insecticides and fertilizers to non-farmers,
•
requiring the monitoring of prescriptions for lethal medications by physicians and
pharmacists, reducing the size of prescriptions, packaging the medications in plastic blisters,
and prescribing medication, when possible, as suppositories, and
passing stricter gun control laws.
«
Other areas of prevention that need further development and research are: suicide prevention
centres, survivors groups, media response to suicide and school suicide prevention programs.
Suicide prevention centers. Suicide prevention centers mainly use the telephone for crisis
intervention but also face-to-face counselling and outreach programs. These centres were initially
establishhed by volunteers in the community as a response to what communities perceived to be
an urgent problem. At present, and despite the high volume of client contacts they register, there
is no conclusive evidence that these centres have an impact on suicide. However, as already
commented in this chapter, they may have a small, but statistically significant, effect on the
suicide rate (130, 135, 136).
Several international organizations coordinate the centers around the world - Befrienders
International, IFOTES, and Lifeline. Befrienders International has now established Samaritan
centers in more than 40 countries. Although these agencies are usually supported by the
community, governments might do more to support them with better funding.
Survivor groups. Those who experience the suicide of a significant other (such as a child,
spouse or parent) are at high risk for suicide. Again, services to help these individuals have
typically been established and run by survivors themselves. These services provide regular
meetings and supportive people to call on in acute crises. As with suicide prevention centers,
governments should provide more funding and professional back-up and advice to support these
services.
To counteract social isolation, governments should also promote community-based programs (eg,
youth centres or senior centres) with the aim of stimulating social interaction and participation of
these groups in the life of the community. Governments should also broadly target emotional and
social well-being, addressing antecedents factors for suicide and increasing protective factors.
Especially among young people, initiatives aimed to promote help-seeking behaviour have to be
strongly stimulated.
26
Media response to suicide. As reported above, research has documented rather convincingly,
that suicides (real and fictional) presented in the newspapers and on television may result in a rise
in the suicide rate in some members of the society in the days after the publicity.
As a result, various agencies have proposed guidelines for the media in reporting suicidal
behaviour, including the Centers for Disease Control in Atlanta (USA), Beffienders International
(the United Kingdom), Australia and New Zealand governments, and WHO (145). An advocacy
for responsible media reporting is an absolute imperative.
School suicide prevention programs. It has been thought that educating people about suicidal
behavior, teaching them the cues to impending suicidal behavior, informing them of helpful
responses to depressed and suicidal individuals, and indicating the community resources
available for those in distress might reduce the incidence of suicidal behavior. Such programs
require a captive audience, and most of the programs devised to date have been for school
children. More thought should be given to providing such educational programs to others in the
community (such as church groups, union members, etc.).
Although evaluation of these programs has indicated beneficial results in some cases (139),
others have argued that they may do harm to some children (138). Providing more general
programs to raise the self-esteem of children and to teach them general coping skills (efforts
which should reduce the incidence of a variety of inappropriate behaviors such as delinquent
behavior, drug and alcohol misuse and eating disorders, as well as suicidal behavior) may prove
to be a more valuable option.
27
CM
O
o
□
s
ll
o
D
L
Figure 1 - Global suicide rates (per 100.000), by gender, 1950-1995.
o
m
O
x
XI
H
CM
30 n
co
o
Q
W
25 -
a>
aj
tr
_P’’
co
, XL
O''
□ •■•
15 -
• a --
♦
10 5 -
CD
■X)
CO
ID
0
CD
pCD
CD
CD
CD
©
CD
CD
CD
1950
1955
1960
1965
1970
1975
1980
1985
1990
1995
—♦—Total
10.1
12.3
10.9
11.6
13.2
14.1
15.8
14
13.9
16
-•fl- Male
16.6
17.5
14.9
16.7
20
24.1
21.4
21
24.7
7
6.7
23.2
8
6.8
6.9
Female
o
o
o
xj-
20 -
I
,p
. XI ,
5.2
Year
8
18/10/2000
18:18
0738753450
AISRAP
PAGE
03
[TX/RX NO 5070]
©Ot
Table 1 - Suicide Rates (per 100.000)
( most recent year available, as of August 1999)
Country
Albania
Antigua and Barbuda
Argentina
Armenia
Australia
Austria
Azerbaijan
Bahamas
Bahrain
Barbados
Belarus
Belgium
Belize
Brazil
Bulgaria
Canada
Chile
China (mainland)
China (SAR Hong Kong)
Colombia
Costa Rica
Croatia
Cuba
Czech Republic
Denmark
Dominican Republic
Ecuador
Egypt
El Salvador
Estonia
Finland
France
Georgia
Germany
Greece
Guatemala
Guyana
Honduras
Hungary
Iceland
India
Iran
Ireland
Israel
Italy
Jamaica
Japan
Jordan
Kazakhstan
Kuwait
Kyrgyzstan
Latvia ______________
Year
1993
1995
1993
1992
1995
1997
1996
1995
1988
1995
1993
1992
1995
1992
1994
1995
1994
1994
1996
1994
1994
1996
1995
1996
1996
1994
1995
1987
1990
1996
1996
1995
1990
1997
1996
1984
1994
1978
1997
1995
1995
1991
1995
1996
1993
1985
1996
1979
1996
1994
1996
1998
Males
-emalos
2.9
1.7
0
2.9
I. 0
5.1
10.0
0.3
0
0.5
3.7
9.6
II. 0
0.9
1.6
9.7
5.4
1.4
17.9
9.1
1.5
I. 8
II. 3
14.9
6.8
9.8
0
3.2
0
o
10.6
3.6
19.0
30.0
1.5
2.2
4.9
9.5
487
26.7
12.0
5.6
25.3
21.5
10.2
14.3
15.9
5.5
8.0
34.2
25.6
24.0
24.3
0
6.4
0.1
15.6
64.3
38.7
30.4
5.4
22.1
5.7
0.9
14.6
0
49.2
16.4
11.4
0.3
17.9
82
12.7
0.5
24.3
0
51.9
1.8
17.6
59.5
7.7
14.1
10.7
10.8
2.0
8.1
12
0.1
6.5
0
15.6
3.8
8.0
0.1
4.6
2.6
4.0
02
0
9.5
I. 9
3.8
II. 8
18/10 '00 WED 09:12
18/10/-2000
18:18 ..
0 /’Jb ZbJ4b0
Lithuania
Luxembourg
Malta
Mauritius
Mexico
Netherlands
New Zealand
Nicaragua
Norway
Panama
Paraguay
Peru
Philippines
Poland
Portugal
Puerto Rico
Republic of Korea
Republic of Moldova
Romania
Russian Federation
Saint Kitts and Nevis
Saint Lucia
St. Vincent and The Grenadines
Sao Tome and Principe
Seychelles
Singapore
Slovenia
Spain
Sri Lanka
Suriname
Sweden
Switzerland
Syrian Arab Republic
Tajikistan
Thailand
Trinidad and Tobago
Turkmenistan
Ukraine
United Kingdom
United States of America
Uruguay
Uzbekistan
Venezuela
Yugoslavia
Zimbabwe
_________
MlSKAr
1898
1997
1997
1996
1995
1995
1994
1994
1995
1987
1994
1989
1993
1996
1996
1992
1995
1996
1996
1995
1995
1986-88
1982-85
1987
1985-87
1997
1996
1995
1991
1992
1996
1994
1985
1992
1994
1994
1994
1992
1997
1996
1990
1993
1994
1990
1990
73.7
29.0
5.9
20.6
5.4
13.1
23.6
19.1
5.6
3.4
0.6
2.5
24.1
10.3
16.1
14.5
30.9
21.1
72.9
0
11.0
2.0
0
12.2
14.3
48.0
12.5
44.6
16.6
20.0
30.9
0.2
5.1
5.6
17.4
8.1
38.2
11.0
19.3
16.6
9.3
8.3
21.6
10.6
13.7
9.8
2.1
6.4
1.0
6.5
5.8
2.2
6.2
1.9
1.2
0.4
1.7
4.6
3.1
1.9
6.7
6.2
4.3
13.7
0
3.0
0
1.8
0
8.0
13.9
3.7
16,8
7.2
8.5
122
0
2.3
2.4
5.0
3.4
9.2
3.2
4.4
4.2
3.2
1.9
92
5.2
18/10 '00 WED 09:12
[TX/RX NO 5070] ©004
J
O
t'-
J
J
I
o
m
o
z;
■pi' <}. 3 Global suicide rates (per 100.000), by
gender and age, selected countries, 1995.
x
X
H
>—1
04
cn
o
Males
70
Q
M
js:
O
o
o
X
co
CL
<r
or
50-
CD
<E
0)
ro
DI
Females
30O
in
*
O')
tn
CD
—
•*’**
*
10
co
pCD
5-14
15-24
25-34
35-44
45-54
55-64
65-74
75+
0.9
Jk5_
14.2
18.9
JLZ6_
24,9
27.6
■ 12..4.
33.3
41
22.1
66.9
29.7
CO
CD
CD
CD
O
CM
'v.
CD
CD
12
11.6
Age group
16.4
18:18
18/10/2000
r H'ac.
AISRftP
0738753450
Table 2 - Percentage of suicides by age group and gender, selected countries’,
(most recent year available for each country).
Age (years)
5-14
15-24
25-34
35-44
45-54
55-64
65-74
75+
Total
Males
0.7
12.7
18.3
20.5
17
13.9
9.6
7.3
100
Females
0.9
13.3
15
15.4
14.7
13.9
13.7
13.1
100
All
0.8
12.8
17.5
19.2
16.4
13.9
10.7
8.7
100
•Does not Include India.
18/10 ' 00 WED 09:12
[TX/RX NO 5070] ©007
3
o
o
m
J
3
I
o
z
X
ptj
X
H
w—I
• u Percentage of suicides by age,
selected countries, 1950-1995
CM
05
o
a
M
o
o
o
X
00
CL
<1
or
CD
<r
44%
47%
%
56%
(S3
If)
S 5-44 years
03
ID
CO
CO
CD
1950
CO
CO
(S3
(S3
(S3
CM
CD
CO
■ 45 + years
1995
a>
o
o
s
J
o
o
J
3
L
io
Table 3 - Psychiatric diagnoses in 5588 cases of suicide
(Source: WHO: Primaiy Prevention of Mental, Neurological and Psychosocial Disorders. Geneva, WHO, 1998)
O
Z.
M
H
?r
O
Diagnosis
Number of
diagnosis
Percentage of total
number of diagnoses
Q
W
o
o
o
\
co
0_
<1
or
LO
<1
Q
ID
CO
ID
CD
CO
S3
Affective disorders
1400
Neurotic and personality disorders 1340
Substance abuse
Schizophrenia
Organic brain syndrome
Other mental disorders
No psychiatric diagnosis
947
612
308
1259
137
24%
22%
16%
10%
5%
21%
2%
co
00
CD
S3
's
CD
The number of diagnoses is greater than the number of cases due to multiple diagnoses, in some cases.
18/10/2000
18:18
0738753450
Standardised suicide rates (per 100 000) in Australia, by birthplace, for
all ages, 1982-1992,.
U.K.
Southern Eastern Western Oceania Asia
Europe
and
Europe
Europe
Ireland
7
14
31
19
8
12
1982
11
16
10
112
8 r
21
12
11
1983
17
17
9
5
17
1984
11
11
7
17
14
20
6
12
1985
11
8
19
14
17
12
13
6
1986
17
8
17
14
28
7
14
1987
17
14
9
20
15
8
13
1988
16
14
8
16
13
7
12
1989
14
8
19
14
5
1990
13
12
8
13
19
14
22
9
14
1991
14
7
17
8
24
13
13
1992
Source: AB S, 1994. Commonwealth of Australia copyright reproduced by
permission.
Year
Australia
18/10 '00 WED 09:12
Total
Overseas
Bom
13
12
11
12
12
13
13
12
11
12
12
[TX/RX NO 5070]
o
s
o
o
J
IO
o
z.
XI
a\
X
H
CM
CD
O
S Frequency of suicides in Samoa in relation to the arrival in the country
(1974) of pesticides containing Paraquat and the control of its sales (1982).
Q
w
3=
o
o
o
CO
CL
<1
or
ID
<1
60
Control of Paraquat availability
50 0)
fO
k—
CD
LT)
CO
tn
co
co
G)
40 -
/
(D
“U
O
□
30 -
co
CD
GD
8
2
\
\
V'
z
/
/
/
'
I
'
i
O) 20 -
i
----- 1
10 0
V
i
Arrival of Paraquat
co
1
/
/
Total suicides
»
- - +—
ir~
I
I
II
»
4^
I
I
A__ _
I
A<b
Paraquat suicides
I
I
I
rS?
<‘b
bV
I
o
d
9
1
L
o
m
- Impact of detoxification of domestic gas (% co) on suicide rates (per ioo.ooo),
England and Wales,1950-1995. (Similar results for Switzerland and Japan)
O
X
X
x
H
>—‘
ci
OJ
o
Beginning of detoxification of domestic gas
14 n
Q
M
o
o
o
12ir’
X
co
£L
<1
Qi
CO
<1
0)
OJ
10
ID
■g
8.
Total suicides
i_
g
‘zj
W
CD
in
co
tn
co
co
J3
6-
- .□ * ’
Jj. -
4 -
Gas poisoning 'v
tl
suicide
\ % of CO in domestic gas
X
K
2 -
fCD
co
co
CD
SD
CD
CM
■'x
E)
CD
0__
1950
I
55
I~
60
<
I
65
70
I“
75
80
I
Ir
~I
85
90
95
O
J
J
o
o
tn
o
- Impact on suicide of detoxification of car emission in the USA,
Z
H
a
x
H
H
as compared with non detoxification in UK.
(Suicide rates per 100,000 by inhalation of car exhaustion gases)
CM
O>
o
UK car exhaust
suicides
18
16
/
4
E
□/
Beginnig of detoxification
in the USA
o 14 a
B 10 -
/
t
t
8 □
3
T
0
0
co
t
’□
Ef
4-
6 -
USA car exhaust
suicides
--El'
4 -
0
»■)
D
D
-4
io
2 0
I
1950
55
S3
2
2
2
CO
—I---------------
’
Source: Clarke & Lester, 1987
60
65
70
T
1
75
- -----
80
I
85
o
o
o
f
J
i12-
Q
W
References
1. World Health Organization (1999). Figures and facts about suicide. WHO/MNH/MBD/99.1,
Geneva: WHO.
2. Murray CJL, Lopez AD (1996). Global burden of disease and injury series, Vol 1. Boston:
Harvard University Press.
3. World Health Organization (1999). Injury: A leading cause of the global burden of disease.
Geneva, Switzerland: Noncommunicable Diseases and Mental Health Cluster, Department of
Violence and Injury Prevention.
4. Moscicki EK (1995): Epidemiology of suicidal behavior. In Suicide Prevention: Toward tlie
Year 2000, MM Slverman, RW Maris (eds), New York: Guilford, pp 22-35.
5. Stoudemire A, Frank R, Hedemark N, et al (1986): The economic burden of depression.
General Hospital Psychiatry, 8: 387-394.
6. Minois G (1999): History of Suicide. Voluntary Death in Western Culture. Baltimore: Johns
Hopkins University Press.
7. Shneidman E (1985): Definition of Suicide. New York: Wiley & Sons.
8. Canetto SS & Lester D (1995): Women and Suicidal Behavior. New York: Springer.
9. Paykel ES, Myers JK, Lindenthal JJ, et al (1974): Suicidal feelings in the general population:
a prevalence study. British Journal of Psychiatry 124: 460-469.
10. Kessler RC, Borges G, Walters EE (1999): Prevalence and risk factors for lifetime suicide
attempts in the National Comorbidity Survey. Archives of General Psychiatry 56: 617-626.
ll.Favazza A (1999): Self-Mutilation. In The Harvard Medical School Guide to Suicide
Assessment and Intervention, DG Jacobs (ed), San Francisco: Jossey-Bass Publishers, pp
125-145.
12. Lester D (1996): Patterns of Suicide and Homicide in the World. Commack, NY: Nova
Science.
13. Lester D & Yang B (1998). Suicide and Homicide in the 20th Century. Commack, NY: Nova
Science,
14. Girard C (1993). Age, gender, and suicide. American Sociological Review, 58: 553-574.
15. Lester D (1998): Suicide in African Americans. Commack, NY: Nova Science.
16. Lester D (1997): Suicide in American Indians. Commack, NY: Nova Science.
17. Cheng TA & Hsu MA (1992): A community study of mental disorders among four aboriginal
28
groups in Taiwan. Psychological Medicine, 22: 266-263.
18. Hunter EM (1991): An examination of recent suicides in remote Australia. Australian and
New Zealand Journal of Psychiatry, 25: 197-202.
19. Rockett IRH, McKinley-Thomas B (1999): Reliability and sensitivity of suicide certification
in higher-income countries. Suicide and Life Threatening Behavior 29:141-149
20. World Health Organisation (1982): Changing patterns of suicidal behaviour. Technical
Report n.74, Copenhagen: WHO/EURO.
21. Padoani W, Marini M, Pauro P, et al (1999): Suicidio e parasuicidio in adolescenza [Suicide
and parasuicide in adolescence]. In “Il Suicidio nell’Adolescenza [Suicide in Adolescence],
AL Berman & DA Jobes (eds), Edizioni Scientifiche Magi: Roma, pp357-378.
22. Cooper PN & Milroy CM (1995): The coroner’s system and underreporting of suicide.
Medicine, Science and the Law, 35: 319-326.
23. Lozano R: Personal communication (1999). WHO: Geneva.
24. De Leo D, Diekstra RFW (1990): Depression and Suicide in Late Life. Toronto-Bern:
Hogrefe/Huber.
25. Sainsbury P & Barraclough BM (1968): Differences between suicide rates. Nature, 220:
1252.
26. Centers for Disease Control (1991): Attempted suicide among high school students - United
States 1990, leads from the morbidity and mortality weekly report. Journal of the American
Medical Association 266: 14-91.
27. Diekstra RFW, Gamefsky NA (1995): On the nature, magnitude, and causality of suicidal
behaviours: An international perspective. Suicide and Life Threatening Behavior 25: 36-57.
28. McIntire MS & Angle CR (1981): The taxonomy of suicide and self-poisoning: A pediatric
perspective. In “Self-Destructive Behavior in Children and Adolescents”, CF Wells & IR
Stuart (eds), New York: Van Nostrand Reinhold, pp 224-249.
29. McIntosh JL, Santos JF, Hubbard RW, et al (1994): Elder Suicide: Research, Theory and
Treatment. Washington DC: American Psychological Association. Pp 7-45.
30. De Leo D, Padoani W, Scocco P, et al (2001): Attempted and completed suicide in older
subjects: Results from the WHO/EURO Multicentre Study of Suicidal Behaviour.
International Journal of Geriatric Psychiatry, in press.
31. Schmidtke A, Bille-Brahe U, De Leo D, et al (1996): Attempted suicide in Europe: Rates,
trends and sociodemographic chracteristics of suicid attempters during the period 1989-1992.
Results of the WHO/EURO Multicentre Study on Parasuicide. Acta Psychiatrica
Scandinavica, 93: 327-338.
29
32. Wichstrom L (2000): Predictors of adolescent suicide attempts: a nationally representative
longitudinal study of Norwegian adolescents. Journal of the American Academy of Child and
Adolescent Psychiatry 39: 603-610.
33. Linden M & Bamow S (1997): The wish to die in very old persons near the end of life: A
psychiatric problem? Results from the Berlin Aging Study (BASE). International
Psychogeriatrics 9: 291-307
j4. Scocco P, Meneghel G, Caon F, et al (2001): Death ideation and its correlates: Survey of an
over-65-year-old population. Journal of Mental and Nervous Disease, in press.
35. Isometsa ET & Lonnqvist JK (1997): Suicide in mood disorders. In “Suicide: Biopsycosocial
Approaches”, AL Botsis, CR Soldatos, CN Stefanis (eds), Amsterdam: Elsevier, pp 33-46.
36. Waem M, Beskow J, Runeson B, et al (1999): Suicidal feelings in the last year of life in
elderly people who commit suicide. Lancet 354: 917-918.
37. World Health Organization (1998): Primary Prevention of Mental, Neurological, and
Psychosocial Disorders. Suicide. Geneva: WHO, pp 75-90.
38. Blumenthal SJ (1988): Suicide: A guide to risk-factors assessment and treatment of suicidal
Patients. Medical Clinics of North America 72: 937-971.
39. Beck AT, Brown G, Berchick RJ, et al (1990): relationship between hopelessness and
ultimate suicide: A replication with psychiatric outpatients. American Journal of Psychiatry
147: 190-195.
40. Guze SB & Robins E (1970): Suicide and primary affective disorders. British Journal of
Psychiatry 117: 437-438.
41. Harris EC & Barraclough B (1997): Suicide as an outcome for mental disorders. British
Journal of Psychiatry 170: 447-452.
42. Roy A (1986): Suicide in schizophrenia. In “Suicide”, A Roy (ed). Baltimore: Williams and
Wilkins, pp 97-112.
43. Caldwell CB, Gottesman II (1990): Schizophrenics kill themselves too: A review of risk
factors for suicide. Schizophrenia Bulletin 16:571-589
44. Beck AT, Steer R, Kovacs M, et al (1985): Hopelessness and eventual suicide: A 10-year
propspective study of patients hospitalised with suicidal ideation. American Journal of
Psychiatry, 142: 559-563.
45. Murphy GE & Wetzel RD (1990): The life-time risk of suicide in alcoholism. Archives of
General Psychiatry 47: 383-392.
46. Hawton K, Catalan J (1987): Attempted Suicide: A Practical Guide to Its Nature and
Management (2nd ed), Oxford: Oxford University Press.
30
47. Gunnell D, Frankel S (1994): Prevention of suicide: Aspiration and evidence. British Medical
Journal 308: 1227-1233.
48. Graham A, Reser J, Scuderi, et al (2000): Suicide: An Australian Psychological Society
discussion paper. Australian Psychologist 35: 1-28.
49. Roy A (1992): Genetics, biology and suicide in the family. In: Assessment and Prediction of
Suicide, RW Maris, AL Berman, JT Maltsberger, et al (eds). New York: Guilford, pp 574588.
50. Schulsinger F, Kety S, Rosenthal D, et al (1979): A family study of suicide. In Origin,
Prevention and Treatment of Affective Disorders, M Schou, E Stromgren (eds) London:
Academic Press, pp 227-287.
51. Asberg M, Traskman L, Thoren P (1976): %-HIAA in the cerebrospinal fluid. A biochemical
suicide predictor? Archives of General Psychiatry 33: 1193-1197.
52. Lester D (1995): The concentration of neurotransmitter metabolites in the cerebrospinal fluid
of suicidal individuals: a meta-analysis. Pharmacopsychiatry, 28: 45-50.
53. Coccaro EF, Siever LJ, Klar HM, et al (1989): Serotonergic studies in patients with affective
and personality disorders. Archives of General Psychiatry 46: 587-599.
54. Mann JJ, McBride PA, Broown RP, et al (1992): Relationship between central and peripheral
serotonin indexes in depressed and suicidal psychiatric inpatients. Archives of General
psychiatry 49: 442-446.
55. Mann JJ (1998): The neurobiolgy of suicide. Nature Medicine 4: 25-30.
56. Van Praag H (2000): Suicide and aggression. In Suicide Prevention, D. Lester (ed.).
Philadelphia: Brunner-Routledge, pp 45-64.
57. De Leo D, Scocco P, Marietta P, et al (1999): Physical illness and parasuicide: Evidence
from the European Parasuicide Sudy Interview (EPSIS/WHO-EURO). International Journal
of Psychiatry in Medicine 29: 149-163.
58. Appleby L, Cooper J, Amos T, Faragher B (1999): Psychological autopsy study of suicides
by people aged under 35. British Journal of Psychiatry 175:168-74
59. Beautrais AL, Joyce PR, Mulder RT (1997): Precipitating factors and life events in serious
suicide attempts among youths aged 13 through 24 years. Journal of the American Academy
of Child and Adolescent Psychiatry 36:1543-51
60. Foster T, Gillespie K, McClelland R, et al (1999): Risk factors for suicide independent of
DSM-III-R Axis I disorder. Case-control psychological autopsy study in Northern Ireland.
British Journal of Psychiatry 175:175-9
61. Heikkinen ME, Isometsa ET, Aro HM, et al (1995) Age-related variation in recent life events
31
preceding suicide. Journal of Nervous and Mental Disorder 183:325-31
62. Kaltiala-Heino R, Rimpela M, Marttunen M, et al (1999): Bullying, depression, and suicidal
ideation in Finnish adolescents: school survey. British Medical Journal 319:348-351.
63. Cavanagh JT, Owens DG, Johnstone EC (1999): Life events in suicide and undetermined
death in southeast Scotland: a case-control study using the method of psychological autopsy.
Social Psychiatry and Psychiatric Epidemiology 34:645-650.
64. Thacore VR, Varma SL (2000): A study of suicides in Ballarat, Victoria, Australia. Crisis
21:26-30.
65. Kemic MA, Wolf ME, Holt VL (2000): Rates and relative risk of hospital admission among
women in violent intimate partner relationships. American Journal of Public Health, 90:14161420.
66. Olson L, Huyler F, Lynch AW, et al (1999): Guns, alcohol, and intimate partner violence: the
epidemiology of female suicide in New Mexico. Crisis 20: 121-126
67. Thompson MP, Kaslow NJ, Kingree JB, et al (1999): Partner abuse and posttraumatic stress
disorder as risk factors for suicide attempts in a sample of low-income, inner-city women.
Journal of Trauma and Stress 12: 59-72
68. Fischbach RL, Herbert B (1997): Domestic violence and mental health: correlates and
conundrums within and across cultures. Social Science in Medicine 45: 1161-76
69. Brown J, Cohen P, Johnson JG, et al (1999): Childhood abuse and neglect: specificity of
effects on adolescent and young adult depression and suicidality. Journal of the American
Academy of Child and Adolescent Psychiatry 38: 1490-6
70. Dinwiddie S, Heath AC, Dunne MP, et al (2000): Early sexual abuse and lifetime
psychopathology: a co-twin-control study. Psychological Medicine 30: 41-52.
71. Santa Mina EE, Gallop RM (1998): Childhood sexual and physical abuse and adult self-harm
and suicidal behaviour: a literature review. Canadian Journal of Psychiatry 43: 793-800.
72. Gamefski N, Arends E (1998): Sexual abuse and adolescent maladjustment: differences
between male and female victims. Journal of Adolescence 21: 99-107
73. Swanston HY, Tebbutt JS, O’Toole BI, et al (1997): Sexually abused children 5 years after
presentation: a case-control study. Pediatrics, 100: 600-608
74. Silverman AB, Reinherz HZ, Giaconia RM (1996): The long-term sequelae of child and
adolescent abuse: a longitudinal community study. Child Abuse and Neglect 20: 709-23.
75. Report of the Secretary’s Task Force on Youth Suicide (1989). Volume 3: Prevention and
Interventions in Youth Suicide Rockville, Md; US Dept of Health and Human Services.
DHSS publication ADM 89-1622.
32
93. Negron R, Piacentrini J, Graae F, et al (1997): Microanalysis of adolescent suicide attempters
and ideators during the acute suicidal episod. Journal of the American Academy of Child and
Adolescent Psychiatry 36: 1512-1519
94. Wenz F (1977): Marital status, anomie, and forms of social isolation: A case of high suicide
rate among the widowed in urban sub-area. Diseases of the Nervous System 38: 891-895
95. Draper B (1996): Attempted suicide in old age. International Journal of Geriatric Psychiatry
11:577-587.
96. Dennis MS, Lindsay J (1995): Suicide in the elderly: the United Kingdom perspective.
International Psychogeriatrics 7: 263-74.
97. Hawton K, Fagg J, Simkin S (1996): Deliberate self-poisoning and self-injury in children and
adolescents under 16 years of age in Oxford 1976-93. British Journal of Psychiatry 169: 202208.
98. De Leo D, Neulinger K, Kosky R, et al (1999): Hanging as a Means to Suicide in
YoungAustralians. A Report to the Commonwealth Ministry of Health and Family Services.
Brisbane: Australian Institute for Suicide Research and Prevention.
99. Yip PS & Tan RCE (1998): Suicides in Hong Kong and Singapore: A tale of two
cities.Intemational Journal of Social Psychiatry 44: 267-279.
100. Schmidtke A & Hafner H (1988): The Werther effect after television films: new
evidence for an old hypothesis. Psychological Medicine 18: 665-676.
101.
Wasserman I (1984): Imitation and suicide: A re-examination of the Werther effect.
American Sociological Review 49: 427-436.
102. Mazurk PM, Tardiff K, Hirsch CS, et al (1993): Increase of suicide by asphyxiation in
New York City after the publication of “Final Exit”. New England Journal of Medicine 3291508-1510.
103. De Leo d & Ormskerk S (1991): Suicide in the elderlv: General characteristics. Crisis 12:
3-17.
104. Bowles JR (1995). Suicide in Western Samoa: an example of a suicide prevention
program in a developing country. In: Preventive Strategies on Suicide, RFW Diekstra, W
Gulbinat, I Kienhorst, et al (eds), Leiden: copublication WHO/E J Brill, pp 173-206.
105. Dudley MJ, Kelk NJ, Florio TM, et al (1998): Suicide among young Australians, 19641993: An interstate comparison of metropolitan and rural trends. Medical Journal ofAustralia
169:77-80
106. American Association of Suicidology (1999). Facts Sheet: Rates of suicide throughout
the country.
34
107. Hawton K, Simkin S, Malmberg A, et al (1998): Suicide and Stress in Farmers. London:
The Stationery Office.
108. Cantor CH, Neulinger K, Roth J, et al (1998): The Epidemiology of Suicide and
Attempted Suicide Among Young Australians. A report to the NH-MRC, Australian Institute
for Suicide Research and Prevention, Brisbane.
109. Platt S (1984): Unemployment and suicidal behaviour: A review of the literature.
SocialScience and Medicine 19: 93-115.
110. Weyerer S, Wiedenmann A (1995): Economic factors and the rate of suicide in Germany
between 1881 to 1989. Psychological Report 76: 1331-1341.
111. Lopatin AA, Kokorina NP (1998). The widespread nature of suicide in Kuzbass (Russia).
Archives of Suicide Research 3: 225-234.
112. Berk JH (1998): Trauma and resilience during war: a look at the children and
humanitarian aid workers in Bosnia. Psychoanalytical Review 85: 648-658.
113.
Lester D (1987): Religion, suicide and homicide. Social Psychiatry 22: 99-101.
114. Faupel CE, Kowalski GS, Starr PD (1987): Sociology’s one law: Religion and suicide in
the urban context. Journal of the Scientific Study of Religion 26: 523-534.
115. Burr JA, McCall PL, Powell-Griner E (1994): Catholic religion and suicide: The
mediating effect of divorce. Social Science Quarterly 75: 300-318
116. Simpson ME, Conklin GH (1989): Socioeconomic development, suicide and religion: A
test of Durkheim’s theory of religion and suicide. Social Forces 67: 945-964
117. Bankston WB, Allen HD, Cunningham DS (1983): Religion and suicide: A research note
on ‘Sociology’s One Law’ Social Forces 62: 521-528
118.
Pope W, Danigelis N (1981): Sociology’s ‘one law’. Social Forces 60: 495-516
119. Stack S, Wasserman I (1992): The effect of religion on suicide ideology: An analysis of
the networks perspective. Journal of Scientific Study of Religion 31: 457-466
120. Stack (1983): The Effect of Religious Commitment of Suicide: A cross national analysis.
Journal of Health and Social Behaviour 24:362-274
121. Kok LP (1988): Race, religion and female suicide attempters in Singapore. Social
Psychiatry and Psychiatric Epidemiology 40:236-239.
122. Neeleman J, Wessely S , Lewis G (1998): Suicide acceptably in African and white
Americans: The role of religion. Journal of Nervous and Mental Disease 186: 12-16
35
123. Linehan MM (1997). Behavioral treatments of suicidal behaviors: defenitional
obfuscation and treatment outcomes The Neurobiology of suicide: From the bench to the
clinic, New York Academy of Sciences, New York, 302-328
124. Salkovskis PM, Atha C, Storer D (1990). Cognitive behavioural problem-solving in the
treatment of patients who repeatedly attempt suicide. A controlled trial. British Journal of
Psychiatry, 157, 871-876.
125. Linehan MM, Heard HL, Armstrong HE (1993). Naturalistic follow-up of a behavioural
treatment for chronically parasuicidal borderline patients. Archives of General Psychiatry 50
971-974.
126. MacLeod AK, Tata P, Evans K, et al (1998). Recovery of positive future thinking within
a high-risk parasuicide group: results from a pilot randomised controlled trial. British Journal
of Clinical psychology, 37, 371-379.
127. Morgan HG, Jones EM, Owen JH (1993). Secondary prevention of non-.fatal deliberate
self-harm. The Green Card Study. British Journal of Psychiatry, 163, 111-112.
128. Cotgrove A, Zirinsky L, Black D, et al (1995). Secondary prevention of attempted suicide
in adolescence. Journal of Adolescence, 18, 569-577.
129. ]Evans MO, Morgan HG, Hayward A, et al (1999). Crisis telephone consultation for
deliberate self-harm patients: effects; on repetition. British Journal of Psychiatry, 175, 23-27.
130. De Leo D, Carollo G, Dello Buono M (1995): Lower suicide rates associated with a telehelp/tele-check service for the elderly at home. American Journal of Psychiatry, 152, 632634.
131. Verkes RJ, van der Mast RC, Hengveld MW, et al (1998). Reduction by paroxetine of
suicidal behavior in patients with repeated suicide attempts but not with major depression.
American Journal of Psychiatry, 155, 543-547.
132. Litman RE, Wold CI (1976). Beyond crisis intervention. Suicidology, Contemporary
developments. ES Schneidman (Ed). Grune & Stratton. New York. 528-546.
133. Gibbons JS, Butler J, Urwins P, et al (1978). Evaluation of a social work service for self
poisoning patients. British Journal of Psychiatry, 133, 111-118.
134. Hawton
1
K, McKeown S, Day A, et al (1987). Evaluation of outpatient counselling
compared with general practitioner■ care following overdoses. Psychological. Medicine, 17,
751-761.
135. Dew MA, Bromet EJ, Brent D, et al (1987). A quantitative literature review of the
effectiveness of suicide prevention centers. Journal of Consulting and Clinical Psychology,
55, 239-244.
36
136. Lester D (1997). The effectiveness of suicide prevention centres; A Review. Suicide and
Life Threatening Behaviour, Vol 27, 304-310.
137. Riehl T, Marchner E, Moller HJ (1988): Influence of crisis intervention telephone
services (“crisi hotlines”) on the suicide rate in 25 German cities. In Current Issues of
Suicidology, HJ Moller, A Schmidtke, R Welz (eds), New York: Springer Verlag, pp 431436.
138. Lester D (1992). State initiatives in addressing youth suicide: Evidence for their
effectiveness. Social Psychiary and Psychiatric Epidemiology, 27, 75-77.
139. Zenere F, Lazarus PJ (1997). The decline of youth suicidal behaviour in an urban,
multicultural public school system following the introduction of a suicide prevention and
intervention program, Suicide and Life Threatening behaviour, Vol 27, 387-403.
140. Oliver RG, Hetzel BS (1972). Rise and fall of suicide rates in Australia: relation to
sedative availability. Medical Journal of Australia 2:919-923.
141. Kreitman N (1972). The coal gas history: United Kingdom suicide rates, 1960-1971.
British Journal of Preventive and Social Medicine 30:86-93.
142.
Lester D (1998): Preventing suicide by restricting access to methods for suicide.
Archives of Suicide Research, 4: 7-24.
143. Clarke RV, Lester D (1987). Toxicity of car exhausts and opportunity for suicide. Journal
of Epidemiology and Community Health 41:114-120.
144. Lester D, Murrell ME (1980). The influence of gun control laws on suicidal behavior.
American Journal of Psychiatry 80:151-154.
145. World Health Organisation (2000). Preventing suicide:A
professionals. In the series Mental and Behavioural
WHO/MNH/MBD/OO.2.
resource for media
Geneva:
Disorders,
146. Taylor SE (1999). Health behaviour and primary prevention in Health Psychology.
McGraw-Hill Companies, US.
147. World Health Organisation (2000). Preventing suicide: How to start a survivors group. In
the series Mental and Behavioural Disorders, Geneva: WHO/MNH/MBD/OO.6
148. United Nations (1996). Prevention of Suicide - Guidelines for the formulation and
implementation of national strategies (Doc.: ST/SEA/245), New York: UN.
149. World Health Organization (1998): Primary prevention of Mental, Neurological, and
Psycho-social disorders. Suicide. Geneva: WHO, pp 75-90.
150.
De Leo D (1999): Cultural issues in suicide and old age. Crisis, 20: 53-55.
151.
Schmidtke A, Weinacker B, Apter A, et al (1999): Suicide rates in the world: Un update.
37
Archives of Suicide Research, 5: 81-89.
152. Isacsson G (2000): Suicide prevention: A medical breakthrough? Acta Psychiatrica
Scandinavica 102: 113-117.
153. Rutz W (2000): The role of family physicians in preventing suicide. In Suicide
Prevention: Resources for the Millennium. D Lester (ed). Philadelphia: Brunner-Routledge,
pp 173-187.
38
Box 1
Suicide in China
Suicide is a major public health problem for China. The Global Burden of Disease project which makes some controversial adjustments to Chinese mortality figures - estimates a 1998
national suicide rate of 32.9 per 100,000 (413,000 deaths), indicating that suicides account for
4.4% of all deaths and 4.2% of all DALYs lost. This makes suicide the fourth most important
health condition in the country. The pattern of suicides is different than elsewhere: unadjusted
mortality data for 1994 (which report a national rate of 21.22 per 100,000) report rural rates 4fold urban rates (27.02 v. 6.74), female rates 26% higher than male rates (23.72 v. 18.81) and
particularly high rates in young rural females 15-44 (34.95) and in the rural elderly 65 and over
(97.33).
Traditionally suicide in China is a culturally acceptable response to a variety of extreme
situations, but other factors - the low rate of treatment for depressive illness and the new social
stressors that have arisen during the economic reforms - also play an important role. The
predominance of women is partly explained by the high proportion of female suicides as
impulsive acts perpretrated by persons without identifiable mental illness (about 40%); these
would probably be suicide attempts in other countries but become completed suicides in China
because of the highly lethal methods employed in rural areas (predominantly insecticides and
herbicides).
The only prevention services currently available are telephone hot-lines and a few crisis centers
in some of the big cities; nothing is available in the countryside. Recently, the Ministry of Health
has given high priority to the development and implementation of a national suicide prevention
program.
Michael R. Phillips, M.D., M.A., M.P.H.
Director
Research Center of Clinical Epidemiology
Beijing Hui Long Guan Hospital
Beijing 100096
People’s Republic of China
39
Box 2
Suicide Among the Inuit
There are few observations or stories about suicide in the Arctic before colonization. Yet, in the
early to mid 1900’s, recorders of suicide among the Inuit suggested that suicide was a trait of the
then called Eskimos. These early reports, however, were based on loosely collected data from
diverse events, not only self-inflicted death (1).
In the last 30 years, the rates of suicide have increased dramatically in Canada’s northern
population (the 1991 census recorded some 43,000 Canadians with Inuit origins). For example,
rates of 59.5 to 74.3 (per 100,000) have been reported in various communities in the Arctic,
compared to around 13.5 (per 100,000) in the general Canadian population. The highest risk
group is the young males and this phenomenon is increasing, something noted in other
indigenous groups (2). Rates as high as 195 (per 100,000) for 15 to 25 year olds have been
recorded. Furthermore, there is little study of attempters in the Arctic (and Canada as a whole),
although these too suggest alarming rates. For example, one study (3) reported a lifetime
prevalence of 14% in an Inuit group, compared to one of about 3% in the general Canadian
population. Thus, it is easy to conclude that the rates of suicide and suicide attempts are high in
the Arctic compared to international rates, although the reliability of the data remain problematic
as underreporting is likely.
Many explanations have been advanced to account for the high rates of suicide and suicidal
behaviour in the Arctic. Understanding the tragedy first requires developing a historical context:
The Arctic has had foreign visitors dating back to the days of the Vikings. However, large-scale
colonization occurred only in the beginning of the 19th century. Since that time, attitudes and
policies of racial and cultural superiority have led to exploitation and suppression of the Inuit
people, their culture and values. The cultural transitions and change were enormous and as
Durkheim (4) demonstrated, when social integration and regulation are too weak or too strong,
suicide is predictable.
The exploitation began in the 19th century with the whaling expeditions and fur trade. Great
diseases occurred with the foreigners’ arrivals, taking tens of thousands of lives, leaving only
about one-third of the population by 1900. The epidemics continued during the first half of the
20 century. The fur trade collapsed in the 1930’s, and Canada introduced a welfare state in the
Arctic. The missionaries came in large scale in the 1940’s and 50’s and an attempt at
assimilation occurred. Oil exploration began in 1959, further adding to social disintegration (4).
The most recent attempt at "integrating” the Inuit was the residential schools, which were
especially suicidogenic (5). The schools separated many children from their families and
communities, and prevented the children from learning their languages, heritage and culture.
Similar assimilation strategies occurred elsewhere with indigenous people; for example, in
Australia, children were taken from their parents’ homes and were placed in non-Aborigine foster
homes (2, 6). Tragically, the cases of physical and sexual abuse in these systems in the Arctic
(and Australia) were quite high and have been associated to the high rates of suicide (1,2, 5, 6).
40
Box 3
Aboriginal Suicide in Australia
Suicide within the Aboriginal and Torres Strait Islander populations of Australia has only within
the last two decades emerged as an issue of public heath concern. While willed or self-willed
death associated with sorcery or physical debility in traditional Indigenous societies might be
considered a ‘suicide equivalent’ phenomenon, it is in sharp contrast to the deaths by hanging of
young men which has now captured national attention.
The enumeration of the Indigenous population of Australia and collection of reliable suicide
statistics remains problematic. However, most recent estimates indicate a total Aboriginal and
Torres Strait Islander population of 386,049, 2.1% of all Australians. The State of Queensland is
home to 104,817 people of Indigenous descent, being 27% of the national total and constituting
3% of the State’s population. The Queensland suicide rate for the period 1990 to 1995 is 14.5 per
100,000, with the Aboriginal and Torres Strait Islander rate being 23.6. The elevated rate is
entirely accounted for by the increased Indigenous male suicides which are concentrated in the 15
to 24 (112.5 per 100,000) and 25 to 34 (72.5 per 100,000) year age periods. These are 3.6 and
2.2 times the rates of these male age-groups for the State as a whole (1), these age-groups
comprising 84% of all Indigenous suicides. Because of the problems with identification of
Aboriginality in death records, these figures are almost certainly an underestimate.
As noted at the outset, Aboriginal and Torres Strait Islander suicide was, until two decades ago,
very uncommon. Understanding why that is no longer the case demands developing an historical
context that foreground a period of enormous transition across the country that occurred, roughly,
through the 1970s. Previously, Indigenous lives and communities had been controlled through
draconian controls and racist legislation which began to lift with little planning or preparation in
the late 1960s. The next decade, a period of ‘deregulation’ (2), was characterised by political and
social instability, the lifting of restricted access to alcohol, rapidly increasing rates of violence
and accidents, high rates of incarceration, and many other manifestations of continuing
disadvantage and underlying turmoil, all with very serious consequences for the stability of
communities and family life. Sadly, much of this has continued. The contemporary context also
includes markedly elevated rates of morbidity and mortality from most causes, including a
homicide rate that is higher by factor of ten, with life expectancy less by nearly two decades that
that expected for non-Indigenous Australians.
Against this background, Indigenous suicide began to be recognised as an issue through the late
1980s. At that time suicides tended to occur in non-remote settings among non-traditional groups
and was often associated with the acute effects of alcohol consumption. At the end of that decade
a national inquiry, the Royal Commission into Aboriginal Deaths in Custody, was convened to
investigate Indigenous deaths in police and prison custody, a significant proportion of which was
suicide by hanging. The national media focus on the Royal Commission provided for the
development of political understandings of hanging that foregrounded the effects of colonisation
and oppression. Thus the contemporary ‘meaningfulness’ of hanging by young Indigenous people
whose manifest disadvantage by comparison to the wider society is often experienced as a result
43
of oppression and discrimination. Since the Commission suicide has continued to increase in the
wider Aboriginal and Torres Strait Islander population, with hanging being by far the most
common method.
Those taking their lives are usually young men who have grown to maturity during or since the
period earlier referred to as ‘deregulation’. They are members of the first generation to be
exposed to the developmental consequences of widespread community and family instability,
much of which reflects the indirect effects of heavy alcohol use (particularly on paternal roles
and, consequently, for the construction of Indigenous male identity). There also appears to be a
cohort effect as this group ages. Indigenous suicide may now be becoming more common at
somewhat later age; some 40% of the male suicides in the Queensland study (1) were of men
twenty five years of age or older. Furthermore, no indigenous settings are unaffected by the
processes of social change, and indigenous suicide now appears to be generalising and becoming
more common in certain remote and ‘traditional’ populations, sometimes taking on ‘traditional’
meanings. This picture of suicide in the indigenous populations of Australia bears distinct
similarities to that among indigenous populations in similar mainstream cultures elsewhere in the
world, for instance in Canada in 1995 (3). These and other health parallels suggest the importance
of common experiences of colonisation and its consequences. They also reinforce the centrality
of history and meaning in any analysis of indigenous suicide.
References
1. Baume PJM, Cantor CH & McTaggart PG. (1997). Suicides in Queensland: A comprehensive
study 1990-1995. Australian Institute for Suicide Research and Prevention, Brisbane.
2. Hunter E. (1999). Considering the changing environment of Indigenous child development.
Australasian Psychiatry, 7, 137-140.
3. Royal Commission on Aboriginal Peoples. (1995). Choosing life: Special report on suicide
among Aboriginal people. Ottawa: Canada Communication Group.
Ernest Hunter
University of Queensland, Australia
44
Box 4
Depression and Suicide
The many faces of depression
Depression is the single mental disorder most often associated with suicide. Several studies show
that up to 80% of people who committed suicide had several depressive symptoms, and up to
50% had fully diagnosable Major Depressive Disorder (MDD), dysthymia or “double
depression” (i.e. dysthymia and recurrent MDD), often combined with anxiety disorder. Anxiety,
a powerful driving force in the suicidal process, is intimately interwoven with depression.
Sometimes depression and anxiety disorder are indistinguishable.
There is a great deal of suffering in the world, and everyone can be a little despondent from time
to time. But for depression to be diagnosed one must (according to DSM IV), for at least two
weeks, show a minimum of five of the following nine symptoms: despondency or irritability,
anhedonia, changed appetite, disturbed sleep, modified motor activity, reduced energy, a sense of
guilt, concentration difficulties and thoughts of death.
It is often difficult to detect and/or diagnose depression in men — who, moreover, seek medical
care more seldom than women. Men’s depressions are not infrequently preceded by acting-out
behaviour, various types of abuse, and violence within and outside the family. Treatment of
depression in men is highly important, since suicide is a typically male phenomenon in most
cultures. For this reason, treatment of depression among men can be commenced even if the
aforesaid criteria for diagnosing depression are not fulfilled.
Among children and teenagers, the faces of depression partially differ from those among adults.
Compared with their elders, depressed children and adolescents tend to show more acting-out
(truancy, bad behaviour, violent tendencies, declining grades, misuse of alcohol or drugs), but
also to sleep and eat more. However, refusal to eat and anorexia, with the concomitant high risk
of suicide, are not unusual in combination with depression — notably among girls, but also in
boys.
Both among young people and — especially — older adults, depression commonly has physical
manifestations, such as stomach ailments, dizziness, cardiac palpitations and pain in various parts
of the body. Depression in the elderly may accompany somatic illnesses, such as stroke, cardiac
infarct, cancer, rheumatism, Parkinson’s, Alzheimer’s, etc.
Suicide risk in depressed people
Relatively few people suffering from depression alone commit suicide. Suicide is committed by
depressed people who have been subjected to stress by an unfavourable psychosocial situation,
such as being poor, unemployed, an outsider, lonely, an immigrant, victimised or disadvantaged
in some other way. They may have experienced occupational and residential segregation, a lack
of social and cultural integration, destructive family patterns, life in rural areas, prison and police
custody, or bereavement. They may belong to indigenous ethnic groups or suffer from conflicts
of sexual identity. A sensitive disposition, with poor stress tolerance and an inability to overcome
45
extra strains in life owing to rigid coping strategies, also contributes to suicidality. In the
psychiatric literature, this kind of disposition is described as a borderline, narcissistic or histrionic
personality type.
Suicide prevention
To prevent suicide, suicidal people’s depression must be detected, diagnosed and treated,
preferably with a combination of antidepressant drugs and psychotherapy. Treatment of anxiety
exacerbated, in suicidal people, by an unfavourable psychosocial situation is also an important
means of suicide prevention (1).
Unfortunately, the fact remains that the majority of depressed people who take their own lives
have not usually received any treatment, either with antidepressant drugs or by psychological
means, for their depression.
Nevertheless, some studies show that suicidality is reduced when depression is treated. In
longitudinal follow-ups of large groups of depressed patients, 15% mortality from suicide is
commonly observed. The report by Angst et al (2) on 5% mortality from suicide among
depressed people given long-term antidepressant treatment for prophylactic purposes is
encouraging. A range of meta-analyses (3-5) also show that a statistically significant reduction in
the number of suicidal thoughts and also a reduction — not statistically significant but
nonetheless marked — in the number of suicide attempts and suicides are attainable among
depressed people treated with antidepressants, in contrast to those given placebo treatment only.
Various forms of psychotherapy, especially the cognitive kind, have proved effective in the
treatment of depression. Cognitive behavioural therapy has documented effects in reducing
suicidality among women with borderline personality structure and several suicide attempts in the
anamnesis (6). An extensive study has shown that good psychosocial support of elderly people,
with increased telephone access to staff, brings about a significant reduction in their depression
and mortality from suicide, as well as enhanced well-being and a decline in their need for hospital
care (7).
Future outlook
Depression is a major, widespread illness and will continue to account for a high proportion of
the global burden of ill-health during the millennium to come (8). Several population studies
show that around 25% of women and 15% of men may be expected to become depressed
sometime in their lives. Many studies suggest that depression is on the rise in the western world,
especially among young people. Ever increasing stress at the workplace, at school and in family
life figures largely in this trend since, it plays a large part in causing depression and precipitating
suicide.
Preventive programmes that are population-oriented, aimed at the public and young people in
particular, should include instruction in how to deal with stress and recognise signs of mental illhealth. Learning how to understand a suicidal person’s communication and cries for help, which
are often poorly expressed in verbal terms, and grasping how important it is to seek help in time
are essential. Besides dissemination of knowledge about the incidence, symptoms and causes of
depression and the treatment options available, population-oriented suicide prevention should
46
include efforts to change people’s stigmatising attitudes towards mental illness and induce them,
as fellow human beings, to offer a helping hand to those who are lonely and socially excluded.
Danuta Wasserman
References
1. Wasserman D (ed). Suicide: An unnecessary death.. Dunitz, London 2000.
2. Angst J, Angst F, Stassen HH. Suicide risk in patients with major depressive disorder. J Clin
Psychiatry 1999, 60 Suppl 2: 57-62.
3. Beasley CM, Domseif BE, Bosomworth JC et al. Fluoxetine and suicide: a meta-anaysis of
controlled trials of treatment for depression, Br Med J1991, 303:685-92.
4. Isacsson G, Holmgren P, Druid H, Bergman U. Psychotropics and suicide prevention. Br J of
Psychiatry 1999, 174:259-265.
5. Montgomery SA, Dunner DL, Dunbar GC. Reduction of suicidal thoughts with paroxetine in
comparison with reference antidepressants and placebo, Eur Neuropsychopharmacol 1995, 5:513.
6. Linehan MM, Armstrong HE, Suarez A, Allmon D, Heard HL. Cognitive-behavioral treatment
of chronically parasuicidal borderline patients. Arch Gen Psychiatry 1991, 48: 1060-1064.
7. DeLeo D, Carollo G, Dello Buono M. Lower suicide rates associated with a Tele-Help/TeleCheck service for the elderly at home. American Journal ofPsychiatry 1995; 152: 632-634.
8. Murray CJL, Lopez AD. The Global Burden ofDisease. Harvard University Press 1996.
47
decrease in suicide rates was observed for males in the 25-54 age group. One explanation for this
decrease in suicide could be that this age group is more easily influenced to the changes in
attitudes that took place in the former Soviet Union during the anti-alcohol campaign. Another
explanation may be that this group is more price-sensitive and was hit harder by price increases.
Decreases in suicide mortality during perestroika cannot be explained by any change in data
quality or modification of routines for compiling mortality statistics. Quantitative and qualitative
studies have shown that the reliability of statistics on suicide and other violent causes of death in
Russia was good in the Baltic and Slavic republics, and also Kazakhstan, Kirgizia and Moldavia.
Effects of perestroika short-lived
Effects of the Gorbachev’s anti-alcohol campaign in the perestroika period did not last long,
probably owing to several factors. One was that unofficial liquor production took over the
Russian market; another is that inadequate financial resources ruled out funding of campaigns to
mould public opinion against alcohol consumption and modify public attitudes towards alcohol.
Considerable stress and adjustment problems for individuals due to substantial economic
difficulties and social changes were probably also contributing to the fact that deaths due to
suicide and other violent causes rose sharply in the republics of the former USSR during the
1990s.
Suicide prevention at society and individual level
A restrictive alcohol policy may be one means of reducing suicide and violent deaths among
alcoholics. However, suicide-preventive measures at individual level along with public health are
essential. Several clinical studies at individual level show that alcohol consumption is correlated
with suicide for people characterised by alcohol abuse and alcohol dependency who, at the same
time, have personality disorders and/or depression or other mental illnesses.
It is therefore plausible to conclude that measures cannot concentrate solely on reducing alcohol
consumption by changing attitudes towards alcohol use, or by regulating the availability of
alcohol in terms of market price or sale. Individual measures which include medical treatment of
any underlying dependency or other psychiatric disorder; reinforcement of protective and
supportive factors; and assistance in developing constructive coping strategies are of equal
importance. Moreover, in any attempt to modify attitudes, measures should be adapted to
patients’ cultural and psychosocial circumstances and fully explore the reasons why they drink.
In this context, it is of interest that, the degree of cultural acceptance of drinking in any society
has a bearing on the number of its members who become alcoholics.
49
References
1. Leon D., Chenet., Shkolnikov V. et al. Huge variation in Russian mortality rates 1984-1994:
artefact, alcohol or what? Lancet 1997;350:383-388.
2. Lester D. The association between alcohol consumption and suicide and homicide rates: a
study of 13 nations. Alcohol Alcohol 1995;30:465-468.
3. Norstrom T. Alcohol and suicide in Scandinavia. Br J Addict. 1988;83:553-559
4. Ryan M. Alcoholism and rising mortality in the Russian Federation. Br Med J 1995;310:646648.
5. Skog O-J, Elekes Z. Alcohol and the 1950-90 Hungarian suicide trend - is there a causal
connection? Acta Sociol 1993;36:33-46.
6. Skog O-J. Alcohol and suicide - Durkheim revisited. Acta Sociol 1991;34:193-206.
7. Skog O-J. Alcohol and the Suicide in Denmark 1911-1924 - experiences from a ‘natural
experiment’. Addiction 1993;88.T 189-1193.
8. Wasserman D. Alcohol and suicidal behaviour. Nord J Psychiatry 1993;47:265-271. Oslo.
ISSN 0803-9488
9. Wasserman D., Varmk A., Dankowicz M., Eklund G. Suicide-preventive effects of
perestroika in the former USSR: the role of alcohol restriction. Acta Psychiat Scand Suppl
No.394-Vol.98-1998
10. Wasserman D., Varmk A., Eklund G. Male suicides and alcohol consumption in the former
USSR. Acta Psychiat Scand 1994;89:306-313.
11. Wasserman IM. The impact of epidemic, war, prohibition and media on suicide: United
States, 1910-1920. Suicide Life Threat Behav. 1992;22:240-254.
D. Wasserman
50
Box 6
Suicidal Behaviour in Children and Young People
During the last two decades there have been increasing concerns, internationally, about suicidal
behaviours amongst young people. While attention has tended to focus upon completed suicide,
there is evidence of a range of suicidal behaviours that extend from ideas and thoughts about
suicide, which are not acted upon to suicide attempts and suicide. The weight of evidence
suggests that up to 25% of young people may have suicidal ideation. For most young people,
therefore, suicidal ideation is common, and, in the absence of other risk factors for suicidal
behaviour, not a risk factor for subsequent suicide. Up to 7% of young people may make suicide
attempts, with the majority of these attempts resulting in no more than minor physical harm.
Suicide occurs far less frequently than suicidal ideation and suicide attempt behaviour. The risk
of suicide amongst children, adolescents and young people tends to increase with age, and is rare
among children aged less than 15 years. In most countries, the male rate of youth suicide exceeds
the female rate, by a ratio of, typically, 4:1. In contrast, young females are approximately twice as
likely as young males to make suicide attempts and to report suicidal ideation.
International research studies consistently suggest that the aetiology of youth suicidal behaviour
is complex and multicausal with a range of elements combining in various ways to influence the
risk of such behaviour. This evidence suggests that risk factors for suicide and attempted suicide
in young people may be classified into several broad domains of related risk factors:
Genetic and biologic factors may influence individual vulnerability to suicidal behaviour.
Evidence from twin, adoption, family and molecular genetic studies suggests that suicidal
behaviours tend to run in families, implying a possible role of genetic factors in suicidal
behaviour. The genetic factor involved may be an inability to control impulsive behaviour, which
is fostered or triggered by depression, other mental illnesses or stress. In addition, there is some
evidence to suggest that a number of neuroendocrine and biologic factors (particularly, serotonin
and its metabolite) may make contributions to suicidal behaviour.
Social and demographic factors may provide social contextual factors that influence both an
individual’s predisposition to suicidal behaviour and their expression of such behaviour. Rates of
suicidal behaviour tend to be elevated amongst young people from socially disadvantaged
backgrounds characterised by low socioeconomic status, limited educational achievement and
low income.
Family characteristics and childhood experiences including parental disharmony and separation,
parental psychopathology, poor inter-familial communication, and exposure to sexual and/or
physical abuse during childhood may influence an individual’s longer-term vulnerability to
psychiatric disorder and suicidal behaviour. Often, young people at risk of suicidal behaviour
tend to come from multiple problem family backgrounds in which several of these family risk
factors are commonly present. This observation suggests, firstly, that it is the density and
chronicity of exposure to a range of risk factors, rather than the occurrence of a single risk factor,
which contributes to increased family dysfunction and the development of subsequent mental
health problems and suicidal behaviour. Secondly, the adverse family backgrounds which
51
characterize young people at risk of suicidal behaviour are very similar to the those which occur
in other adolescent psychosocial disorders (including, for example, depressive disorders,
substance use disorders and offending behaviours), suggesting that the major life pathways which
lead to serious suicidal behaviour overlap and correlate, very substantially, with those which lead
to a range of adolescent psychosocial and mental health problems.
Stressful or adverse life events or circumstances may precipitate suicidal behaviour. Most
commonly, these events include interpersonal losses and conflicts and disciplinary or legal crises.
However, such events occur commonly amongst young people and may precipitate suicidal
behaviour only when they occur in those individuals vulnerable to suicidal behaviour.
Psychiatric morbidity, including, in particular, affective disorders, substance use disorders,
antisocial behaviours and anxiety disorders, are frequent precursors of suicidal behaviour. The
majority (80-90%) of young people who die by suicide or make serious suicide attempts have at
least one mental disorder. Depressive disorders consistently emerge as those most commonly
associated with suicidal behaviour. In addition, young people with psychotic disorders, including
schizophrenia, have an elevated risk of suicidal behaviour.
Frequently, young people with serious suicidal behaviour have histories of extensive mental
illness including co-morbid or multiple mental disorders, a history of previous suicide attempts,
and a history of prior contact and care with mental health services.
Personality factors and cognitive styles, which reflect individual variations in temperament or
related factors, may act to encourage the development of suicidal behaviour. Personality
disorders may be present in up to one third of those who die by suicide, with the most common
disorders being borderline and antisocial disorders.
In contrast to the large volume of research into risk factors for suicidal behaviours in young
people there is comparatively little research focused upon identifying individual, family and
community factors which may protect against the development of suicidal behaviour in young
people. There is emerging interest in this area with available evidence suggesting that the factors
likely to protect against suicide include social supports, family cohesion, peer affiliation, good
adaptive, social and coping skills and problem solving behaviours, positive and life affirming
beliefs and values; high self-esteem, and holding attitudes and moral values against suicide.
Annette L. Beautrais
Christchurch School of Medicine, New Zealand
52
Box 7
Access to Means of Suicide
While suicidal thoughts are relatively common in the general population, only a small
proportion of those experiencing suicidal thoughts attempt suicide, and only a proportion
of those who attempt it actually kill themselves. One of the variables that influences
whether people act on their suicidal impulses, and with what degree of lethality is the
availability and access to different methods of committing suicide.
Despite the many potential methods of suicide, only relatively few are commonly used.
In developed countries, these are self-poisoning by prescribed or over-the counter
medicines or by motor vehicle exhausts, hanging, firearms, drowning, jumping from
high places or in front of moving vehicles or self cutting. On the other hand, poisoning
with agricultural chemicals is the most common method in developing countries.
The popularity of different methods changes within and between countries with time,
determined by relative availability and social and cultural acceptability of each method.
Research shows that availability, familiarity with the method, and technical skills
necessary to use the method effectively are all relevant variables in the choice and
frequency of method used. Thus farmers tend to shoot themselves and medical
practitioners, nurses and pharmacists take medicines. The fatality is dependent on the
length of time that elapses between a potentially lethal method and death and on whether
there are people and facilities available to resuscitate the person. Thus an overdose of
drugs usually offers some opportunity for reconsideration and rescue whereas hanging or
firearms do not. An overdose of pesticide, while extremely dangerous everywhere, in
rural China or Sri Lanka is much more likely to be fatal than such an overdose in the
Western world where there is faster access to sophisticated Accident and Emergency
units.
While some suicides have been planned in detail for weeks or even months, many
suicides are impulsive, occurring soon after a ’’last straw” life event in the context of
multiple social stresses and lack of social supports which have often induced underlying
depressive symptomatology. This is important because it means that if access to a
commonly used method of suicide is restricted and not available at the time of the
impulse, then the suicide is prevented for long enough for the suicidal impulse to wane.
Therefore reducing access to one means of suicide does not result in complete
substitution by another method, and this is exemplified in the detoxification of domestic
gas which has been associated with a decline in suicide rates (REF England, Australia,
Netherlands)and in the marked decline in fatal self poisoning due to barbiturates in
countries where there has been better prescribing practices for depressed patients,
legislative restriction on barbiturate prescribing and the substitution of benzodiazepines.
Motor vehicle exhaust suicides could be reduced by the introduction of carbon
monoxide sensors (to cut off the engine at a toxic level), exhaust modification (to make
attaching a hose more difficult) and modifications to catalytic converters which decrease
53
Chapter 8
Collective Violence
Status:
Draft: 3rd
Peer Reviewed: Yes
Date of Current Draft: 4 October 2000
Table of Contents
I.
II.
III.
IV.
V.
VI.
VII.
Introduction
Definitions
Data Sources, Trends and Emerging Patterns
Trends in Violent Political Conflict over the Centuries
A.
Global Burden of Conflict
B.
Problems of Collecting Data
C.
Direct and Indirect Health Impact
Population Displacement
A.
Increased Mortality and Morbidity
B.
Communicable Diseases
C.
HIV/AIDS
D.
Violence against Women and Rape
E.
Increased Vulnerability
F.
Disability
G.
Mental Health
H.
Impact on Health Services
I.
Human Resources
J.
Impact on Infrastructure and Development Resources
K.
Community Participation
L.
Macroeconomic Impact
M.
Impact on Food and Agriculture
N.
Forced Resettlement
O.
Contributory Causes to Violent Political Conflict
Level of Weapons Technology
A.
Indicators of States at Risk
B.
Responding to Collective Violence
Prevention
A.
1. Promotion of Human Rights and Application of International Treaties
2. Early Warning Systems
3. Building and Boosting Resilience
4. Community Involvement
5. Increasing Control by Those Affected
Providing Services During and in the Aftermath of Crises
B.
Post-Conflict Re-establishment of Services and Systems
C.
Other Public Health Roles: Documentation, Advocacy, Health as a Bridge to
D.
Peace
Recommendations
Definitions
A.
Data and Surveillance
B.
Prevention
C.
Boosting Resilience
D.
Improving the Quality of Humanitarian and Other Responses
E.
Enhanced Post-Conflict Responses
F.
Research Needs
G.
2
Introduction
In today's globalized world, war and its effects in remote areas grab our attention. We know
something about East Timor, Chechnya, Sierra Leone - places which otherwise would be off
the maps of many. Despite this, however, many of the victims, events, responses and
complexities remain hidden, sometimes deliberately.
The objectives of this chapter are to:
•
Describe the pattern, magnitude, forms and impact of collective violence on health
and health systems;
•
Summarise existing information on risk factors, scope for prevention and the
development of policy responses
•
Identify important research priorities for the future
This chapter suggests that if we are to effectively respond to collective violence and war we
will need to:
•
Develop an understanding of the context in which collective violence occurs, and identify
the known and presumed risk factors
•
Establish the extent to which prevention of collective violence may be possible;
•
Recognize the sensitivities and scope for manipulation concerning data on collective
violence; and promote improved surveillance and recording of conflict-related
experiences
•
Recognize the importance of understanding and bolstering resilience and protective
factors which assist individuals, societies and systems to maintain themselves during
periods of conflict;
•
Highlight the mechanisms being adopted by humanitarian agencies to improve their
performance and accountability by developing and adopting good practice guidelines;
•
Highlight the value of improving our responses after conflicts so as to ensure that new
cycles of violence are not re-established;
•
Draw attention to the wide range of possible interventions in response to political conflict:
including primary prevention to avoid violent political conflict occurring, secondary’
prevention to reduce the impact of wars and collective violence, and tertiary prevention to
minimise the longer term damage to health systems and to the determinants of health.
3
Definitions
Collective violence may be defined as the use of force by groups to achieve political,
economic, or social objectives. Various forms of political violence have been explicitly
recognized: these include wars and related violent political conflicts which occur within or
between states, state violence which takes the form of genocide, repression, disappearances
torture and other human rights abuses, and structural violence which is characterized by
economic, political, and social discrimination and inequity directed at one or more societal
groups*.
A clear example of structural violence was the policy of apartheid in South Africa. The
social, economic and political rights of the black majority population were systematically
undermined and abused by the apartheid state; this was accompanied by state-repression in
the form of torture, detention without trial, and political assassinations. The discriminatory
impact on health status and on access to health care has been well described1; reversing the
long-term impact of these forms of discrimination still represents a major challenge to the
current democratic government which was elected in 1994.
All deaths which occur in the organized contest for political or economic control of a territory
may be considered war-related deaths. The Correlates of War project (COW) and Stockholm
International Peace Research Institute (SIPRI) have defined minor armed conflicts as those in
which 25 - 1000 battle deaths occur over the entire period of the conflict; intermediate armed
conflict as those in which 25 -1000 deaths occur per year in a conflict; and major conflicts as
those in which more than 1000 deaths per year occur. These terms have limitations and do
not capture the extensive contextual and other difference that will be present in countries of
different sizes and resource bases.
Even if definitions can be agreed, the number of conflict deaths attributed to a particular
conflict vary considerably among researchers and even more so between parties to the
conflict. Up to 101 major conflicts have been recorded since the end of World War II, with
20-30 ongoing at the beginning of the 21st century. Minor differences in how specific
conflicts are categorized has relatively little effect on the total war-related deaths recorded
each year, as a small number of major conflicts accounts for the majority of all deaths.
Complex humanitarian emergency (CHE) is a term which describes “A situation affecting
large civilian populations which usually involves a combination of factors including war or
civil strife, food shortages, and population displacement, resulting in significant excess
mortality” 2. Other analysts3 (1999) prefer the term 'complex political emergencies' (CPEs)
because this highlights the political roots and contributions to the crisis. CPEs typically occur
across state boundaries; have political antecedents relating to competition for power and
resources; are protracted in duration; are embedded in and reflect existing social, political,
economic and cultural structures and cleavages; and are often characterised by 'predatory'
social domination. Leaning identifies four characteristic outcomes of CHEs - population
" Other forms of organised violence, such as gang warfare and criminal violence associated with
banditry, which do not have political objectives, are not discussed here.
4
dislocation; destruction of social networks and ecosystems; insecurity affecting civilians and
noncombatants and human rights abuses4.
Kaldor uses the term 'new wars' to describe those which blur the distinctions between war
(usually defined as violence between states or organized political groups for political
purposes), organized crime (violence undertaken by privately organized groups for private
purposes, usually financial gain) and large-scale violations of human rights (violence
undertaken by states or politically organized groups against individuals)5. The 'new wars' are
said to have a number of common features6 : they entail an element of 'identity politics' - the
claim to power on the basis of a particular national, clan, religious or linguistic identity; they
employ techniques characterised by an attempt to control the population by getting rid of
those of a different identity, through the use of forced resettlement, mass killings, and
intimidation, and they link what appear to be local, decentralised conflicts with a 'globalized
war economy', interacting with people, processes and aspirations present at the global level.
Warring factions may prey upon the civilian populations in their midst, seizing their food and
assets. Some commentators dispute this categorisation, arguing that many features of these
new wars are not 'new' at all and have been present in many conflicts over the last century.
Genocides (intent to destroy, in whole or in part, a national, ethnic, racial or religious group7),
politicides (attempts to eliminate a different political group), and democides (attempts to
eliminate a different social group) are all forms of ’war on the civilian population'. Although
in practice these events have often occurred just prior to, during, or following major clashes
between military forces, they sometimes occur independent of military conflicts. Genocides
and democides, directed at civilian rather than military populations, have been responsible for
the largest number of conflict-related deaths in the 20th century. Major genocides in the 20th
century have included the systematic extermination of Armenians by the Young Turks in
1915, the Turkish massacre of Kurds in 1937-38, the holocaust in which Jews, gypsies,
homosexuals and communists were systematically annihilated by the Nazis in the second
world war, the massacre of Hutus by Tutsi perpetrators in Burundi in 1972, the Khmer Rouge
'auto-genocide' in which around 1.5 million Cambodians were killed in the mid-1970s, the
1994 genocide against Rwandan Tutsis.
5
Data Sources, Trends and Emerging Patterns
Several research groups collect and analyze data on the victims of conflict, including the
Department of Peace and Conflict Research (DPCR) at Uppsala University, Sweden, which
focuses on conflict resolution; the Stockholm International Peace Research Institute (SIPRI)
which has developed a detailed, standardized reporting format for producing an annual report
on conflict impacts; the Unit for the Study of Wars, Armaments, and Development (AKUF)
at the University of Hamburg; the Interdisciplinary Research Program on Root Causes of
Human Rights Violations (PIOOM) in the Netherlands which monitors deaths and other
violations of rights around the world; and the Correlates of War (COW) project, the most
widely cited source on the magnitude and causes of conflicts from the 19th century to the
present.
Trends in Violent Political Conflict over the Centuries
One set of estimate suggests that 1.6, 6.1, 7.0, 19.4 and 109.7 million military-related deaths
occurred in the 16th, 17th, 18th, 19th and 20th centuries respectively8 9. Such aggregated data are
of limited use in that they hide the circumstances in which populations have died. At least 6
million people are estimated to have died in the capture and transport of slaves over 4
centuries, and at least 10 million native people of the Americas died at the hand of European
settlers.
According to White, about 180 million people lost their lives from direct or indirect warrelated causes in the twenty five events of greatest magnitude in the 20th beginning with
colonial war in the Congo and ending with repeated episodes of genocides in Rwanda and
Burundi. Rummel estimates 191 million deaths, also citing 60% of them as occurring among
noncombatants. Two specific events - Stalin’s terror and Chinese deaths in the Great Leap
Forward (1959 - 1962) are the source of the greatest imprecision in terms of magnitude of
human destruction. Immediate causes of death in the 25 events of greatest magnitude
included about 39 million military and 33 million civilians who died in wars. About 12
million other people were killed in genocides, and 40 million in democides or politicides.
Famine related to war, genocide, or democide also killed an estimated 40 million people.
Deaths among the military comprise 22% of all these conflict related deaths.
No consensus has been reached on which of the many estimates of collective violence deaths
is most accurate. Few data are produced with the degree of precision and rigour expected for
public health surveillance, and estimates are often very politicized with some investigators
even denying that certain genocides occurred. More precise measurement would be to help
differentiate deaths among military personnel from those occurring in the general population.
Deaths among armed forces are usually recorded in military vital event systems and therefore
are more accurate, usually varying by no more than 10% - 50% . This contrasts with conflicts
fought by non-state groups, among whom death data are more easily manipulated and less
readily confirmed, in which estimates vary by up to 100%. Genocides, politicides and
democides are even more subject to manipulation and are harder to confirm: estimates for
mass killings of noncombatant groups may vary as much as 5 - 10 fold.
A small number of events cause most of the deaths in each category. The two world wars
were the source of about two thirds of all deaths among soldiers and civilians killed in wars.
6
Democide and politicide in many countries during WWII and in China in the 1950s and
1960s account for more than 75% of these deaths. About half of all genocide deaths in the
twentieth century occurred during WWII. Most famine deaths occurred in the Soviet Union
in the 1920s or in China during 1959 - 1962.
A total of 165 wars or tyrannies in the 20th century each killed more than 6,000 people. Five
were responsible for more than 6 million deaths - World Wars I and II, the Russian Civil War,
Stalin’s rule, and Mao Tse Tung’s rule. All of these events were characterized by a majority
of deaths occurring among civilians, alongside experiences of democide, genocide, or famine.
Together, these five events accounted for about 85% of all conflict-related deaths worldwide
in the 20th century.
Twenty one other wars or tyrannies caused between 600,000 and 6 million deaths, 61 caused
60,000 - 600,000 deaths, and 78 caused 6,000 - 60,000 deaths. Most of the events that caused
the greatest loss of life occurred early in the century. Only one of the seven biggest killing
events occurred after 1950 (famine in China), while sixteen of the next 23 occurred after that
year.
In the entire period since WWII, the AKUF estimates there to have been a total of 190 wars.
Only a quarter of these were international wars. Most of these wars lasted less than 6 months.
Wars lasting more than six months often extended for many years. The total number of wars
was under 20 in the 1950s, above 30 in the 1960s and 1970s, and reached a peak of more
than 50 during the late 1980s before beginning to recede towards the end of the 20th century.
The number of wars declined after 1992, but the remaining wars are, on average, of longer
duration. The number of conflicts on each continent declined with the exception of Africa.
About 20% of these conflicts have accounted for more than 80% of all deaths. An estimated
total of 51 million people are believed to have lost their lives from all direct and indirect
causes related to conflicts since WW II; around 17 million deaths can be considered to be
direct effects of these conflicts.
Although almost all conflicts occurred in developing countries since the end of the Second
War, the post-Cold War period saw an increase in the occurrence of Europe-based conflicts
(Table 1). These have occurred mostly in areas of the former Soviet Union and Eastern bloc
and reflect power struggles within and between national and ethnic groupings.
Table 1: Armed Conflicts by Region: 1989-1998 10
Europe 2
M.East
Asia
Africa
Americas
89
3
4
19
14
8
90
6
6
18
17
5
91
9
7
16
17
5
92
10
7
20
15
4
93
5
7
15
11
3
7
94
5
5
15
13
4
95
1
4
13
9
4
96
0
5
14
14
2
97
2
3
14
14
2
98
3
15
14
2
Table 2 highlights the fact that modern-day conflicts are increasingly within rather than
between states. Although interstate conflicts occur they are rare.
Table 2: Ongoing Conflicts in Terms of Inter or Intrastate Involvement (Numbers)11
1989
’90
’91
’92
’93
’94
’95
’96
’97
’98
43
44
49
52
42
42
34
33
30
32
Combined (intra- and inter-state)
1
2
1
2
4
0
0
1
3
2
0
0
1
2
1
2
Intrastate
Interstate
3
3
1
1
Global Burden of Conflict
The WHO estimates that about 588,000 people died in wars in 1998. That makes war the
fourth most common type of injury death in that year, after unintended injuries, suicides, and
homicides. Deaths from war varied from less than 1 per 100,000 population in high income
countries, to 12 per 100,000 in low/middle income countries. The rate varied by region from
near 0 in China, India, and the Americas to 33/100,000 in the eastern Mediterranean and
51/100,000 in Africa. War ranked as the 19th most common cause of death in high income
countries and the 17th most common cause of death in low/middle income countries. War
ranked as the 13th most common cause of death for 0-1 year olds, 5th for 5-14 year olds, and
5th for 15-44 year olds in 1998.
An estimated 5 percent of all deaths during the 20th century were due to the immediate or
secondary impact of collective violence. This was higher than in the 17th - 19th centuries, in
which 2 percent of deaths are estimated to have resulted from collective violence. The 40fold rise in the number of deaths among soldiers in the 20th century greatly exceeded a
doubling of the globe’s mid-century population. Military deaths per million population rose
18-fold from the 19th to the 20th century (Table 3). Genocide and democide-related deaths
also rose in the 20th century as the centralization of large political and economic systems and
the emergence of new technologies made mass killings possible12 .
Table 3: Estimated Average Annual Military Deaths in Wars, Worldwide, by Century
Century
Average Annual
Military Deaths
17th
18th
19th
20th
9,500
15,000
13,000
458,000
World Mid-Century Average Annual
Population in
Military Deaths per
Millions
Million Population
500
19.0
800
18.8
1,200
10.8
2,500
183.2
8
Further research will be needed to determine the different forms of collective violence, as
well as their nature, implications and trends for affected population groups. Standardized
indices will assist in analysis of the massive figures on war-related deaths: specification of
the population from which the deaths occur will generate indicators of the proportion of the
population killed, comparison with the normal death rate among that group prior to initiation
of violence will permit identification of the magnitude of excess mortality rates, and
specification of the time period over which deaths occurred permits determination of the
velocity of death.
Problems of Collecting Data
Most poor countries lack reliable health information and vital registration systems; in the
absence of such baseline data it is particularly difficult, if not impossible, to determine the
proportion of morbidity, mortality, and disability which is clearly conflict-related. Complex
emergencies invariably disrupt surveillance and information systems13 thus further weakening
the potential to assess conflict-related impact. Assessing the impact of collective violence
thus often requires analysis of data which does not normally enter into health data systems.
Innovative methods have, in some situations, been used to more accurately define the extent
of loss of life in conflicts. In Guatemala data were analyzed to define the numbers of violent
deaths during the civil conflict which engulfed the country from 1960 - 1996. Three in
country data sets were developed from witness and victim reports. Total deaths estimated
from these sources range from 8,500 to 24,505. If each source were mutually exclusive total
reported deaths would be 54,065, but duplicate entries reduced the total to about 48,000
reported deaths. The existence of these multiple, overlapping sources permitted more
accurate estimation of the total deaths unaccounted for of 84,000 people. Summing
documented and undocumented estimated total deaths due to the civil conflict produced a
total estimate of 132,000 +/- 13,000 (95% confidence interval). The largest number of
registered deaths in a year occurred in 1982, at 18,000. Yet official government reports
register much fewer deaths due to violence in that year. These more than 100,000 deaths
remain invisible in systems that evaluate the burden of disease on the basis of vital event
•
• 14 15
registries
The effect of conflict on particular sub-groups of the population such as war orphans,
unaccompanied child refugees, and internally displaced populations (IDPs) may be especially
difficult to determine. Population size and density may vary tremendously over short periods
of time as people move to where safety and resources are greatest; measuring health impact
and health status with such shifting populations and uncertain denominators is always
problematic. Data on numbers of IDPs and refugees may on occasions be manipulated to
make a political point or maximize resource access.
Determining precisely how an emergency unfolded and precisely what occurred in a given
complex emergency, as well as establishing an documenting the health impact are significant
challenges. As mentioned earlier, parties to the conflict will seek to have their own
viewpoints reflected and to play down the impact on civilians. Where popular opposition to
conflicts has been generated, it has often been as a result of perceived or actual excessive
impact on civilians, as the USA found in the Vietnam war, NATO countries found during its
war with Yugoslavia, and Israel found during the Palestinian intifada. Both measurement and
9
information biases may be present, and civil society organizations, both indigenous, and
global, play a valuable role in documenting and ‘witnessing’ occurrences of collective
violence and human rights abuses, and the response by affected communities.
Estimates of the magnitude of impact are extremely imprecise: deaths in the Rwandan
genocide have been assessed as varying from 500,000 to one million. In East Timor, tens of
thousands of people were considered missing after the conflict and months later it was still
unclear whether the original population estimates had been erroneous or whether tens of
thousands of people were missing. Little was known about the health burden of the conflict
in the Democratic Republic of Congo although recent estimates have suggested that over 1
million people may well have lost their lives in the Congo forests.
Direct and Indirect Health Impact
This section seeks to highlight the variety of ways in which conflict affects health and health
systems. It does not seek to be comprehensive, but to reflect the range of impacts, thus
highlighting how other health-related objectives suffer in the presence of violent political
conflict.
Mass destruction of the enemy as a tactic of warfare has been common throughout history,
although we are currently more aware of this given today's technologies of mass
communication. Laying of siege, laying waste to essential goods and services, poisoning
water supplies, or enslavement of a losing enemy often accompanied warfare in pre-modem
times. In European war since the establishment of nation-states in the 17th century, soldiers of
one nation engaged in direct battle almost exclusively with soldiers of a rival nation. Anti
colonial wars, often based on guerilla warfare, further blurred the distinction between the
military and civilians. This distinction was further eroded with the breakdown of national
states since the end of the Cold War. In many internal conflicts, often pitting the state against
a section of the civilian population, torture, disappearances and other forms of repression
have been practiced in pursuit of political and ideological objectives.
The end of the Cold War unleashed a great deal of violence. Given the widespread
availability of small arms16, parties to conflicts did not have to be affluent, well organized, or
supported by one of the global superpowers, in order to obtain modem weapons or perpetrate
mass destruction. Even uniformed fighters in many conflicts do not hold clear political
allegiances nor have a clear definition of who represents 'the enemy'. One side-effect of this
change has been the large numbers of violent deaths of civilian UN employees and workers
with non-governmental agencies (see Box). At least 382 deaths among humanitarian workers
occurred in the years 1985-199817. More UN civilian personnel than UN peacekeeping troops
have been killed over the same period.
While conflicts within states are most common, conflicts between states still occur. The IraqIran war of 1980-1988 is estimated to have left 450,000 soldiers and 50,000 civilians dead18.
The Eritrea-Ethiopian conflict in the last year of the 20th Century was largely fought between
two conventional armies, using heavy weaponry and trench warfare, and claimed tens of
thousands of lives. In recent years we have also witnessed coalitions of multinational forces
waging war using massive air attacks as in the Gulf against Iraq in 1991 following the latter's
10
invasion of Kuwait, and in the NATO-led campaign against Yugoslavia (1999) in an effort to
stop the internal violence and repression against the Kosovar Albanian population.
Under the Geneva conventions, the rules of war require the application of principles of
proportionality and distinction in the choice of targets. Proportionality involves an assessment
of ways to minimize likely civilian casualties when a military objective involves targeting
which is not exclusively military. Distinction focuses on avoiding civilian targets wherever
possible.
Whichever type of war we focus upon, civilians comprise a substantial proportion, and often
the majority, of casualities in most conflicts still in progress at the beginning of the 21st
century. Measuring the impact and costs of conflict is complex due to methodological and
theoretical constraints, inconsistencies in definitions and terms, and restricted access to
affected areas. No standard definitions of direct and indirect effects exist, and data are
subject to extensive political manipulation. The manipulation of data is exacerbated the
closer in time one is to the events being reported. Both NATO and the Yugoslav government
devoted attention not only to the air battle but to the ‘battle over the air-waves’: the Federal
Republic of Yugoslavia and NATO repeatedly tussled over the number and circumstances of
civilian casualties. At the height of the conflict, US government estimates of deaths among
Albanian Kosovars in 1999 varied from 100,000 - 500,000. After the war the estimated
number of deaths was substantially lower than 10,000, including about 2,000 deaths resulting
from allied bombings19. A UN spokesman in Angola argued that only general estimates of
war-related deaths was possible: 'You simply cannot count bodies in any war. There are other
priorities', he said20.
War typically entails a ‘conscious attempt by armed parties to subdue or inflict harm on the
individual members of an opposed group, to dominate or shatter the social structures of their
enemy, and/or to capture, damage or destroy their enemy's material resources' .21 While
wars have probably always reflected such societal destruction and targeting, international
community efforts to prevent annihilation of populations and massive economic destruction
have been most visibly inadequate in recent conflicts.
Population Displacement
One consequence of the targeting of entire communities and their livelihoods have been the
large numbers of displaced people (Table 4). Refugee (those seeking refuge across
international borders) numbers have risen from 2.5 million in 1970, 11 million in 1983, to
18.2 million in 1992 and 23 million In 1997 22 23. Following the end of the Cold War there
were estimated to be 30 million internally displaced people24, the vast majority fleeing
conflict zones. Those displaced within countries have less access to resources and support
from the international community and may be at ongoing risk from violence perpetrated by
the state and other powerful local actors25.
11
7A
Table 4: Millions of Internally Displaced and Refugees by Continent and Year (Adapted from )
21
92
93
94
25
96
97
98
Internally
displaced
Africa
13.5
EAsia/Pa 0.34
S.Asia
3.09
Mid-East 1.3
Europe 1.05
Americas 1.13
14.2
0.68
2.69
1.45
1.76
1.22
17.4
0.7
1.81
0.8
1.63
1.35
16.9
0.6
0.88
1.96
2.77
1.4
15.73
0.61
1.78
1.71
5.2
1.4
10.19
0.56
1.6
1.7
5.08
1.28
8.51
1.07
2.4
1.48
4.74
1.22
7.59
0.8
2.25
1.48
3.7
1.62
8.77
0.53
2.12
1.58
3.27
1.77
Refugees
Africa
5.41
EAsia/Pa 0.7
S.Asia
6.33
Mid-East 3.55
Europe 0
Americas 0.15
5.32
0.81
6.9
2.79
0.12
0.12
5.73
0.5
4.72
2.85
2.53
0.1
5.81
0.8
3.9
2.98
1.95
0.1
5.9
0.69
3.32
3.83
1.78
0.12
5.19
0.64
2.81
3.96
1.81
0.07
3.62
0.65
3.18
4.37
1.88
0.07
2.9
0.72
2.97
4.3
1.34
0.06
2.73
0.67
2.93
4.38
1.32
0.36
Ratio of IDPs to refugees
Africa
2.5
2.7
E.As/Pa 0.5
0.8
S.Asia
0.5
0.4
Mid-East 0.4
0.5
Europe
14.7
Americas 7.5
10.1
3.0
1.4
0.4
0.3
0.6
13.5
2.9
0.8
0.2
0.7
1.4
14
2.7
0.9
0.5
0.4
2.9
11.7
2
0.9
0.6
0.4
2.8
18.3
2.4
1.6
0.8
0.3
2.5
17.4
2.6
1.1
0.8
0.3
2.8
3.2
0.8
0.7
0.4
2.5
4.9
90
T1
Refugees are typically enumerated in order to plan and provide relief, but relatively little
attention has been devoted to establishing the precise composition of refugee and internally
displaced populations (IDPs). Older adults, refugees not in camps, and IDPs, may neither be
identified nor receive required attention.
Table 4 illustrates patterns of refugee and internally displaced movements over the last
decade. It is notable that the ratio of IDPs to refugees varies dramatically between regions: in
Africa, Europe and the Americas there are many more times as many IDPs as refugees;
whereas in Asia and the Middle East the opposite holds.
12
Increased Mortality and Morbidity
Many deaths may result from reduced access to health services or public health programs, but
there is invariably be dispute regarding the extent to which such outcomes should be
considered 'conflict-related'. Where the impact on health is due to the weakened economy
and environmental destruction, it may be even more difficult to agree on whether and how to
attribute these ill-effects to the conflict. Some have suggested measuring the opportunity
costs of development foregone as a result of the conflict. Countries in conflict have made
systematically less progress, when compared with others of similar socio-economic status and
in the same region, in extending life expectancy, and reducing infant mortality and crude
death rates27. Such analyses, however, may be confounded by the simultaneous influence of
the HIV/AIDS pandemic, which in some settings is also exacerbated by conflict and
instability28 29.
Refugees and internally displaced persons typically experience high mortality, especially in
the period immediately after their migration30 31. Deaths from malnutrition, diarrhea and
infectious diseases occur especially in children, while some communicable diseases such as
malaria, tuberculosis and HIV infection, as well as a range of non-communicable diseases,
injuries and violence typically affect adults. The prior health status of the population, their
prior access to key determinants of health (housing, food, shelter, water and sanitation, health
services), the extent to which they are exposed to new diseases and the level of resource
availability, all affect their health status.
Not surprisingly, the impact on health can be extensive in terms of morbidity, mortality and
disability (Table 5). Over thirty years of war in Ethiopia led to approximately one million
deaths, about half of whom were civilians32. About one third of the 300,000 soldiers
returning from the front after the end of the conflict had been injured or disabled and at least
40,000 people had lost one or more limbs in the conflict.
Table 5: Examples of Direct Health Effects of Collective Violence (Adapted from 33)
Category of
impact
Mortality
Morbidity
Disability
Causes
External causes: mostly weapon-related mortality________________________
Infectious diseases: Preventable diseases measles, polio, tetanus, malaria_____
Non-communicable diseases: deaths otherwise avoidable through medical care
(e.g. asthma, diabetes, emergency surgery)_____________________________
External causes: injuries from weapons, mutilation, anti-personnel landmines,
bums, poisoning__________________________________________________
Other external causes: sexual violence_________________________________
Infectious diseases
• water-related: cholera, typhoid, shigella
• vector-bome: malaria, onchocerciasis
• other communicable: tuberculosis, ARI, HIV/AIDS, other sexually
transmitted infections________________________
Reproductive health:
• Increased prematurity, low birth weight, stillbirths, delivery complications
• longer-term genetic effects of exposure to chemicals and radiation_______
Nutritional: acute and chronic malnutrition plus variety of deficiency disorders
Mental health: anxiety, depression, post-traumatic stress disorder___________
• Physical
• Psychological
• Social
13
Health status may worsen substantially in wartime. Infant mortality rises in association with
reduced access to health and immunisation services, impairment of the basic infrastructure
necessary to promote health, poorer nutrition for children and their mothers, and population
displacement. Preventable diseases such as measles, tetanus and diphtheria may become
epidemic. In the mid-1980s, infant mortality in Uganda rose above 600 per 1000 in some
war-affected areas34. UNICEF showed that declines in infant mortality were reported for all
countries in Southern Africa over the period 1960-1986, with the exception of Mozambique
and Angola both of which were affected by vicious civil wars and aggression by apartheid
South Africa in this period35.
The comprehensive economic embargo against Iraq in the 1990s, although not a war as such,
shared many similar characteristics, notably the very high civilian-to-military casualty rate
and deterioration in many indices of health status36.
In Zepa (former Yugoslavia), a UN-controlled ’safe-haven’ which was subsequently overrun
by the Bosnian Serbs, perinatal and childhood mortality rates doubled after only one year of
war. In Sarajevo, deliveries of premature babies had doubled and average birthweights fallen
by 20% by 1993, two years into the war. In Bosnia, fewer than 35% of children were
immunized, compared with 95% before the war37 38; in Iraq dramatic declines in
immunization coverage have been experienced since the Gulf War and the subsequent
imposition of economic and political sanctions. Recent evidence from El Salvador indicates
that it is possible, with selective health care interventions and major resource inflows, to
improve certain health indices during ongoing conflicts39; although other health problems are
more refractory to such inputs.
Communicable Diseases
The occurrence and transmission of communicable diseases increases due to the decline in
immunisation coverage, population movement, reduction in public health campaigns and
outreach activities (see Table 8) and the lack of access to health services. A war-related
measles epidemic in Nicaragua was attributed in large part to the declining ability of the
health service to immunise those at risk in conflict-affected areas40 while deterioration in
malaria control activities was associated with epidemics in Ethiopia41 and Mozambique42
highlighting the vulnerability of complex disease control programmes in periods of conflict.
Increased rates of malaria in Nicaragua were attributed to war-related population and troop
movements, inability to carry out timely disease control activities, and shortages of the health
personnel needed to conduct control programmes in peripheral areas43. In Ethiopia,
epidemics of louse-borne typhus and relapsing fever were associated with crowded army
camps, prisons, and relief camps, as well as the sale of infected blankets and clothes to local
commumties by retreating soldiers44.
In Rwanda, epidemics of water-related disease
(shigella dysentery and cholera) led to the death, within a month, of 6-10% of the refugee
population arriving in Zaire45. The crude death rate of 20-35 per 10000 population per day
was two to three times higher than that previously reported in refugee populations.
HIV/AIDS
'High risk situations' for HIV transmission may also occur in times of conflict and their
aftermath46. HIV infection has reached high levels in many army forces; the ability of these
14
men to command sexual services from local women, through payment or force, the movement
of troops to different parts of the country, and their ultimate return to divergent regions of the
country after demobilisation, present significant risks to women47 48 49. Military forces
(including peace-keeping forces) may stimulate a market for sexual services, attracting
women from surrounding states into commercial sex work50, thus fuelling HIV and STD
transmission.
Violence against Women and Rape
Violence against women and rape as a weapon of war have been documented in many
conflicts. Although the scale of rape in war has often been hidden, recent conflicts and the
systematic use of rape within them, have attracted media, academic and service attention.
Estimates of the number of women raped in Bosnia range from 10,000 to 60,00051 and reports
of rape in wartime have been documented from Bangladesh, Liberia, and Uganda, amongst
others. Rape is used to terrorize and undermine communities, to force people to flee, and to
fragment community structures. The impact on the violated women may be far-reaching in
both physical and psychological dimensions. Recent work has sought to document and reflect
upon these experiences in order to enhance our responses to such violence52 53. The
International Criminal Court has indicated that systematic rape in wartime should be
considered a crime against humanity.
Increased Vulnerability
Raised mortality rates reflect the combined effects of poor nutrition, increased vulnerability to
communicable diseases, diminished access to health services, poor environmental conditions,
and psychosocial distress. A study of the impact of the war in Bosnia drew attention to the
creation of new ’vulnerable' populations such as those in isolated enclaves or forced to flee as
a result of ethnic repression54. Reviews of the health of refugees and displaced populations
have revealed massively raised mortality, at its worst, up to 60 times the expected death rates
during the acute phase of displacement 55 56 57. In Monrovia, Liberia, the death rate among
civilians displaced in 1990 during the civil war, was seven times higher than the pre-war
death rate (MSF, Holland, quoted in 58).
Disability
Data on war-related disability are scant. A nation-wide disability survey conducted in 1982
after the liberation struggle in Zimbabwe, revealed that 13% of all physical disabilities, were
the direct result of the war. Estimates of landmine-related amputations and disabilities are
sobering: 36000 in Cambodia (6000 in 1990 alone; i.e. one in every 236 Cambodians has lost
at least one limb after detonating a mine). Over 30 million mines were laid in Afghanistan in
the 1980s, the costs of which are both medical and social59. In Hargeisa Hospital,
Somaliland, 74.6% of the land-mine related injuries treated from February 1991 to February
1992, were in children between the ages of five and fifteen years60.
15
Mental Health
The mental health impact is influenced by a range of factors including the nature of the
conflict, the form of trauma experienced (or directly inflicted, as in the case of torture and
other repressive violence), the individual and community response to these pressures, the
cultural context in which they occur, and the psychological health of those affected prior to
the event61 62. Psychological stresses are also associated with displacement, both forced and
voluntary, and result from loss and grief, social isolation, loss of status, loss of community,
and in some settings, acculturation to new environments63. Manifestations of such stress
include depression and anxiety, psychosomatic ailments, intra-familial conflict, alcohol abuse
and antisocial behaviour. Single and isolated refugees, as well as women who are single
handedly managing a family, may be at particular risk.
Summerfield64 and Bracken et al65 caution against assuming that populations do not have the
ability and resilience to respond to these adverse circumstances; authors in South Africa have
similarly argued that not everybody exposed to massive degrees of trauma become 'victims’ their ability to respond is strengthened by their perception of being part of a legitimate
struggle66. The medical model which labels individuals with a particular complex of
symptoms and signs as having 'post traumatic stress syndrome' is culture-bound and may fail
to take account of ongoing stressors: it cannot address the complexity of human response,
including the interpretation and adaptation to the effects of violence67. It is important to
appreciate the subjective meaning of the violence or trauma, the way in which distress is
experienced and reported, the type of support available to the individual, and the therapies
available. It is increasingly apparent that recovery is linked to the reconstruction of social and
economic networks, cultural institutions and respect for human rights68 . A useful
contribution to the debate suggests that conflict-affected populations be divided into three
groups: those with disabling psychiatric illnesses, those with severe psychological reactions to
trauma, and the majority who are able to adapt once peace and order are restored69. The two
former groups will benefit greatly from context-appropriate service provision.
Depression, substance abuse and suicide are also important consequences of collective
violence70. Sri Lanka before the two decades of civil war had a much lower suicide rate
overall (higher for Tamils, very low for Sinhalese) than is presently the case. Similar data
have been reported from El Salvador; in both cases at least in part as a consequence of
political violence.
Impact on Health Services
The impact of conflict on health services are wide-ranging (see Table 6). Damage to the Iraqi
health system in the 1991 international community response to the Iraqi invasion of Kuwait
and its repression of its Kurdish minority was dramatic. Health services were accessible to
90% of the population and the country was able to immunise the vast majority of children
under the age of five years before the war. By the end of the conflict, many hospitals and
clinics had been severely damaged or closed: those still operating were forced to cope with
much larger catchment populations, and damage to water supplies, electricity and sewage
disposal, exacerbated both the determinants of health and the operation of health services71.
16
Table 6: Impact of Conflict on Health Services (Adapted from 72)
Category of
____ impact____
Reduced access to
services
Compromised
service
infrastructure
Human resources
Equipment and
supplies
Health
activity
services
Impact on health
policy formulation
Relief activities
Manifestation
Reduced security (landmines, curfews)
Reduced geographic access (poor transport)
Reduced economic access (increased charges for health services)
Reduced social access (fear of service providers or of being identified as
conflict participant)__________________________________________
Destruction of clinics
Disrupted referral systems
Destroyed vehicles and equipment
Poor logistics and communication_______________________________
Injury, killing and kidnapping of health workers
Displacement and exile
Poor morale
Difficulty keeping health workers in public sector and especially in
insecure areas
Disrupted training and supervision_____________________________ _
Lack of drugs
Lack of maintenance
Poor access to new technologies
Inability to maintain cold chain for vaccines_____________________
Shift from primary to tertiary care
Increased urbanization of provision
Reduction in peripheral and community-based activities
Contraction of outreach, preventive and health promotion activities
Disrupted surveillance and health information systems
Compromised vector control and public health programs (partner
notification, case-finding)
Tendency towards vertical programs
Reliance on range of organisations to provide project-based services
Undermined national capacity
Inability to control and coordinate NGO and donor activities
Reduced information upon which to make decisions
Reduced engagement in policy debates locally and internationally
Impaired community structures and reduced participation
__
Limited access to many areas
Increased expense in delivering services
Increased pressure on host communities, systems and services
High degree of verticality
Insecurity of relief personnel
Impaired coordination and communication between agencies
Poorer supply of drugs during and following conflicts have been associated with increases in
medically preventable causes of death such as asthma, diabetes and many infectious diseases.
The quality of available care may deteriorate greatly, whether in hospital or out-patient
settings. The lack of personnel, diagnostic equipment, electricity, water, and drugs all
contribute to these problems.
17
Human Resources
Human resources are seriously affected by conflict and in countries such as Nicaragua and
Mozambique have been specifically targeted. Violations of medical neutrality have been
reported from many conflicts, including those in South Africa, the Occupied Territories of
Palestine, Philippines, and El Salvador73. Qualified personnel often retreat to safer urban
areas or may leave their profession. In Uganda, half the doctors and 80% of the pharmacists
left the country in search of safer opportunities between 1972 and 1985. In Mozambique, the
country was left with only 15% of the 550 doctors present in the country before
independence74. In East Timor, only 20 Timorese medical practitioners were present after
Indonesia left the territory. In Kosovo, although there were numerous Albanian medical
professionals, most had not been able to work in the official health system while the area was
under the control of the Federal Republic of Yugoslavia.
Impact on Infrastructure and Development Resources
Conflict will also indirectly have an impact on health, through its influence on infrastructure
and the determinants of health, such as water and samtation (Table 7). In both southern
Sudan and Uganda, hand pumps in villages were specifically destroyed by activities of
government troops in rebel-held areas, and by guerillas in government-held areas . In the
international coalition against Iraq following its invasion of Kuwait and its repression of
Kurds and Shiites, water supplies, sewage removal and other sanitation services were
drastically affected by saturation bombing.
18
Table 7: Indirect Impact of Conflict on Health and Development (Adapted from 76)
Category of
impact
Infrastructure
damage
Disrupted human
settlement
Environmental
damage
Impact on social
organisation
Macroeconomic
Manifestation
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
Damage to water supplies, sanitation, sewage and garbage disposal
Disrupted electrical power and gas supplies
Destruction of bridges, roads, railways, airports and docks
Communication system dysfunction and targeting_____________________________
Displacement within country as IDEs and across borders as refugees
Movement to cities and safer areas
Forced resettlement as military strategy____________
Environmental destruction from use of e.g. napalm, defoliants, saturation bombing
Environmental contamination from chemicals, radiation
Reduced access to areas as a result of landmines and unexploded ordinance
Reduced availability of wood fuel and other local resources
Destruction of natural resources e.g. forests, gems, wildlife, to fuel ongoing conflict
Impact (positive or negative) on community participation
Direct targeting of community leaders and representatives
Decreased accountability of political systems
Human rights abuses: detention, torture, disappearances, political assassination
Changed (positive or negative) gender relations
Increased inter-group hostility and tension____________________________________
Diversion of economic and productive resources from social to military sector
Hyper-inflation and price manipulation in favour of few entrepeneurial and military
elites
Destruction of local markets
Loss of production, trade and external markets
Increased cost of imports
Increased susceptibility to donor and international financial institution pressure on
how economy functions
Loss of tourism
Community Participation
The impact of conflict on community participation may be positive or negative. In some
countries, there is evidence of conflict enhancing community mobilisation, participation and
control over local decision-making.
Experiences from Nicaragua77, Mozambique78,
Vietnam79, Eritrea80 and Tigray81 suggest that some conflicts may present opportunities for
community mobilisation and organisation as communities come together to respond to
external threats. Such activity may, however, represent a target for opposing groups, and
there is evidence to this effect from Nicaragua and Mozambique.
Macroeconomic Impact
The economic impact of conflict may be profound82 83. In Ethiopia, military expenditure
increased from 11.2% of the government budget in 1974/75 to 36.5% by 1990/91, the health
budget declined from 6.1% in 1973/4 to 3.5% in 1985/86 and 3.2% in 1990/9184. There are
also significant effects on productivity and exports, reductions in the ability to collect tax
revenue, loss of human capital, loss of tourism and other usual sources of income, and great
scope for price manipulation, speculation and profit-making, often at the expense of the
19
majority of the population.
Impact on Food and Agriculture
The specific targeting of food production and distribution activities during periods of conflict
is extensive85 (Table 8). Food production was directly disrupted in Ethiopia through
preventing farmers from planting and harvesting their crops and the looting of seeds and
livestock by soldiers; in Tigray, the conscription of men, the mining of land, the confiscation
of food, and the slaughter of cattle were widespread86. The loss of livestock deprives farmers
of an asset needed to put land into production: it therefore has an adverse effect both
immediately and in the future. In Eritrea, about 40% of the land area was not cultivable due to
similar disruption of activities and access to land. In numerous latter-day conflicts,
humanitarian supplies have been used by competing sides, as a means of controlling
populations, fuelling fighting capacity, or luring populations to areas in which control can be
Forced Resettlement
Forced population resettlement, used by a number of governments for security and ideological
reasons, may also have severe adverse impacts upon health. In the two to three years after
1985, more than 5.7 million people, 15.4% of the total rural population, had been moved to
villages as part of an enforced Ethiopian government programme87.
Table 8: Mechanisms Compromising Food Security during Periods of Violent Political
Conflict (Source: Adapted from 88)
Category of
_____impact
Destruction
Disruption
to
agricultural
and
market activities
Selective
provision of food
Population
relocation
Manifestation
Direct destruction of food stores
Looting of seeds, equipment, and animals
Destruction of markets________________________________________________
Encirclement of towns and curfews limiting access to fields and markets
Reduction in ability to use coping strategies such as migration
Seizure of relief goods and supplies
Disrupted cattle grazing, veterinary services
Manipulation of commodity prices______________________________________
Providing food only to areas w'here political support is provided or desired
Diversion of food aid to military force use________________________________
Forced displacement
Establishment of 'protected villages' or other forms of forced settlement to enable
control over population to be exerted
20
Contributory Causes to Violent Political Conflict
Good public health practice requires a focus also on the identification of risk factors and the
determinants of violent political conflicts, and the development of approaches that can
improve the resolution of conflicts without war.
A range of proximal and distal risk factors for conflicts have been identified. Proximal risk
factors include immediate tensions within and between states in the presence of weapons
availability.
Level of Weapons Technology
The technologic level of available weapons influences the extent of destruction which occurs
in conflicts. It is important to note, however, that even primitive weapons, such as the
machete, can contribute to the occurrence of massive human destruction, as occurred in the
Rwandan genocide89.
The move from the arrow to the crossbow increased the range and destructive force of
projectile weapons in the modem era. This was followed, in succession, by the development
of simple firearms, rifles, machine guns, and submachine guns. The ability to fire more
bullets, more quickly, with greater range and accuracy, has greatly increased the destructive
power of such weapons.
The size of the field of combat has also expanded rapidly. Until the 1800s warfare among
nation states took place on a ‘field of battle’. Mobilization of mass numbers of citizen
soldiers in the Napoleonic wars created larger battle fields, but mobile warfare with rapidly
moving positions in large geographic areas only occurred later with the development of
railways and the mechanization of mass transport. Subsequent development of tanks,
submarines, fighter-bombers, and laser-guided missiles created the possibility of a battle field
without geographic limits. More recent wars, such as that waged by NATO against
Yugoslavia in Serbia (1999) have been labeled as 'virtual war'90 given the extent to which the
conflict was fought with fighter bombers and missile attacks without ground force
involvement.
Indicators of States at Risk
The Carnegie Commission on Preventing Deadly Conflict91 has attempted to identify
indicators of states at risk of collapse and internal conflict (Table 9). These factors interact
with one another to predispose to violent political conflict; on their own none may be
sufficient to lead to violence or state disintegration. Demographic pressures, for example,
become particularly important in the presence of resource constraints, maldistribution of
development gains, and lack of democratic processes through which to decide on how best to
facilitate development despite adversity. A particularly important factor associated with the
occurrence of violent political conflict is the presence of inter-group inequalities, especially in
the presence of widening gaps between groups92
21
Table 9: Indicators of States at Risk (Adapted from 93 94^
Inequalities
Widening inequalities, especially those manifest between, rather
than within, groups______________________________________
Demographic pressures
High infant mortality; rapid changes in population including
massive refugee movements; high population density; youth
bulge; insufficient food supply or access to safe water; ethnic
groups sharing and disputing land, territory or environmental
resources______________________________________________
Lack of democratic processes
Criminalisation or deligitimization of the state; human rights
violations; kleptocratic and corrupt processes of governance_____
Regimes of short duration__________ Rapid changes of regimes_________________________________
Ethnic composition of the ruling elite Political and economic power exercised (and differentially
differing from the population at large
applied) through ethnic and religious identity; desecration of
ethnic symbols by opposing sides___________________________
Deterioration or elimination of public Reduction in the size and performance of social safety nets which
services_________________________ ensure a minimum standard of service available to all___________
Sharp and severe economic decline
Uneven economic development; differential benefits or losses to
one or other group or geographic zone as a result of significant
changes in economy; massive economic transfers or losses over
short periods of time_____________________________________
Legacy of vengeance-seeking group History of inter-group rivalry with previous disputes settled
grievance_______________________ through violence________________________________________
Massive, chronic or sustained human Sufficiently adverse social, political, economic or environmental
flight
conditions to propel large numbers of the population into
displacement within or across borders
Many of these risk factors are identifiable in advance of overt collective violence. Increased
pressure and competition for resources, within a system which inequitably distributes political
and economic power, is a potent input to many conflicts. A key question is whether current
trends in globalisation are likely to increase the frequency and magnitude of these negative
features. There is increasing recognition that both intranational and international violent
conflict is fueled by unequal access to resources and power. Trends in the global economy
have hastened the pace of global integration and enrichment for some countries and groups
within countries, and the fragmentation of societies, associated with the marginalisation and
impoverishment of others. Risks of violent conflict are exacerbated by the massive and rapid
movements of international finances, by aspirations fanned by the global media and
commumcations systems, by the stimulation of ethnic and religious nationalism often
encouraged during periods of economic stress, and by the ready availability of small arms,
other weapons and tools of mass destruction. Conflict is much less likely in situations of
economic growth than in contracting economies and intense competition over resources.
Figure 1 suggests potential linkages between current trends in globalisation and the
occurrence of violent political conflict95
22
Key natural resources, such as diamonds in Sierra Leone, the Democratic Republic of Congo
(DRC), and Angola, oil in Angola and Southern Sudan, and timber and gems in Cambodia,
have all played some role in fuelling and prolonging conflicts. In other settings such as
Afghanistan, Burma, and Colombia, control over drugs production and distribution makes a
significant contribution to promoting violence.
How best to intervene in the presence of such negative features is unclear. Some key nodes
for action include enhancing accountability of decision-making, reducing absolute and
relative poverty, reducing inequity between groups, reducing access to weapons, and ensuring
that development assistance is targeted to where it is likely to make the greatest impact on
poverty reduction96.
Possible Linkages between Globalisation, Inequalities and Conflict
Compromised role of
state
Increasing privatisation
Reduction in social
safety nets
Reduced size of public
sector
Increased degrees of relative
deprivation
Increased competition for
resources
Increased weapons availability
I
I
Globalisation
------ ►
Widening
inequity
>
| ConfiicI
t
Inability of state to
manage political
challenges and maintain
control over use of force
Responding to Collective Violence
Prevention
Attention must be focused on prevention: it is clear that the impact of violent political
conflict, whether in the form of wars, genocide, or state repression, can be horrendous preventing these from occurring must be our primary response. Many 'risk factors' and risk
situations have been identified in the previous section; finding mechanisms for preventing
these occurring is a major challenge.
23
Promotion of human rights and application of international treaties. Among the most
important means of prevention are the promotion and application of internationally agreed
treaties and laws. Respecting international human rights and the rights of the child are
important aspects of developing an appropriate response. International humanitarian law
guides the conduct of armed forces during periods of conflict, but are limited when dealing
with internal conflicts rather than those between states. Key treaties such as the Ottawa
Treaty, with its ban on production, distribution and use of anti-personnel landmines need to
be more vigorously enforced. Those states which have not yet signed up to this treaty, which
include some of the most powerful states in the world today, should be pressured to do so.
New initiatives around light weapons, such as that promoted through the European code of
conduct on light weapons transfer should be widely promoted. Recent measures to integrate
the monitoring of small arms flows with other early warning systems for conflict97 should be
actively applied. Other measures to decrease the production, sale and distribution of
weaponry are also to be encouraged.
The establishment of the International Criminal Court which will ensure a permanent
mechanism for dealing with war crimes and other crimes against humanity, should be actively
promoted. The role of the state remains extremely important in taking forward each of these.
Over thousands of years, warrior codes and rules of conduct have sought to limit the degree of
destruction to economic systems and civilian populations. There was some pragmatism in
this: land, machinery and people were the assets gained through fighting wars; without them
one had only the risk with none of the benefits. International humanitarian law has developed
over time in an attempt to constrain the ways in which wars are waged between states, and to
regulate how armed forces relate to one another. The production of biological, chemical, and
nuclear weapons and tactics of indiscriminant bombing with conventional explosives created
unprecedented levels of mass destruction in the 20th century. Rules to limit the development
and use of such tactics and technologies followed, starting with biological and chemical
weapons early in the century, including those governing treatment of prisoners such as the
Geneva Conventions, and leading to controls over nuclear weapons, chemical and biological
warfare agents, anti-personnel land mines, and blinding laser weapons late in the century.
These rules, however, only apply effectively to nation-states and multi-state coalitions
interested and capable of asserting authority over combatants in their territories.
Early warning systems. Despite increasing attention by the United Nations and other
organisations to identify vulnerabilities and early warning systems which can more accurately
predict where violent political conflicts are likely to occur, we still have little ability to act on
this information. Evidence of increasing tensions between groups, often as a result of
deepening poverty and inequities in the distribution of resources within societies, need to be
identified early and appropriate responses developed. The role of the state and of the
international community, plus key donors, are extremely important in avoiding patterns of
development which contribute to violent conflict occurring.
The health sector may play a valuable role in detecting inequalities in health status, in access
to health care, and in access to other societal resources. Detecting these early and promoting
societal responses to their presence, especially if the gaps between those in good and poor
health are widening, is especially important. The recent World Health Report98 has identified
the reduction of health inequity as a key indicator of health sector performance. The health
sector, through its surveillance mechanisms, may be able to play a role in detecting increases
24
in inter-group tensions and social distress through monitoring trends in diseases of poverty
and their distribution, as well as the occurrence of inter-personal and inter-group violence.
The health sector can play a valuable role in highlighting the health, social and economic
impact of violence, thereby contributing to the search for more equitable systems of
governance and economic and social development.
Aside from totally avoiding violent political conflict, effort can be directed at improving
system performance should violent conflict occur. Disaster planning should include
improving early assessments of vulnerability and of needs, enhancing coordination between
the key actors, and bolstering global, national, and local capabilities to provide effective
health services in situations of adversity. The World Health Organisation has developed
surveillance mechanisms such as HINAP and HEDIP to help identify and respond to such
crises earlier rather than later.
Globalisation also presents opportunities for heightened awareness and knowledge of violent
conflicts and their causes worldwide. The new technologies offer opportunities to exchange
ideas and place pressure upon political leaders to make pro-poor and pro-peace choices. The
establishment of transnational civil society organisations also presents some opportunities for
influencing and pressing for appropriate forms of external aid and support before, during and
in the aftermath of conflicts. Conflict-affected groups can use the new technologies to ensure
that their voices are heard and their experiences documented and placed in the public domain.
Building and boosting resilience. Systems exposed to adverse environments typically adapt
and respond: such adaptations may be positive while others may be dysfunctional, leading to
a worsening of the situation for individuals, communities and systems. The World Health
Organization, supportive academic and non-governmental organizations, can play a valuable
role in conceptualising and documenting adaptations in order to support and bolster those
responses which are most effective, efficient and equitable.
During conflicts adaptations and responses occur at individual, community and societal
levels. Individuals and communities seek to identify mechanisms of ensuring survival, often
entailing one or more of migration, responding to violence with political opposition and/or
violence, or promoting innovative survival and coping strategies to ensure that basic needs
are met. Adaptation occurs at system level with new actors filling the gaps left by retreating
and contracting public sector services. Indigenous health care providers may become more
important, as may other forms of private sector provision. The private health care sector
expands, both through the provision of services by non-profit NGOs, and through the
hemorrhage of public sector workers into the private sector, whether officially or unofficially.
Community involvement. Local systems of democracy and accountability, where they exist,
may be seriously disrupted and involvement in community affairs discouraged during violent
political conflicts. People may fear playing an active role in debates around social policy or
in advocating for the needs of marginalized groups being addressed. This is particularly so in
undemocratic political systems and in the presence of actual or threatened state violence
against opponents.
In some conflicts, however, positive community responses may occur, facilitating
opportunities for health system and societal development. Such responses appear to be more
common in those conflicts which were ideology-based, as in Nicaragua, Vietnam and
Mozambique, in all of which community participation and control was actively promoted as
25
part of a broader socialist political programme. In the popular conflict against the Ethiopian
Derg, community-based political movements in Eritrea and Tigray engaged strongly in
building community structures for participation and decision-making, facilitated the
development of multi-sectoral health promotion strategies, and identified innovative
community-financing systems" 10°.
Gender relations may change substantially during conflicts: in part a reflection of greater
absence of men on a day to day basis within the community, but also reflecting the new roles
absorbed by women during periods of instability101. Changed gender relations may also place
women at risk, for example, through forcing an engagement in transactional sex in order to
maintain livelihoods and earn sufficiently to meet family needs.
Increasing control by those affected. The experiences of those most affected by collective
violence as individuals, members of families and communities, service providers and users of
services, need to be elicited and understood if those responses which are health-promoting for
the greatest number of people are to be supported. Participants in collective violence are not
passive objects and will themselves use opportunities presented through health and social
services, and the provision of relief supplies, to strengthen their own positions and to
mobilise resources for their constituents. This intensely political scenario must be understood
and appreciated if service providers and concerned agencies are to minimise the harm they do
and maximise opportunities for promoting improved health.
Providing Services during and in the Aftermath of Crises
Different forms of conflict logically require different forms of health sector and humanitarian
responses. Some conflicts, whether primarily internal or directed at other states, attract
substantial international attention while others appear to be largely ignored by the media,
donors and service providers. Ideological, economic and political factors influence greatly
the nature of the response; consistency is often lacking despite equivalent loss of life,
population displacement, and infrastructure destruction. Ongoing conflicts in Burma,
Algeria, Sierra Leone, Sudan, Liberia and Angola, for example, have attracted far less
attention and funds in recent years than those in Kosovo or East Timor.
Every conflict has winners and losers. Those who benefit do so through manipulating
scarcity, seizing assets, or selling resources such as gems, timber, or drugs. The winners may
have an interest in perpetuating conflict; humanitarian aid itself may become a resource over
which groups compete, and such assistance and resources may directly or indirectly stoke the
conflict.
Humanitarian workers may be directly targeted in latter-day conflicts102 (also see Box) and
this has led to increased efforts to work closely with the military and security sector, which,
despite some benefits such as in improving logistics support, may bring negative
consequences, additional dangers, and threaten the neutrality and impartiality which many
agencies aspire to.
During internal conflicts, due to scarcity of resources and government difficulties in accessing
populations under the control of insurgents, indigenous and international NGOs, traditional
practitioners, and church groups fill part of the vacuum left by the public sector, especially in
26
rural areas and among populations more directly affected by violence. NGOs often provide a
patchwork of services which are relatively independent of the state and do not necessarily fit
in with other service provision and priorities. They may communicate poorly with one
another, adopt different approaches and standards of care and health worker remuneration,
and focus attention mostly at a very local level. Ensuring a modicum of equity of access and
quality in different parts of the service is an early casualty
Extending emergency responses to promoting more wide-ranging services is another key
challenge. Surgery services developed in response to anti-personnel landmine injuries could
be extended to other forms of injury surveillance and treatment. Mechanisms to protect and
maintain key elements of service provision and functioning, including information systems
and supplies, are crucial to assuring ongoing system functioning. Complex humanitarian
emergencies severely affect health care services both in the countries affected by conflict as
well as those to which refugees have fled.
A major challenge in conflict-affected settings is to seek to minimise the direct and indirect
adverse impacts of the conflict on the health services, personnel and resource availability.
Even where the ’official' health system is destroyed, however, health workers may still be
present within their communities and may be able to offer services. The extent to which
services are able to be maintained depends on earlier disaster preparedness activities (e.g.
training, pre-positioning of drugs and other supplies) as well as on the nature, severity, form
and intensity of the conflict.
When refugees flee from their country and cross borders their usual sources of health care are
lost. They become dependant on what they can provide for themselves and what can be
provided for them in the host country to which they have fled either through existing host
country services, or through additional services offered by non-governmental and UN
agencies. Host government services may become overwhelmed, both by the number of
service users and by their need for services, if large numbers of refugees suddenly move into
an area and seek to use local health services.
Host-country services typically do not receive additional resources despite the increased
needs and therefore have to cope as best they can with the additional service load. This may
disadvantage local community members. While host communities will often accept
compromises in services available to them, these may fuel tensions between the host and
refugee populations if services available are inequitable in the longer term. Antagonism may
also be experienced if refugees are offered access to host services at no or lesser cost, while
local community members may be required to pay user charges, both informal and formal, to
obtain care.
For host communities and health service providers, a key challenge is to utilise opportunities
presented by the influx of refugees, and with them other organisations and resources, to
ensure that services access and quality are improved and system capacity strengthened.
When the Kosovar Albanians fled into Macedonia and Albania, for example, there were some
attempts, by WHO and others, to support the existing health and welfare systems of these
countries to cope, rather than simply bringing in a parallel system provided through NGOs.
Even in circumstances where the international community through UNHCR, and a wide range
of expatriate NGOs are providing services to the refugee community, an impact on local
health services may well be felt. This may result from the recruitment by these agencies of
27
local health workers, thus depleting indigenous systems of their usual human resources, or
from additional needs placed on other levels within the health services through requirement
for referral services, rehabilitation services, and chronic disease services which the NGOs
may be ill-prepared or unwilling to address.
Humanitarian organisations seek to reduce excess loss of life, and to re-establish an
environment in which maintaining and promoting health is possible. A much debated issue is
the extent to which these interventions should focus solely on immediate and short-term
needs, or should have longer-term objectives in mind. Many relief and humanitarian
organisations see their primary role as saving lives which have been placed at risk as a result
' of extraordinary events. In such situations, doing whatever is necessary, within the available
resources, is deemed appropriate even if some of what is done cannot be replicated or
sustained over the longer term. The emergency phase is seen as lasting a variable period
which ends when excess mortality is reduced to less than 1/10,000 per day, or at least more
closely resembles the pre-emergency and host-population levels of mortality.
Organisations which typically espouse a development rather than relief-oriented approach
have sought to place the issue of 'developmental relief onto the agenda, arguing that early
attention to the difficult issues of efficiency, effectiveness, sustainability, equity and local
ownership will be beneficial in the longer term. If one adopted the latter approach, greater
effort would be given to activities such as training, building local capacity, and keeping costs
down, rather than seeing these as desirable, but not practical given the acute needs faced in
relation to saving lives.
An additional key concern facing the range of organisations offering services in response to
humanitarian crises, is the importance of coordination. Organisations need to work together
very closely if they are to reinforce each others action, maximise the use of available
resources, minimise duplication and overlap, and enhance effectiveness, equity and
efficiency. The Code of Conduct for Humanitarian Organisations (see Box, Part A), as put
forward by the International Federation of Red Cross and Red Crescent Societies103, states a
number of key principles which many, but not all, humanitarian organizations identify as
underpinning their work. Such Codes are aspirational, however, and there are no effective
measures for enforcing them nor for evaluating the extent to which such principles are
effectively implemented.
Related guidelines for improving service provision in armed
conflicts have been put forward (see Box, Part B) by other commentators104 and deserve
attention.
28
Box 1
Proposed Codes of Conduct for Humanitarian Activity
A. Principles of Conduct as stated by the International Red Cross and Red Crescent Movement
and NGOs in disaster response programs
1. The humanitarian imperative comes first.
2. Aid is given regardless of race, creed or nationality of the recipients and without adverse
distinction of any kind. Aid priorities are calculated oh the basis of need alone.
3. Aid will not be used to further a particular political or religious standpoint.
4. We shall endeavour not to act as instruments of government foreign policy.
5. We shall respect culture and custom
6. We shall attempt to build disaster respons on local capacities.
7. Ways shall be found to involve program beneficiaries in the management of relief aid.
8. Relief aid must strive to reduce future vulnerabilities to disaster as well as meeting basic
needs.
9. We hold ourselves accountable to both those we seek to assist and those from whom we accept
resources.
10. In our information, publicity and advertising activities, we shall recognize disaster victims as
dignified humans, not hopeless objects.
B.
1.
2.
3.
4.
5.
6.
7.
8.
Providence principles of humanitarian action in armed conflicts
Relieving life-threatening suffering: humanitarian action should be directed toward the relief
of immediate life-threatening suffering.
Proportionality to need: humanitarian action should correspond to the degree of suffering,
wherever it occurs. It should affirm the view that life is as precious in one art of the globe as
another.
Nonpartisanship: humanitarian action responds to human suffering because people are in need,
not to advance political, sectarian, or other extraneous agendas. It should not take sides in
conflicts.
Independence: in order to fulfil their mission, humanitarian organisations should be free of
interference from home or host political authorities. Humanitarian space is essential for
effective action.
Accountability: humanitarian organizations should report fully ontheir activities to sponsors
and beneficiaries. Humanitarianism should be transparent.
Appropriateness: humanitarian action should be tailored to local circumstances and aim to
enhance, not supplant, locally available resources.
Contextualization: effective humanitarian action should encompass a comprehensive view of
overall needs and of the impact of the interventions. Encouraging respect for human rights and
addressing the underlying causes of conflicts are essential elements.
Subsidiarity of sovereignty: where humanitarianism and sovereignty clash, sovereignty should
defer to the relief of life-threatening suffering.
Recognition by the humanitarian community of these problems has led to measures aimed at
improving practice and accountability, including the development of minimum standards for
service provision in emergencies (the Sphere Project), the promotion of a humanitarian
ombudsman, and the research efforts through, among others, the WHO. Promoting the
derivation and uptake of good practice is particularly difficult in humanitarian agencies given
rapid staff turnover, resistance to revealing failures and limitations given perceptions of
negative funding consequences, and a culture of doing rather than reflecting. Interventions
may be inadequately based on evidence and despite most agencies valuing the concept of
coordination, few wish to be coordinated. Poor quality services have significant adverse
29
consequences: increased morbidity, mortality, disability, further spread of communicable
diseases, community dissatisfaction and break-down, and psychosocial distress. Clear policy
objectives for interventions are often lacking and mechanisms for working with new players
such as the military and the private sector remain inadequately developed. Despite recognition
that the accountability of relief efforts to affected populations should be enhanced,
mechanisms to assure this are in their infancy105
Ongoing humanitarian challenges include understanding how best to upgrade host population
health services alongside efforts to improve those available to refugees; how to most
humanely and efficiently provide good quality services; and how to maintain the role of
communities in structuring both the determination of priorities and the pattern of service
provision. A key issue relates to how and whether to bolster and support resilient health and
social systems and individual adaptations to conflict: our level of knowledge regarding these
responses, and the potential to further support them, is weak. A persistent challenge to
humanitarian workers is to institutionalize a sensitive and inclusive evidence-based culture
and to build sustainable mechanisms for crystallizing policy advice from the vast and
valuable foundation of field experience106.
Another key set of concerns relates to the ethics of intervention and how assistance is
distributed. In some settings, such as Somalia, agencies felt obliged to hire armed guards and
militia from various factions in an attempt to ensure a degree of safety in maintaining their
operations. In many settings, there is a degree of expectation and tolerance that a proportion
of aid resources will be diverted to the military combatants waging the conflict - agencies
have at times felt that such 'leakage' is acceptable given that most resources still reached their
destination. However, in some settings, such a large proportion of food aid and other
resources was siphoned off by the warring parties that agencies decided to withdraw from
providing services. Other ethical concerns also arise from working with warring factions and
indirectly conferring upon them a degree of legitimacy. Mechanisms for deciding whether to
be silent, to speak out against abuses observed, or to withdraw from service provision are all
possible options, none of which can be lightly taken. The broader debates regarding how aid
may directly or indirectly support peace - or war - have been clearly raised in recent
commentaries107.
Post-conflict Re-establishment of Services and Systems
There is increasing awareness of the issues which need to be addressed as countries emerge
from major periods of conflict108. Supporting countries and agencies to do so is a significant
challenge attracting effort and analysis. Table 10 suggests some of the typical responses to
post-conflict health system rebuilding, highlights the rationale and problems of such
responses, and proposes some improvements for debate.
30
Table 10: Post-conflict Health Challenges
Component of
post-conflict
health sector
development
Establishment of
policy framework
Donor
coordination
Attitudes to
working with
government
Infrastructure
development
Specific disease
problems
Bringing together
conflicting sides
Role of private
sector
Promotion of
equity
Typical responses
Towards a more appropriate response
Activities seen as
’projects'; limited
attention to establishing
a policy framework
Early planning and development of policy framework within
which project activity can be based; donor support to Ministry of
Health policy development capacity including collation and
dissemination of information, and facilitation of communication
and debate between key stakeholders_______________________
Identify areas of common interest and build around these;
strengthen Ministry of Health capacity to give direction and
provide policy framework, coordinate donors and NGOs, and
identify areas of need
All agree that
coordination should
take place but none
wishes to be
coordinated_________
Government often
bypassed with support
going through NGOs
and UN authorities
Attempt to reconstruct
whatever previously
existed_____________
Vertical programs with
large amount of donor
funding and high levels
of donor control______
Promotion of corridors
of tranquility and other
methods to cease
conflict temporarily to
promote disease control
Encouragement to
diversifying range of
providers and
deregulating private
sector______________
Considered important
and to be addressed in
the longer-term
Emphasis on
training
Often overlooked,
fragmented and
uncoordinated
Emphasis on
information
systems and on
data-based
decision-making
Information considered
a luxury and not shared
or lost
Reform to international aid system so as to allow development
type activity and funding to take place earlier on in a period of
post-conflict recovery
Review needs and distribution of services - identify opportunities
to rationalise and make more equitable distribution of available
services; respond to changed population distribution in region
Consider implications of range of vertical programmes on
linkages between different parts of health sector
Consult and involve relevant stakeholders: including national and
local public sector, NGOs, and private sector_________________
Recognise symbolic nature of health care in restoring inter
community relationships; see reconciliation as a long-term issue
requiring rebuilding of own community and of trust between
communities; promote opportunities to work together but
appreciate role of time in reducing distrust___________________
Recognise but seek to constrain private sector provision; promote
state role in providing policy framework and setting quality
standards; consider incentives to promote equitable access and
delivery of services of public health importance.
Identify equity as key objective of post-conflict reform; recognise
that equity concerns may, for interim period, be over-ridden in
order to promote greater stability; build links between competing
groups, different areas, men and women, as key elements of postconflict reform_________________________________________
Identify human resource development as integral component of
maintaining and developing appropriate services; develop
strategies for integrating workers trained in separate systems;
invest in management training_____________________________
Prioritise documentation, health intelligence, central repository
for information; use new technologies e.g. world-wide web for
making information available; make funding conditional on
making information available to all; facilitate learning from other
post-conflict settings
31
Other Public Health Roles: Documentation, Advocacy, Health as a Bridge to Peace
Surveillance and documentation remain core areas for public health action. This chapter has
argued that data are often imprecise and that improving them will allow more appropriate
planning, analyses of experiences and establishing what does and does not contribute to
improving health in these adverse settings. We have also argued that data are extremely
politically sensitive thus highlighting the role which organizations such as the World Health
Organization can play in placing more objective assessments on the table for debate arid
analysis. However, an obsession with the precision of data may not be warranted in most
situations of collective violence; a concern with validity is, however, paramount.
Advocacy is a fundamental component of public health action. NGOs, UN agencies, and
health professionals all have important advocacy roles. Some NGOs, such as Amnesty
International, have an explicit mandate to speak out about human rights abuses, as do certain
parts of the United Nations such as the Special Rapporteur on Human Rights. Some NGOs,
such as Medecins sans Frontieres, see the witnessing and recording of abuses, and the
advocacy around protecting basic needs and human rights, as central to their mandate. Other
agencies are more reticent to speak out against key parties to the conflict for fear that this will
undermine their ability to deliver much needed services. In some such cases NGOs have
provided information on abuses of which they are aware, to third parties and the press in
order to ensure that the international community are appraised of such abuses. Health
professionals have a privileged position in being networked globally, and through
organisations such as Physicians for Human Rights and health professional associations, are
able to monitor, record and advocate against human rights abuses. In South Africa under
apartheid, different groups of health professionals played different roles, with some clearly
supporting the structural violence of apartheid and the use of violence by the state against the
community, whereas others worked individually and in organisations to highlight and oppose
apartheid and its manifestations in state violence109.
Data and the analysis of experience are necessary if advocacy is to be effective. In addition,
building up some understanding of effective interventions, what works and does not, and in
what circumstances and with what costs, are essential to motivating for more appropriate
action. While policy is not made in a vaccuum, it can be better informed by data and
experience.
Desire for improved health, and for safety and security, are universal. Health promotion
efforts may help identify adverse influences on health which are amenable to control. In
societies which are riven with tension and conflict, promoting health may offer opportunities
to identify longer-term objectives across communities which will help lay down future
patterns of development which are health enhancing and community-building.
Healthrelated peace-building activities provide avenues for facilitating the reestablishment of social
structures and livelihoods. Critically examining the potential role of health services in peace
building while gaining understanding of the limitations of such efforts, remains crucial.
32
Recommendations
We face considerable challenge in developing skills, awareness and knowledge of effective
preventive and responsive interventions with which to respond to collective violence. Here
we highlight a small number of recommendation that spring from the analysis presented
above. First and foremost, however, is the importance of recognizing that the health system
response is but a small element of possible mechanisms for preventing and responding to
violent political conflict. Clearly, political measures need to be taken to predict and avoid
conflict and to develop means for limiting its scope and resultant damage. Strengthening the
United Nations is the only global mechanism we currently have to undertake such analyses
and to promote patterns of development that will help reduce the risk of conflict. Working
together with concerned states and with global civil society organisations, the promotion of
forms of development which help reduce inequity and conflict between sub-groups within the
population is fundamental, as are mechanisms to ensure that states fulfil their obligations
towards the populations that live within their borders.
Definitions
Public health personnel should refine the definitions they use to categorise different forms of
collective violence, as well as to describe and quantify their impact.
Data and Surveillance
a) Health and health service-related indicators should be identified and mechanisms to collect
them promoted so as to ensure that early departures from health among particular groups and
early indications of inter-group tension and rivalry can be identified, b) Knowledge of
innovative data and surveillance techniques which have developed over the last two decades
for describing and analysing the health status of conflict-affected populations should be
further refined to improve our understanding of the impact of conflict on internally displaced
populations, populations which have integrated with host communities, and specfic sub
groups of the populations such as child soldiers. C) Improved methods of analysing how
health systems are affected, and respond to adversity, is essential if good practice in
stimulating preparedness, boosting resilience, and generalising effective responses is to be
ensured; d) development of effective mechanisms for 'capturing’ the experiences of conflictaffected populations in order to record and analyse these, placing them in the public domain
and playing some role in ensuring that population needs are addressed. E) Establishing a
post-event analysis to describe more objectively the build-up to the violence, the response and
impact, and the behaviour of the UN and NGO agencies, would provide significant
opportunities for learning. Some such analyses have been conducted, notably following the
Rwandan genocide110 and provide an extremely valuable opportunity to further enhance
prevention and response activities.
33
Prevention
Effort must be directed at prevention. Close collaboration with the UN and other agencies is
essential if vulnerabilities to conflict are to be identified early and more appropriate forms of
development promoted. Current forms of globalisation appear as though they may increase
the risk of further conflicts occurring. Urgent work to determine how globalisation and
development patterns are linked is required. Key interventions include strengthening the
United Nations, reducing arms transfers, promoting the adoption of treaties governing the
production, distribution and use of anti-personnel landmines. Enhancing respect for and
adoption of human rights laws and international humanitarian law are key interventions to
promote at country level.
Monitoring the down-sides to globalization and promoting more equitable forms of
development and more effective pro-poor development assistance will all contribute to
reducing conflict occurrence.
Boosting resilience: The responses of individuals, communities and systems to insecurity and
adversity need to be better documented and analysed. Effective responses should be
supported and knowledge of these practices generalised in order to assist struggling health
systems to cope with situations of adversity.
Improving the quality of humanitarian and other responses - standards and accountability:
Efforts to improve the standards and accountability of those responding to collective violence
should be supported and bolstered. Current efforts such as the Sphere Project to establish
minimum standards for humanitarian interventions, and the Ombudsman project to enhance
accountability to affected communities should be actively promoted. Improving the
knowledge base upon which good practice is determined and promoting means of extending
knowledge and uptake of these findings should be supported.
Enhanced post-conflict responses - reducing likelihood of future conflicts occurring: Health
system development in the aftermath of conflicts must take cognisance of the lessons which
are beginning to emerge about how best to institute and formulate appropriate policies in
these settings. Further documentation of good practice is required and support needed for
organizations like WHO to play their part in facilitating more effective policy formulation
and implementation; Opportunities to overcome many of the factors which led to the conflict
occurring should be identified and targeted for action in the post-conflict environment. The
potential, and limitations, of health and health sector activity as a contributor to peacebuilding
should be actively explored and documented. Increased appreciation of the sensitivity of
health sector action conflicted societies is desirable at the same time as building upon
emerging good practice examples of what can be done to promote peace in unstable settings.
Research Needs
There is a clear need for further research, documentation and analysis if we are to improve
our understanding of the factors contributing to violent political conflict occurring, as well as
to understanding the scope for prevention, improving the response to violence when it occurs,
and responding to the health and system challenges which exist in the aftermath of major
periods of violence.
34
A key question is to identify why certain countries which have a number of features of
societies which are likely to descend into major conflicts are able to contain this (eg, Mexico,
Nigeria, Cuba), whereas in others conflict has occurred or, worse, have led to the almost total
collapse of some states (eg, Somalia, Sierra Leone, Liberia, Angola, former Yugoslavia),
while other conflicts have been contained and the nation state as a whole has continued to
function and continue to provide relatively good public services (eg, Sri Lanka, Indonesia,
Algeria).
35
Box 2
Small Arms and Light Weapons
An issue of emerging importance on the international agenda is the presence of large
quantities of small arms and light weapons in many parts of the world. During recent years,
the focus of the international community has shifted to these weapons because they have
become the weapons of choice in most of today’s armed conflicts. Although no clear
definition exists, small arms are generally accepted to include pistols, rifles, sub-machine
guns and ammumtion for them. Light weapons are generally accepted to include small calibre
cannons, light support weapons, combat grenades, anti-personnel mines, mortars, anti-tank
weapons, anti-tank mines, shoulder-fired surface-to-air missiles and their ammunition.
While the extent to which such weapons are available is not argued to cause conflicts, it is
argued that proliferation of such weapons has increased regional instability, threatens the
security of civilian populations, and is a major impediment to post-conflict development.
Examples of such situations include countries like Angola, Liberia, and Afghanistan. The
abundance of weapons, their use, threat of use, and misuse tends to degrade all of the
dimensions of an individual’s health - his or her social, mental, and physical well being. It is
estimated that xxx,xxx small arms-related deaths occur world-wide each year. The number of
non-fatal injuries is unknown. Original data from a variety of contexts of armed conflict
suggests that at least 35% of those injured or killed by such weapons are civilians. In addition
to these direct effects, indirect health effects also arise through factors such as population
displacement.
Although most of the international community’s attention is focused on legal and
humamtanan aspects that relate to transfer of these weapons, it is important to recognise their
public health impact and the role public health agencies can play in preventing death, injuries
and other adverse effects resulting from the abusive use of these weapons.
The role of public health agencies is multiple. Epidemiologic methods can be applied,
whether through descriptive or analytic studies. More complete data on victims and the
circumstances in which their injuries occur is needed to set priorities for interventions and to
evaluate the impact of these interventions. Analytic studies are needed to elucidate risk and
protective factors, and public health needs to develop the appropriate programmes to provide
emergency pre hospital and hospital care, as well as long term physical and psycho-social
rehabilitation.
In short, complex connections between various factors need to be elucidated, understood, and
communicated to those whose decision-making impacts on root causes. Despite the fact that
the issues involved cut across domains of the health, social, and political sciences, in many
ways this is a traditional type of challenge for public health.
Weapons tend to be viewed in terms of their political attributes of imparting power or
advantage. In some senses this has dissuaded those in the health sector from becoming
involved in complex issues such as arms transfer. However, weapons do not only possess
political attributes - they also possess the inherent capacity to inflict death and suffering.
Conveying this in a compassionate and credible manner is an important challenge for those
concerned with the health of populations.
36
Box 4
Child Soldiering: Questions and Challenges for Health Professionals
The issue of involvement of children1 in armed conflict as participants has not received much
attention as a health issue although there are numerous health implications and an important
role for health professionals in preventing child recruitment and in the demobilisation and
reintegration process and follow-up thereto.
Clearly the involvement of children as combatants in armed conflicts exposes them to risks of
death and combat-related injury. What is less obvious are the other health implications,
including mental and public health aspects, some inherent and some arising from particular
conditions or situations, or the type of use and abuse to which child soldiers are prone. What
follows is a brief introduction to some of these issues.
The generally quoted estimate of “at least 300,000 children currently participating in armed
conflicts”2 should be treated with scepticism. First, there is no way of knowing what the
numbers really are, and this was the most conservative estimate at the time in order to ensure
that it could be challenged as being exaggerated. Secondly, a figure of this sort is irrelevant
because it fails to convey the essentially cumulative nature of the problem: today’s “300,000”
are not the same children as yesterday’s or tomorrow’s. Some of the children will have been
killed, demobilised, invalided out, or become adults (that is passed their 18th birthday), while
new children will have been recruited. The effects of their participation in combat do not
cease simply because of increased age or demobilisation. Finally, this does not of course
include the thousands of under-18s who are in armed forces or groups but not currently
engaged in armed conflict.
In so far as this issue is considered only in the context of participation in combat, inevitably
the distribution of child soldiers by region predominantly reflects the distribution of conflicts,
although the relative sizes of armies affects the absolute numbers of children involved as
compared with the proportion of the population or the number of situations in which children
are involved. Thus the large number of conflicts in Africa at present, does not necessarily
mean that there are more child soldiers in Africa than say in Asia where the sizes of the
armies tend to be much greater. Unless children are routinely recruited into armed forces,
they normally become involved only after the conflict has been in progress for some time.
However, once children start becoming involved the numbers involved escalate rapidly and
the ages decrease. At any time, the majority will be teenagers but the ready availability of
simple-to-use lightweight automatic weapons means that children as young as ten now serve
as front line troops. Although even younger children are reported to be involved, they
normally serve as spies, messengers, guards, camp servants, and similar roles. In those
situations where girls are recruited as well as boys, normally about one-third of the child
soldiers are girls, and most of them are combatants even if their role also requires them to
provide sexual services, whether or not dignified as “wives”.
1 The term “child” is being used in this context to cover all those up to the age of 18 years in line with the
general definition in the Convention on the Rights of the Child.
2 See, for example, Coalition to Stop the Use of Child Soldiers booklet “Stop Using Child Soldiers”, (Coalition
and International Save the Children Alliance, London, 1998)
37
The research on child soldiers3 undertaken for the UN Study on the Impact of Armed Conflict
on Children (Machel Study)4 showed that the most frequent combat-related injuries of child
soldiers (as opposed to adult soldiers) were loss of hearing, loss of sight and loss of limbs.
These partly reflect the greater sensitivity of children’s bodies, for example, the eardrum, and
partly the uses to which children are more prone to be put, such as laying and detecting anti
personnel landmines. However, child recruits are also prone to non-combat related health
hazards caused by carrying heavy loads, including weapons, malnutrition, infectious diseases
such as malaria, skin and respiratory infections.
Although not an intrinsic part of military service, the fact is that girl recruits are often
expected or required to provide sexual services as well as to fight. This exposes them to high
risk of sexually transmitted diseases, HIV/AIDS, as well as to the dangers of abortion or child
birth. (The military are one of the high risk groups for both contracting and spreading HIV
infection). Younger boys too are often sexually abused. In addition, child recruits are often
given drugs and/or alcohol in order to enable or encourage them to kill and to commit
atrocities, creating problems of substance dependency in addition to the general health
hazards.
Teenagers recruited into regular government armed forces are usually subjected to the same
military discipline, including initiation rites, toughening up exercises, punishments, and
denigration designed to break the will, as adult recruits. The impact on adolescents is often
damaging mentally and emotionally as well as physically. The whole issue of the mental and
emotional impact on children and adolescents of their involvement in armed forces and armed
groups and participation in combat is an area that merits further consideration.
As WHO pointed out in its contribution to the Machel Study5:
“The vulnerability of children engaged in combat is directly related to the age and
developmental maturity of the child. A child’s development is a process of mastery of
its ever-enlarging environment. Violence and fear run contrary to that need of
mastery. The repeated direct exposure of children as perpetrators to violence may
lead to persisting patterns of problematic behaviour and functioning. Many children
may be withdrawn, depressed and display difficulties in social relationships and at
school. Others, particularly the “successful” child soldier, are likely to adopt an active
role, becoming the agent of aggressive behaviour rather than becoming its passive
victim. ... Children going through the development stages of socialization and
acquisition of moral judgement in such an environment are ill-prepared to be
reintegrated into a non-violent society. They acquire a premature self-sufficiency,
devoid of the knowledge and skills for moral judgement and for discriminating
inappropriate risk behaviours whether reflected in violence, substance abuse or sexual
aggression. Their rehabilitation constitutes one the of the major social and public
health challenges in the aftermath of armed conflict.”
3 Published as Rachel Brett and Margaret McCallin: Children: The Invisible Soldiers (Stockholm, Radda
Bamen, 2nd edition 1998)
4 The Study was requested by the Committee on the Rights of the Child and mandated by the UN General
Assembly in 1994. The Final Report is contained in UN Document A/51/306, which presented to the UN
General Assembly in 1996.
5 Family and Reproductive Health and the Division of Emergency and Humanitarian Action: “The Impact of
Armed Conflict on Children: A Threat to Public Health” (WHO, Geneva, July 1996), p 54
38
Key issues for medical professionals include:6
•
The need for thorough (but sensitive) medical screening of all former child soldiers at
the earliest possible opportunity. This may take place at the time of formal
demobilisation, but may also occur when child soldiers are captured, escape, or
otherwise leave service.
•
Such screening may need to be done in stages, addressing the most vital problems first
and only moving on to more sensitive issues, such as sexual abuse, sexually
transmitted diseases, and so on at a later stage when greater confidence has been
established between the health professional and the child. Issues of culture and gender
sensitivity inevitably arise (such as the appropriateness of opposite sex health
professional questioning about sexual abuse), and the availability of continuing,
appropriate health services and support for male and female former child soldiers for
whom longer-term treatment or care is needed.
•
Be aware of the need to look for “hidden” injuries, for example vitamin deficiencies
or hearing impairments (which may be more prevalent amongst child soldiers than
amongst adult ones, and unless identified and addressed may continue to impact on
the child’s development or ability to reintegrate into society, education, vocational
training and employment).
•
It is important to realise that many child soldiers may not be identified as such in the
course of a demobilisation process. In particular, girls may be listed as “dependents”
or “camp followers” even though they have also been fighters, and even if this puts
them in a position of dependency on those who (individually or collectively) abducted
or otherwise forced or persuaded them to join in the first place. Younger children may
also be “screened out” either because they are politically embarrassing and/or because
they cannot comply with demobilisation requirements such as handing in a gun.
•
Adolescent girl soldiers frequently suffer from loss of menstruation brought on by
malnutrition and trauma.7 This in itself may add to their worries and they may need
reassurance that this is normal given their circumstances.
•
In general, attention needs to be given to the mental/psychosocial state of child
soldiers as well as to their physical health situation. There may be nightmares,
hallucinations, chronic anxiety, avoidance, flashbacks, regression in behaviour,
increased substance abuse as a coping mechanism, poor concentration and memory, a
sense of guilt, refusal to acknowledge the past, poor control of aggression, obsessive
thoughts of revenge, feelings of estrangement from others. In addition, their social
behaviour may be altered, with “militarised behaviour” leading to non-recognition or
6 This section draws heavily on the UNICEF/Save the Children: Report of the workshop “The Challenge of
Child Soldiers”, Colombo, Sri Lanka, 24 March 2000
It has been reported that all the adolescent girls in the GUSCO rehabilitation programme in Northern Uganda
suffer from this problem.
39
poor recognition of the norms of civilian society.
•
As with all children, each child is different, an individual, as well as falling within a
particular age group. Any programmes obviously need to take account of this. For
child soldiers, there may be external factors which also need to be considered: what
was their role; how were they recruited; have they “won” or “lost” the war; or has it
ended inconclusively; or is the war ongoing; how long were they involved; were they
protected, perhaps remaining within a family structure, or abused and exploited; what
is the situation to which they will be returning. These and other factors may not only
affect how the children see themselves but also how others perceive them - as a
threat, as heroes, stigmatised as killers and those who committed atrocities; as victims,
as “used goods” in the case of girls who have been married or sexually exploited.
There may also be security issues making it impossible for them to rejoin their
families or to be reintegrated into their home communities.
•
At the same time as addressing the results of child participation in conflicts, health
professionals may be able to play a valuable educational role in helping to prevent
children being recruited - including as volunteers - by raising awareness amongst at
risk groups of children and adolescents themselves and their families and
communities of the dangers of becoming involved, including the psychological and
mental health impacts. At the same time, they may also be in a position to alert
governments and armed groups themselves to the dangers for children of becoming
soldiers, and the possible long term impact on society of training children in arms and
involving them in fighting.
40
References
1 World Health Organization. Apartheid and health. Geneva, World Health Organization
1983.
2 Burkholder BT, Toole MJ. Evolution of complex disasters. Lancet 1995;346:1012-15.
3 Goodhand, J. and D. Hulme. 1999. “From wars to complex political emergencies:
understanding conflict and peace-building in the new world disorder,” Third World Quarterly,
Vol. 20, no. 1: 13-26.
4 Leaning, J. 1999. Introduction in Leaning et.al. (eds). 1999. Humanitarian crises: the
medical and public health response. Cambridge, MA, Harvard University Press pp 1-11.
5 KaldorM. 1999. New Sold wars. Organized violence in a global era. Cambridge, Polity
Press.
6
KaldorM. 1999. New & old wars. Organized violence in a global era. Cambridge, Polity
Press.
7
Orentlicher DF. 1999. Genocide. In Gutman R, Rieff D (Eds). Crimes of war. What the
public should know. London, WW Norton and Co. pp 153-157
8
Sivard R.L. World Military and Social Expenditures (14th Edition).
Washington, World
Priorities, 1991.
9
Sivard R.L. World Military and Social Expenditures (16th Edition).
Washington, World
Priorities, 1996.
io
Wallensteen P and Collenberg M. Armed conflict 1989-1998. Journal of Peace
Research 1999; 36(5): 593-606.
Wallensteen P and Collenberg M. Armed conflict 1989-1998. Journal of Peace
Research 1999; 36(5): 593-606.
12
Garfield R, Neugut A. Epidemiologic Analysis of Warfare. JAMA 1991; 266(5): 688-692).
13 Zwi A, Ugalde A, Richards P. The effects of war and political violence on health services. In
Kurtz L. (Ed.), Encyclopaedia of violence. Academic Press, 1999, pp 679-690..
41
14
Ball P, Seltzer W, Scheuren F, Spirer HF. Multiple or N-System Estimates of the
Number of Political Killings in Guatemala.
15
Ball P, Kobrak P, Spirer HF. State Violence in Guatemala, 1960 - 1996. Washington,
AAAS, 1996.
16
Lumpe L. (Ed). 2000. Running guns. The global black market in small arms. London,
Zed Books.
17 Sheil M, Gutirrez Ml, Bolton P, Spiegel P, Thieren M, Burnham G. Deaths among
humanitarian workers. British Medical Journal 2000; 321: 166-168.
18
Sivard R.L. World Military and Social Expenditures (16th Edition).
Washington, World
Priorities, 1996.
19 Washington Post, 1999. 10/24/99 That Kosovo ‘Genocide’ - the Numbers Don’t Add Up.
20 Black I. Why count the bodies when it's not our war? Guardian. 1993; 18 September
1993. pp.27 London.
21
22
23
Meyers B. Disaster study of war. Disasters, 15 318- 30, 1991.
Deacon
Reed, H.; Haaga, J.; & Keely, C. (Eds.). (1998). The demography of forced migration:
Summary of a workshop. Washington, DC: National Academy Press.
24
Reed, H.; Haaga, J.; & Keely, C. (Eds.). (1998). The demography of forced migration:
Summary of a workshop. Washington, DC: National Academy Press.
25
Hampton J (Ed). 1998. Internally displaced people: a global survey. London,
Earthscan, Norwegian Refugee Council and Global IDP Survey.
26
International Federation of Red Cross and Red Crescent Societies. World Disasters
Report 1999. Geneva, 1999.
42
27 Kumaranayake L, Zwi A, Ugalde A. Costing the direct health burden of political violence
in developing countries. In Brauer J and Cissy W (Eds; Economics of Conflict and Peace.
Avebury, 1997: 292-301.
28
Khaw A. J., Salama P, Burkholder B, Dondero T.J. HIV risk and prevention in emergency-
affected populations: A review. Disasters 2000; 24(3): 181-197.
29 Smallman-Raynor M. and Cliff A. Civil war and the spread of AIDS in central Africa.
Epidemiology of Infectious Diseases, 107(1)69- 80, 1991.
30
Toole M.J. and Waldman R.J. Prevention of excess mortality in refugee and displaced
populations in developing countries. JAMA, 263 (24), 3296- 302, 1990.
31
Toole MJ, Waldman RJ and Zwi AB. (2000). Complex humanitarian emergencies. In
Black R, Merson M and Mills A. Textbook of International Health.
32
Kloos H. Health impacts of war in Ethiopia. Disasters, 16 347- 54, 1992.
33
Zwi AB. Numbering the dead. Counting the casualties of war. In Bradby, H (Ed.).
Defining violence. Understanding the causes and effects of violence. Aidershot, Avebury
Press, 1996: 99-124.
34
Dodge C.P. Health implications of war in Uganda and Sudan. Soc. Sci. Med. 31 (6), 691-
8, 1990.
35
UNICEF. 1989. Children on the front line. The impact of apartheid, destabilization and
warfare on children in southern and South Africa. 3rd ed. UNICEF, New York.
35 Garfield R. The impact of economic sanctions on health and well-being. Relief and
Rehabilitation Network Paper #31, November 1999.
37 Mann J., Drucker E., Tarantola D., et al. Bosnia: The war against public health. Medicine
and Global Survival, 1 130- 46, 1994.
38
Horton R."... on the brink of humanitarian disaster". Lancet, 343 1053- 1994.
43
39 Ugalde A, Selva-Sutter E, Castillo C, Paz C, Canas S. The health costs of war: can they
be measured? Lessons from El Salvador. British Medical Journal 2000; 321: 169-172.
40 Garfield R.M., Frieden T., and Vermund S.H. Health-related outcomes of war in
Nicaragua. American Journal of Public Health 1987; 77 (5): 615- 8.
41
Kloos H. Health impacts of war in Ethiopia. Disasters 1992, 16 347- 54.
42 Cliff J. and Noormahomed A.R. Health as a target: South Afriica's destabilization of
Mozambique. Soc. Sci. Med. 1988; 27 (7), 717- 22.
43 Garfield R.M., Frieden T., and Vermund S.H. Health-related outcomes of war in
Nicaragua. American Journal of Public Health 1987; 77 (5): 615- 8.
44 Kloos H. Health impacts of war in Ethiopia. Disasters 1992, 16 347- 54.
45 Goma Epidemiology Group. Public health impact of Rwandan refugee crisis: what
happened in Goma, Zaire, in July 1994? Lancet, 345 339- 44, 1995.
46 Zwi A.B. and Cabral A. J. High risk situations for AIDS prevention. Br. Med. J. 303 1527- 9,
1991.
47 Kloos H. Health impacts of war in Ethiopia. Disasters, 16 347- 54, 1992.
48 Khaw A.J., Salama P, Burkholder B, Dondero T.J. HIV risk and prevention in emergencyaffected populations: A review. Disasters 2000; 24(3): 181-197.
49 Smallman-Raynor M. and Cliff A. Civil war and the spread of AIDS in central Africa.
Epidemiology of Infectious Diseases, 1991; 107(1) 69- 80.
50 Mann JM, Tarantola DJM, Netter TW (Eds). 1992. AIDS in the world. Cambridge,
Harvard University Press.
51
Ashford M-W, Huet-Vaughn Y. The impact of war on women. . In Levy BS, Sidel VW
(Eds). 1997.
War and public health. Oxford: Oxford University Press, pp186-196.
44
52 Turshen M , Twagiramariya C. (Eds). 1998. What women do in wartime. Gender and
conflict in Africa. London, Zed Books.
53
Stiglmayer A. (Ed). 1994. Mass rape. The war against women in Bosnia-Herzegovina.
Lincoln, University of Nebraska Press.
54 Russbach R. and Fink D. Humanitarian action in current armed conflicts: Opportunities
and obstacles. Medicine and Global Survival, 1 188- 99, 1994.
55
Toole M.J. and Waldman R.J. Prevention of excess mortality in refugee and displaced
populations in developing countries. Journal of the American Medical Association 1990; 263
(24), 3296-302.
56
Centers for Disease Control. Famine-affected, refugee, and displaced populations:
Recommendations for public health issues. Morbidity and Mortality Weekly Report, 41
(RR-13), 1992.
57
Toole M.J. and Waldman R.J. Refugee and displaced persons. War, hunger and public
health. Journal of the American Medical Association 1993 270 (5), 600-5.
58 Centers for Disease Control. Famine-affected, refugee, and displaced populations:
Recommendations for public health issues. Morbidity and Mortality Weekly Report, 41
(RR-13), 1992.
59 Stover E., Keller A.S., Cobey J., et al. The medical and social consequences of land
mines in Cambodia. JAMA, 272 (5), 331- 6, 1994.
60 Physicians for Human Rights (USA). 1992. Hidden enemies. Landmines in Northern
Somalia. Physicians for Human Rights (USA), Boston.
61
Summerfield D. The psychosocial effects of conflict in the Third World. Development in
Practice, 1 159-73, 1991.
62 De Girolamo G. International perspectives on the treatment and prevention of
posttraumatic stress disorder. In International Handbook of Traumatic Stress Syndromes.
45
(Edited by Wilson J.P. and Raphael B.).pp. 935-46.Plenum Press, New York and London,
1993.
63
Quirk G.J. and Casco L. Stress disorders of families of the disappeared: A controlled
study in Honduras. Soc. Sci. Med. 39 (12), 1675- 9, 1994.
64 Summerfield D. The psychosocial effects of conflict in the Third World. Development in
Practice, 1 159- 73, 1991.
Bracken P.J., Giller J.E., and Summerfield D. Psychological responses to war and
atrocity: the limitations of current concepts. Soc. Sci. Med.1995; 40 (8), 1073- 82.
66
67
Gibson K. Children in political violence. Soc. Sci. Med. 1989; 28(7) 659- 68.
Summerfield D. The psychosocial effects of conflict in the Third World. Development in
Practice, 1991; 1 159-73.
68
Bracken P.J., Giller J.E., and Summerfield D. Psychological responses to war and
atrocity: the limitations of current concepts. Soc. Sci. Med. 1995; 40 (8), 1073- 82.
Silove D, Ekblad S, Mollica R. The rights of the severely mentally ill in post-conflict
societies. Lancet 2000; 355: 1548-1549.
70
Ugalde A, Selva-Sutter E, Castillo C, Paz C, Canas S. The health costs of war: can they
be measured? Lessons from El Salvador. British Medical Journal 2000; 321: 169-172.
71 Lee I. and Haines A. Health costs of the Gulf War. Br. Med. J. 303 303- 6, 1991.
72
Zwi AB. Numbering the dead. Counting the casualties of war. In Bradby, H (Ed.).
Defining violence. Understanding the causes and effects of violence. Aidershot, Avebury
Press, 1996: 99-124.
73
Wackers GL, Wennekes CTM. 1992. Violation of medical neutrality. Amsterdam,
Thesis Publishers.
74
Walt G. and Cliff J. The dynamics of health policies in Mozambique 1975-85. Hlth Policy
46
World Health Organization.
99
Barnabas, G.A. Community mobilisation in Tigray. Unpublished. Addis Ababa. 1995.
100 Barnabas GA, Zwi AB. Health policy development in wartime: establishing the Baito
health system in Tigray, Ethiopia. Health Policy and Planning 1997; 12(1): 38-49.
101
El Bushra J, Piza-Lopez E. 1994. Gender, war and food. In . In War & Hunger-
Rethinking international responses to complex emergencies. (Edited by Macrae J. and Zwi
A.).1994 London, Zed Books pp180-193.
102
Sheik M, Gutierrez Ml, Bolton P, Spiegel P, Thieren M, Burnham G. Deaths among
humanitarian workers. British Medical Journal 2000; 321: 166-168.
103 International Federation of Red Cross and Red Crescent Societies. World Disasters
Report 1995. Netherlands, Martinus Nijhoff.
104
Minnear L, Weiss T.G. Humanitarian action in times of war: A handbook for
practitioners. Boulder, Lynne Rienner, 1993.
105 ***i
Banatvala N and Zwi A. Public health and humanitarian interventions: improving the
evidence base. British Medical Journal, 2000.
106
Banatvala N and Zwi A. Public health and humanitarian interventions: improving the
evidence base. British Medical Journal, 2000.
107 Anderson M.B. Do no harm. How aid can support peace - or war. Boulder and London,
Lynne Rienner, 1999.
108 Kumar K (Ed). Rebuilding societies after civil war. London and Boulder, Lynne Reiner,
1997.
109
*** Ambulance of the wrong colour.
110
***ODI Rwandan evalaution
49
G-o rvi h -
Chapter 9
Public Health Challenges - What Lies Ahead
Status:
Draft: To be completed after Regional Consultations
8
Public Health Challenges - What Lies Ahead
The purpose of the final chapter is twofold: 1) to summarize the crosscutting patterns, issues, and
gaps in the field of violence prevention, and 2) to discuss the important public health challenges
in the field and ways to address those challenges in different regions of the world.
Cultural perspectives, expertise, and insight for moving the field forward within and across the
different regions of the world will be summarized from four regional consultations to be held in
the Fall, 2000. Each of the consultations will focus on five public health challenges: 1) how to
improve the collection, availability, and quality of data for documenting the problem of violence,
2) how to improve our understanding of the etiology of violence, particularly the role of social
and cultural factors (poverty, inequality, social norms), 3) how to accelerate the development,
implementation, and evaluation of prevention programs and policy responses, 4) the contributions
and limitations of the public health approach, and 5) how we can advocate for greater
involvement of the health and other sectors in violence prevention.
Appendix A
Statistical Annex
Table la. Population for all member states by sex, age, 1999
Table lb. Basic indicators for all member states
a. Total population size
b. Total population density
c. Infant mortality rate (per 1,000 births)
d. Life expectancy at birth - male
e. Life expectancy at birth - female
f. Average years of education for population aged more than 25 years
Table 1c. Basic indicators associated with the crime rate for all member states
a. % of population living in urban areas
b. Divorce rate
c. GDP/capita adjusted for purchasing power (USS)
d. % of GDP in health expenditure
e. % of GDP in social resource expenditure
f. % of GDP in family child care services
Table 2. Injury mortality by manner of death, sex, and country, 1995-1997
a. Homicide
b. Suicide
c. War
e. Unintentional
f. Undetermined
g. All injuries
Table 3. Mortality caused by intentional injury by sex, age, WHO region, 1998
Table 4. Mortality caused by homicide by sex, age, WHO region, 1998
Table 5. Mortality caused by suicide by sex, age, WHO region, 1998
Table 6. Mortality caused by war by sex, age, WHO region, 1998
Table 7. Mortality caused by intentional injury by sex, age, and country, 1995-1997
Table 8. Mortality caused by homicide by sex, age, and country, 1995-1997
Table 9. Mortality caused by suicide by sex, age, and country, 1995-1997
Table 10. Mortality caused by war by sex, age, and country, 1995-1997
Table 11. DALYs lost due to intentional injury by sex, age, WHO region, 1998
Table 12. DALYs lost due to interpersonal violence by sex, age, WHO region, 1998
Table 13. DALYs lost due to self-inflicted violence by sex, age, WHO region, 1998
Table 14. DALYs lost due to war by sex, age, WHO region, 1998
Table 15. Ten leading causes of mortality and DALYs lost by region, 1998
Table 16. Intentional injury by method, sex, and country, 1995-1997
Table 17. Homicide by method, sex, and country, 1995-1997
Table 18. Suicide by method, sex, and country, 1995-1997
C0<^ H
Appendix B
Contact Information for Study Contributors
<SOM H 1+-3--II
Appendix C
Useful Resources
General
Books/Joumals/Articles
Pediatrician’s Guide to Media Violence
American Medical Association
International Organizations
World Wide Web Sites
Bonn International Centre for Conversion
www.bicc.de
International Center for the Prevention of Crime
www. crime_prevention inti. org
International Action Network on Small Arms
www.iansa.org
National Center for Injury Prevention and Control
www.cdc.gov/ncipc/injweb/websites.htm
Trauma.org
www.trauma.org/trauma.html
United Nations Children’s Fund
http ://www.unicef. org
United Nations Development Fund for Women
http: //www. unifem. undp. org/
United Nations High Commissioner for Refugees
http://www.unhcr.ch
WHO Violence and Injury Prevention Department
www:whodnt/violencejinj\iryjprevention/
WHO Collaborating Centres on Violence and Injury Prevention
www.who.int/violence__injury_prevention/pages/who_collaboratingcentres .
htm
Youth Violence
Books/Joumals/Articles
International Organizations
World Wide Web Sites
Center for the Study and Prevention of Violence
www.colorado.edu/cspv
Youth Violence and Suicide Prevention Team, Division of Violence
Prevention, National Center for Injury Prevention and Control, Centers for
Disease Control and Prevention
www.cdc.gov/ncipcZdvp/yvpt/yvpthtm
Partnerships Against Violence Network
www.pavnet.org
Child Maltreatment
Books/Joumals/Articles
Child Abuse and Neglect: The International Journal,
Elsevier Publications, Editorial Office, 1825 Marion Street, Denver CO,
80218, USA (a monthly professional journal)
Child Maltreatment,
Sage Publications, 2455 Teller Rd, Thousand Oaks, Ca. 91320, USA (a
quarterly professional journal)
The Battered Child, Fifth Edition, edited by Mary Edna Helfer, Ruth
Kempe and Richard Krugman, The University of Chicago Press, Chicago,
II. 60637, USA, 1997 (a basic textbook)
Classic Papers in Child Abuse, edited by Anne Cohn Donnelly and Kim
Oates, Sage Publications, Thousand Oaks, Cal, 91320, USA, 2000. (a
compilation of significant work in the field)
International Organizations
International Society for Prevention of Child Abuse and Neglect
Membership Office
P.O. Box 809343
Chicago II 60680
USA
World Wide Web Sites
International Society for Prevention of Child Abuse and Neglect
ISPCAN.org
The Center for Effective Discipline
http://www. stophitting. com/
Violence Against Women by Intimate Partners
Books/Joumals/Articles
International Organizations
World Wide Web Sites
Women Against Violence Europe
http.7/www. wave-network.org/
The Trust Fund in Support of Actions to Eliminate Violence Against
Women
http://www.unifem.undp.org/trust.htm http://www.un.org/womenwatch/
The Human Rights of Women: A Reference Guide to Official UN
Documents
http://www 1 .umn.edu/humanrts/instree/women/engl-wmn.html
Human Rights Watch: Women's Rights Division
http ://www. hrw.org/about/proj ects/women.html
Network of East-West Women
http://www.neww.org/index.htm
Elder Abuse
Books/Joumals/Articles
The dimension of elder abuse, perspectives for practitioners
(Bennett, Kingston and Penhale edited by Macmillan Press Ltd, London,
1997 ISBN-0-333-62568-4)
Abuse and Neglect of Older Canadians: Strategies for change
Michael MacLean, Editor, Canadian Association of
Gerontology,Toronto, 1995-(ISBN 1 -55077-068-3)
Understanding elder abuse in minority populations
edited by Toshio Tatara, 1999,United States,(ISBN 0-87630-919-8)
Elder Abuse. International and Cross-Cultural Perspectives by Jordan
Kosberg PhD, Juanita Garcia PhD) Editors n The Haworth Press, Inc1995- USA (ISBN 1-56024-711-8)
Journal of Elder Abuse & Neglect - An International Journal published by
Haworth Maltreatment & Trauma Press, Inc. edited by Rosalie S. Wolf and
Susan McMurray Anderson(UBSN: 0894-6566)
Journal on Adult Protection, published by Pavilion Publishing, Brighton
edited by Hilary Brown, Paul Kingston, and Barry Wilson
The Ageing and Development Report 1999
edited by Randel J, German T, and Ewing D
The Status of Widows in 10 Countries: Seclusion and Exclusion, 1999
Edited by Owen M and Young K: Unpublished, OAK Foundation
Margaretowen@compuserve.com or
www. oneworld. org/empo weringwido ws
Institutional abuse: Perspectives across the life course (1999), edited by N.
Stanley, J. Manthorpe & B. Penhale.
International Organizations
Action on Elder Abuse (UK)
Astral House 1268
London rd
SW64ER
England
Tel: 01816792648
Fax: 01816796069
Australian Network for the Prevention of Elder Abuse
Aged Rights Advocacy Service
Abuse Prevention Program
45 Flinders St., Adelaide, SA 5000
Tel. (08) 8232 5377
Fax (08) 8232 5388
Canadian Network for the Prevention of Elder Abuse
c/o Institute for Human Development, Life Course and Aging
University of Toronto
222 College Street, Suite 106
Toronto, ON M5T 3JI
tel. 416 978 1716
416 978 4771
cnpea@hotmail.com
website: www.mun.ca/elderabuse
International Network for the Prevention of Elder Abuse
c/o Institute on Aging
UMass Memorial Health Care
119 Belmont Street
Worcester, MA 01605
USA
tel. 508 334-6166
fax 508 334-6906
wolff@ummhc.org
www.inpeabuse.org
National Center on Elder Abuse
1225 I Street, NW, Suite 725
Washington, DC 20005
saravanis@nasua.org
www. gwj apan. com/NCEA
National Committee for the Prevention of Elder Abuse
c/o Institute on Aging
UMass Memorial Health Care
119 Belmont Street
Worcester, MA 01605
USA
tel. 508 334-6166
fax 508 334-6906
wolfr@ummhc.org
www.preventelderabuse.org
Latin American Committee for the Prevention of Elder Abuse (COMLAT
IAG)
ARENALES 1391 - Sth Floor ”B”
(1061) Buenos Aires - Argentina
Tel/Fax: 54-11-48119590
e-mail: lsdaichman@intramed.net.ar
e-mail: lmachado@attglobal.net
Sexual Violence
Books/Joumals/Articles
International Organizations
World Wide Web Sites
The Global Alliance Against Traffic in Women
http ://www. inet. co. th/org/gaatw/
Coalition Against Trafficking in Women
http://www.mi.edu/artsci/vvinsdiughes/catw/
Research, Action and Information Network for the Bodily Integrity of
Women
http://www.rainbo. org/
Self-directed Violence
Books/Joumals/Articles
Jacobs DG (1999): The Harvard Medical School Guide to Suicide
Assessment and Intervention. Jossey-Bass Publisher: San Francisco
De Leo D, Schmidtke A, Diekstra RFW (1998) : Suicide Prevention: A
Holistic Approach. Kluwer Academic Publishers:
Dordrecht/Boston/London
International Organizations
International Association for Suicide Prevention,
c/o General Secretary, Mrs. Vanda Scott, Le Barade’, 32330 Gondrin,
France,
tel/fax +33 562 29 19 47, Email: iaspl960@aol.com
International Academy for Suicide Research,
c/o President, Prof. Armin Schmidtke, Dept. Psychiatry, University of
Wuerzburg, Germany, tel +49 931 201 7667, fax +49 931 201 7669,
Email: clips-psychiatry@mail.uni-wuerzburg.ge
World Wide Web Sites
Australian Institute for Suicide Research and Prevention
http://www.gu.edu.au/school/psy/aisrap/
Suicide Information and Education Centre
www.siec.ca
Collective Violence
Books/Joumals/Articles
Anderson M.B. Do no harm. How aid can support peace - or war. Boulder
and London, Lynne Rienner, 1999.
Carnegie Commission on Preventing Deadly Conflict. 1997. Preventing
deadly conflict. Final report. New York, Carnegie Corporation (available
also at http://www.ccpdc.org
Leaning J, Briggs SM, Chen LC. (Eds) 1999. Humanitarian crises: the
medical and public health response. Cambridge, MA, Harvard University
Press
Levy BS, Sidel VW (Eds). 1997. War and public health. Oxford: Oxford
University Press.
Medecins Sans Frontieres. 1997. Refugee health,
mergency situations. London, Macmillan.
An approach to
Perrin P.
1996. Handbook on war and public health.
International Committee of the Red Cross.
Geneva,
Weiss TG, Collins C. 1996. Humanitarian challenges and interventions.
World politics and the dilemmas of help. Boulder, Westview Press, 1996
International Organizations
World Wide Web Sites
UN Office for the Coordination of Humanitarian Affairs
http://www.reliefweb. int/oclia_ol/
Relief Web
http://www.reliefweb.int/w/rwb.nsf
AlertNet
http://www.alertnet.org
Refugees Daily
http://www.unhcr.cli/news/media/daily.htm
UNHCR News
http ://www.unlicr. ch/news/ne wswire/news wire. htm
Security Watch
http://www.isn.ethz.ch/infoservice/index.cfm?service=cwn&menu=l
Updates on World Conflicts
http://wwv.cartercenter.org/UPDATES/updates.html
Weekly News
http://www.idpproject.org/weekly_news.htm
Appendix D
Sample Testimonies
Nepal (Child Prostitution): “Maya, an attractive young Nepalese girl, came from a village so poor that
often com was the only food available. When she was only 13, her cousin said he would take her to
Kathmandu and find her work. That was 15 years ago. Her cousin took her to a room in Bombay which
she was assured was Kathmandu. There, he sold her. When she refused to perform on her first day in the
brothel, she was beaten and red chili powder put up her nose. Maya surrendered, and lived a life full of
mental and physical torture. She had already been in Bombay for 13 years when she tested HIV positive
and was sent back to Nepal. Her story is not much different from those of hundreds of Nepali girls, some
dying in villages in Nepal after being sent back with AIDS from the brothels of Bombay.”
USA (Youth Violence): “Twenty years ago what kids worried about was getting bad grades. But now,
kids are worried about getting killed.. .I’m scared too and I don’t want to die. I have a whole lot of life to
go and make my own goals and be what I want to be.” (Miguel, a 12-year-old resident of California)
Nicaragua (Child Soldiers): “I joined the guerrillas when I was about fifteen years old. This was not
exceptionally young. Many of the fighters were youths. It gave us a perfect cover for our activities,
because youths are always in the streets, playing or whatever. We were inconspicuous...We were
organized in small clandestine units and had as little contact as possible with other units, to minimize
what we could tell if we were captured. We would prepare caches of weapons, make molotov
cocktails, lay ambushes and carry out strategic actions, such as burning buses or executing informants.
It wasn’t easy. The dictator’s guards caught eight of the twelve members of my unit and killed seven
of them. There were many sympathizers of the dictatorship living in the neighborhood and we always
had to be careful about informants. But the guards could also hit us hard without the help of
informants. One time, a patrol chased a fellow fighter, named Jorge, and me after we’d managed to
firebomb one of their buses. They followed us right into the neighborhood and caught Jorge. They shot
him in the stomach, again and again, repeatedly, and left him to die in the streets, with his entrails
falling out. People came out of their houses to see. Including a kid called Noel Gutierrez, who lived on
my alleyway. I think he was eleven or twelve. The guards grabbed him and started beating him, asking
him who in the neighborhood were guerrilla fighters. Noel was screaming that he knew nothing, that
he was just a kid. So they began pulling his fingernails out, one by one. But he didn’t tell them
anything, even though he knew who I was. When they finished torturing him, they beat him senseless,
put a grenade down his pants and walked away...They were brutal, you know, barbarians,
savages...Noel was a good kid - we named the neighborhood school after him...”
USA (Gang Violence): “11-year old Tony lives in a large, poor, densely populated housing project in
Chicago. When he is asked to draw a picture of the neighborhood, he draws a 15-story red brick building,
commenting that ‘it’s one of the gun towers.’ He proceeds to draw a large gun coming out of the side
window and explains; ‘usually the gangs shoot out of the side windows, but sometimes they shoot out of
the front windows too.’ Then he draws a car with a gun coming out of the window as well and says; 'that’s
when they do a drive-by and shoot you.’ Stick figures of rival gang members appear and Tony tells
‘usually they just throw things - bottles and bricks. But if you try to come on their territory, they’ll try to
kill you. One day a bullet came through the window of our apartment and almost hit my little sister.’ He
also talks about his cousin, who was shot and killed in one of the white buildings, because he was
mistaken for a gang member from one of the red buildings. ‘I hear shooting every night and see dead
bodies on the street. I see and hear all these things all the time, on the way to school and the way from
school.’”
1
Panama (Intimate Partner Violence): "I married my husband at a very early age. My dad saw me
talking to him and drew his conclusions. He threw me out of the house and forced me to live with the boy.
It was his mistake. He had always been bad; he hit my mother a lot. Although, I had just entered
university, I dropped out. At first, my husband and I lived at his parents’ house. The first four years were
relatively peaceful. He did not abuse me physically, just with words. As time went by he started hitting
me out ofjealousy and he would not allow me to get out of the house. But I had to work, because he could
not be bothered with it. With my work, we earned quite some money and we bought cows and a car. He
would deny my contribution, though, and one day he decided without my consent to sell the cows and car.
He once hit me when I was three months pregnant. He hit me with a metal chain in the stomach and pelvis.
I aborted and got an infection. I had to stay hospitalized for more than a month and because of the
incident, they had to remove my uterus. During that time he only visited me once. He had sold the
television my brother had given me and stopped by to give me some money. He still hits me - he always
hits me - in the face. But it’s the insults that hurt the most. I tell you, it’s worse than if he had stabbed me
in the back with a knife.”
Peru (Marital Rape):“... When my husband raped me again, I had to go to the health center to receive
medical attention. They sent me to the police, who asked me to take a medical exam to prove I was
raped. Medical exams cost money and I did not have money so I went back to the center. A nurse there
helped me, gave me money to take the exam. I took it, went back to the police and they called on my
husband. He denied that I was his wife and the policemen believed him. They told me that, if I wanted
to press charges against him, I would have to find people who could testify that we were married. I
could not proceed the case, because getting people to testify costs money. And I don’t have money...”
Mozambique (Elder Abuse): “It’s a sad, sad story. My physical pain is enormous, because I was burnt
alive. Still, the pain I have inside is worse: No one takes care of me and no one visits me. My son even
prefers that I died...It happened a long time ago. My daughter-in-law lost her baby in the sixth month of
her pregnancy and she blamed me for it. She accused me of witchcraft. From then on the injustices started:
I was not allowed to cook anymore, my grandson was prohibited from seeing me and eventually I was
forced to move out of my son’s house. I had to live into a pahlota, a straw hut which had no facilities
whatsoever. One day my grandson had an accident while he was walking home from school. My son and
his wife accused me from causing it. I told them I was innocent, but they would not listen. The next day,
my daughter-in-law threatened to kill me. She said that as long as I was alive, the family would have
misfortune. I was so confused that I didn’t notice they were making concrete plans. I only realized the
severity of their hatred, when I woke up in the night and saw my pahlota on fire. I was suffocating from
the smoke and I couldn’t get out of the flames. I started to scream, but no one would listen. Then, I must
have fainted. When I recovered I found myself in the hospital, with my neighbors. Their 12-year old son
had woken up from a toothache that night when he saw the flames and heard me screaming. He called his
parents and they saved me.”
Nicaragua (Violence against the Elderly): “I was walking in the Don Camilo neighborhood, looking
for someone to assault. Don Camilo is about two or three neighborhoods away from ours, ‘cause we never
attack our own people. Our own people help us, they hide us when the Police comes. And we protect them
from other gangs and all. They’re like our friends, our family. Anyway, my parents had kept me locked
up for three weeks because of all the Police activity. I felt really destructive and just wanted to take it out
on somebody... ! also needed a drink, because I hadn’t had one in three weeks, and so needed some
money to buy it. I was walking around with my gun and thought ‘hey, why not do a house?’ I wanted to
find one with, say a couple of old-timers. It would be the easiest job I ever did! So I hung around a bit,
looking into houses, you know, checking them out. After a while I figured out that in one house there was
2
nobody except this old woman who would probably die of a heart attack when she saw me. So I thought,
‘great, let’s do it!’
Man, I have never heard anybody scream so loudly! But worse, she didn’t stop. I tried to shut her up by
beating her up, but she just kept on screaming and screaming, louder and louder! I figured I’d best get the
hell out of there. But her screaming brought the whole neighborhood out and I about twenty guys chased
me, all shouting that they were going to kill me.”
India (Suicide): “X drank a lot, but not too much. He was a hard working man who never quarreled with
anyone. He and his wife had been married for more than 25 years and their marriage had brought five
children; three daughters and two sons. The boys were still in school, but the two youngest girls were
already working to support the family. His eldest daughter had been married off two years before and her
dowry had left the family in great debt. When his second daughter reached the marriageable age, X went
out to find her a suitable husband. Although the family could not afford another dowry, he set the wedding
date. He was convinced that he could somehow borrow money just as he had done with the first marriage.
It was a big disappointment to X when his friends and relatives closed the door on him when he asked
them for a contribution. He still owed them money from the last dowry. As the wedding date approached,
his wife and children started to bother him. They told him to call the wedding off if he was not able to
come up with the money for the dowry. X became desperate and started to yell at them, as they pushed
him more and more. Too stubborn to call the marriage off, he tried every place he could think of to borrow
money. It was no use. One day the disillusioned father was sitting outside his house. He asked his
neighbors for advice and they told him not to worry. God would help him one way or another. The next
day he did not go to work. In another quarrel with his wife, he wondered why he had been given daughters
at all. His wife, fed up with his complaints, went into the kitchen. Making use of the time alone, X took
poison and laid himself down. His wife continued her housework, thinking he was fast asleep from
exhaustion. Later, when she went to wake him, she discovered he was dead.”
Belarus (Attempted Suicide): “I am married and have two adult sons, both students. I used to work
as an engineer in the same plant as my husband. We were making a good living but our salary
decreased extremely last year and became insufficient. That was the only reason that my husband
started his own business. He founded a small firm and asked me to take are of the bookkeeping. Thus,
I was obliged to do a lot of extra work as a bookkeeper and a housekeeper after a whole day at the
plant. I did not have enough experience to do the bookkeeping properly. I had to rewrite my papers
over and over again, which caused anger and discontent with my husband. We worked hard, but it
didn’t pay off. Our company stayed small and we didn’t grow rich. My husband and I started to
quarrel, something that had never happened before. I was constantly nervous and tensed and did not
get any rest. After a year of business disappointments, I was on the edge of a nervous breakdown. One
day, I had again made mistakes in the bookkeeping and my husband wanted me to correct and rewrite
the documents. I can hardly recall what I felt at that moment, but I got hold of a knife and before my
husband’s eyes, I cut myself in my abdomen. Even now I don’t realize that I could have died. I spent a
long time in a mental hospital and now I am doing well.”
Rwanda (Collective Violence): “My husband and I come from different ethnic backgrounds. After we
got married, I moved to live in his area. When the atrocities started, people from our community stoned
me, raped me and mutilated certain intimate parts of my body. I had to flee and live with my father’s
family - my mother’s family had perished during the killings. I am forty years old now and not only do
I have scars that will never go away, the genocide has also destroyed my marital and family relations. My
husband has recently asked me to return home. At first I refused, because I vividly remember the
hostilities and because the situation remains oppressed. But, I have to think of the future of my two-yearold daughter. It is only because of her that I have decided to go back.”
3
Vietnam (Health Consequences-Collective Violence): “I got married when I was twenty. My husband
is a farmer and I love him very much, because he is nice and intelligent. We live in the Tran Yen district,
Yen Bai province. I have been pregnant six times, but we only have one child, a 2-year-old girl. The first
time I got pregnant I had an abortion in the fourth month. The second time I gave birth to a baby whose
face and little hands were severely malformed. It died after three days. When my third pregnancy was
again premature and I again lost the baby, people in our community began to suspect that our house was
haunted by an evil spirit. My husband could no longer keep quiet; he became very angry and burnt our
house. We moved to the Bao Dap commune where I became pregnant for the fourth time. My husband
and I were hoping that this time, now that we had moved into this new community, into this new house,
we would receive a healthy child. When I again had a miscarriage, I went to see a doctor. They could not
find anything wrong. I got pregnant again and this time I gave birth to a girl with a cleft palate. We started
hoping again, but my last pregnancy ended in the birth of a twin-malformation that died just after the
delivery. I don’t know why this happens to us. I am healthy and I don’t work hard. In fact, when I’m
pregnant my husband often asks me to stay at home and rest. He is also in good health. Although, at first,
when he just came back from the front, he did not look well. During the war with the Americans, he had
fought in the South. He saw the American airplanes flying over many times, spraying Agent Orange. I
don’t think I will ever give birth again. I’m afraid this is my fate.”
Afghanistan (Collective Violence):“Hundreds of men, women and children had come a long way to
attend a wedding ceremony in Kandahar City. I went along with my mother and two sisters, but was on
a different bus because men and women traveled separately. Near to the place where the wedding was
going to take place, their bus, which was full of women and children, drove over an anti-tank mine. It was
a terrible moment. The explosion spread sorrow and screams throughout the whole party. In all 45 women
and children were killed from the explosion and 35 more were severely wounded. My sisters and my
mother died on that bus.”
4
Position: 980 (5 views)