COMMUNITY FINANCING OF PRIMARY HEALTH CARE : THE PRICOR EXPERIENCE
Item
- Title
- COMMUNITY FINANCING OF PRIMARY HEALTH CARE : THE PRICOR EXPERIENCE
- extracted text
-
4.' 1
is L
■■
|
1
IL
|L
COMMUNITY FINANCING OF
PRIMARY HEALTH CARE:
THE PRI£OR EXPERIENCE
s,
it
A Comparative Analysis
L i
14.
March 1987
fti
Bi
L
L
L
I
i1
' . I
‘J
11
f
1
u
Primary Health Care Operations Research
Center for Hyman Services
5530 Wisconsin Avenue
Chevy Chase, Maryland 20815
■■ T
ACKNOWLEDGEMENTS
PRICOR acknowledges the efforts of its researchers with
respect and appreciation. They worked long and hard in
difficult settings and contributed significantly to our
knowledge of PHC design and financing. Researchers whose work
is reported here include:
Benin
Ms. Elizabeth Coit, Universalist Service Committee,
Boston, Massachusetts
Dr. Eusebe Alihonou, Cotonou, Benin
Bolivia
Dr. Rene Gonzalez, Institute de Investigaciones
Medico Sociales, Cochabamba, Bolivia
Dr. Martin Miller, Fundacion Integral de Desarrollo,
Casilla, Bolivia
/
Brazil
Dr. Timothy Baker, The John Hopkins University,
Baltimore, Maryland
Ms. Karen Johnson Lassner, Centro de Pesquisas de
Assistencia Integrada a Mulher e a
Crianga, Rio de Janeiro, Brazil
Dr. Marilyn K. Nations, University of Virginia
Medical School, Charlottesville, Virginia
Dr. Maria Auxiliadora de Souza, Federal University
of Ceara, Brazil
Dominica
Mr. Peter Cross, Management Sciences for Health,
Boston, Massachusetts
Dr. Desmond McIntyre, Ministry of Health, Roseau,
Dominica
Dominican Republic
Dr. Alan Udall, Clapp and Mayne, Inc., San Juan,
Puerto Rico
Dr. Jose A. Herrero, Secretariat of Health and
Social Assistance, Santo Domingo,
Dominican Republic
Haiti
Dr. Antoine Augustin, Alliance pour I’Enfance et
le Developpement Conununautaire,
Port-au-Prince, Haiti
Honduras
Dr. Frederick Hartman, Management Sciences for
Health, Boston, Massachusetts
India
Dr. Henry Elkins, Management Sciences for Health,
Boston, Massachusetts
Jamaica
Ms. Patricia Desai, University of the West Indies,
Kingston, Jamaica
Mr. Bobby Zachariah, Price Waterhouse Associates,
Kingston, Jamaica
Korea
Dr. Yeo Shin Hong, Seoul National University
Seoul, Korea
Liberia
Dr. Andrew Cole, Christian Health Association of
Liberia, Monrovia, Liberia
Mali
Dr. Mamadou Traore, Ministere de la Sante Publique
et des Affaires Sociales, Bamako, Mali
Mexico
Ing. Luis de la Macorra, Promotora de Planificacion
Familiar, Queretaro, Mexico
Philippines
Dr. Trinidad S. Osteria. Institute of Southeast
Asian Studies, Pasir Pajang, Singapore
Professor Ida M. Siason, University of the
Philippines in the Visayas, Iloilo
City, Philippines
Senegal
Dr. Clive Gray, Harvard Institute of International
Development , Cambridge, Massachusetts
Somalia
Dr. Norman Lane, Medical Services Consultants, Inc. ,
Arlington, Virginia
Swaziland
Ms. Laurie Dunn, University of Swaziland, Malkerns,
Swaziland
Ms. B. Dlamini, Vilakati, Ministry of Health,
Mbabane, Swaziland
Thailand
Ms. Orathip Tanskul, National Economic and Social
Development Board, Bangkok, Thailand
Zaire
Dr. Lusamba Dikassa, The Basic Rural Health
Services Project, Kinshasa, Zaire
Dr. Frank, Baer, The Basic Rural Health
Services Project, Kinshasa, Zaire
Many pioneered operations research in specific topics and
countries and generated additional operations research that is
only now occurring.
This analysis is based largely on investigator reports
and the personal knowledge of PRICOR study monitors, Its
authors are Wayne Stinson, Marty Pipp, La Rue Seims, and Pat
Sayer. Time and distance constraints precluded review by the
majority of principal investigators. We apologize for any
errors or misinterpretations of their studies.
COMMUNITY FINANCING OF PRIMARY HEALTH CARE:
THE PRICOR EXPERIENCE
A Comparative Analysis
TABLE OF CONTENTS
PAGE
!
1.
Introduction and Overview
1
2.
Influencing the Political and Policy
Environment........................
10
Demand for Health Care:
Interests and Resources
15
3.
Community
4.
PHC Costs and Costing Methodology
27
5.
Problem Identification and Objective
Setting............................
32
Matching Needs with Resources:
Selection and Design Decisions
37
6.
Scheme
7.
Mobilizing Community Participation
45
8.
Resource Generating Methods
57
9.
Inputs and Activities Financed
75
10.
Evaluation
88
BIBLIOGRAPHY
99
APPENDICES
1.
Brief summaries of community financing
studies
.
100
2.
Case studies of costsharing schemes
145
3.
Case studies of credit schemes that
promoted preventive health activities
149
CHARTS
PAGE
1.
2.
3.
4.
5.
6.
7.
8.
PRICOR-Supported Community Financing
Studies: Problems, Objectives, Proposed
Solutions, and Results
.......
4
Costing Methodologies Used in PRICORSupported Studies..................
29
Decisionmaking in Community Financing
Design Studies.....................
38
Cross Tabulation of Sources and Uses of
Community Financing..................
42
Community Participation in PRICORSupported Financing Studies.....
47
PRICOR-Supported Studies that Documented,
or Helped to Establish, Drug Sales Systems
58
PRICOR-Supported Studies that Documented, or
Helped to Establish, Fee for Service Systems
65
PRICOR-Supported Studies that Documented, or
Helped to Establish, Cost-Sharing Schemes
73
9.
PRICOR-Supported Studies that Documented, or
Helped to Establish, Methods of Compensating CHWS..76
10.
PRICOR-Supported Studies that Documented, or
Helped to Establish, Methods of Financing
Preventive/Promotive Activities...........
85
BOXES
1.
Matching Needs with Resources in Bolivia
44
2.
Sale of Shares for Capitalization in Thailand
61
3.
Factors Considered in Setting Price Levels
in Benin...............................
69
TABLES
PAGE
1.
2.
3.
Zaire:
Care .
Health Center Visits for Curative
91
Brazil/Lassner: Proportion of Total and
Direct PHC Costs Covered by Revenue Generated
from Community Financing Strategies........
95
Zaire: Percentage of Health Center Operating
Costs Covered.............................
96
CHAPTER 1
INTRODUCTION
Research and action projects over the past decade have
pioneered new methods of improving people’s health, using
simple techniques, mainly local manpower, and a few basic
drugs, vaccines and supplies from national or international
sources. Relatively low cost interventions - oral rehydration
therapy, immunization, growth monitoring, breast feeding,
vitamin A, among others - have been widely promoted as ways to
increase chances for children’s survival in the first
difficult years. There has been encouraging progress in a
number of countries. Enough progress has been made to give
health care planners hope that at least some "health” may be
achievable for major parts of the world’s population by the
year 2000 - if only ways can be found to pay for it.
In the early years of the primary health care movement,
especially at and after the Alma Ata conference, enthusiasts
in many countries promoted the low cost virtues of community
based preventive and promotive activities and suggested that
much of the cost of these activities could be paid for by the
community residents themselves. This argument had some
validity when set against the weak technical rationale for
curative medical care, hospitals, and European or American
professional standards. Primary health care is low cost and
relatively affordable in these terms. The problem is that
neither governments nor peoples value it sufficiently to
overcome the forces favoring curative care or to expend the
necessary resources. The problem of primary care financing
remains acute and tops the agenda of such major organizations
as the World Health Organization and the Agency for
International Development.
SUSTAINABILITY REQUIRES MORE THAN RESOURCE GENERATION
Discussions have distinguished four basic finance-related
tactics. The first, and often most promoted, element is new
resource generation, emphasizing user fees and, now, employer
financing. The subject of community-managed financing, as
distinct from user fees managed by the government, was widely
discussed in the early 1980s but seems to be less emphasized
currently. The second tactic is cost containment, seen both
as a way to make basic services more affordable and as a way
to save money on politically popular but expensive hospitals.
The third tactic is resource reallocation, particularly from
hospitals to preventive care and from ineffective
interventions to effective ones. A final but strongly
promoted tactic is reorganization of health services,
particularly favoring the private sector. Recent AID Health
Financing Guidelines support all four of these tactical
approaches.
1
Recognizing financing’s urgency, the USAID-financed
Primary Health Care Operations Research Project (PRICOR)
selected community financing as one of four subject areas in
which it would work between 1981 and 1986. Sixteen studies
were funded either to document or to develop community/user
financing activities. Six of these studies were in Africa,
three in Asia, and seven in the Latin American/Caribbean
region. Six additional studies produced significant data on
health care expenditures, costs, and productivity. In all
cases, PRICOR supported American and LDC researchers through
funding and technical guidance but did not do the research
itself.
Community financing has been broadly defined in the
literature to include user payments of all kinds, whether
managed directly (1) by community groups/other non-health
system personnel, or (2) by health system/government
personnel. Among PRICOR-supported studies, both Dominica and
Liberia/Cole I are considered community financing studies,
even though the Dominica system was managed by the Central
Medical Stores and district health officers, while the
Liberian one was managed by a village health committee, Both
systems are discussed here because users of PHC goods and
services helped pay for them, but study results showed that
the distinction between professionally managed activities and
community managed ones is just as important for financing as
it is for primary health care.
Though all studies reported here related to costs and
financing, they varied greatly in their venues, objectives,
and outcomes. Locations varied from the sometimes violent
urban slums of Rio de Janeiro to the relatively traditional
rural communities of Zaire, Liberia, and Bolivia. Some
studies tried to guide design of specific community financing
schemes, while others were meant to influence the policy
environment. Some studies were prospective, that is, intended
to study and influence future conditions, while others, like
traditional evaluation, were mainly intended to analyze
current conditions and their causes. Some studies were
national in scope, while others encompassed only a few
communities. Some studied government ministries, others
private groups. Brief summaries of these studies are given in
Chart 1 and Appendix A. Further detail on specific studies
may be obtained from PRICOR.
OPERATIONS RESEARCH IS A MANAGEMENT DESIGN PROCESS
PRICOR’s objective in these studies was to resolve
specific operational problems, not to produce generalizable
knowledge.
Often the key factors affecting problem
resolution were decisionmaker perceptions and preferences,
rather than objectively verifiable conditions. The least
workable problem solutions were ones developed by traditional
2
researchers using abstract models or working in isolation from
implementers.
Operations research of conmmunity financing is an
economics question, but also a sociological/anthropological
one (because of the importance of community organizations) and
a management one. Both the service delivery mechanism and its
financing have to be studied because people will only pay for
functioning, reasonably high quality activities. Design of
community financing schemes in PRICOR-supported studies often
required strengthening and partial redesign of the PHC system
itself to increase users’ ability and willingness to pay.
THIS PAPER APPLIES A COMPARATIVE FRAMEWORK
This report describes and analyzes how PRICOR-supported
decisionmakers investigated and resolved specific operational
problems in PHC and financing. It makes extensive use of
charts to present study activities and results within a
comparative framework. It attempts to draw patterns where
they are apparent and to at least describe processes and
results within a common framework when they are not. It is
identified as a comparative analysis even though studies were
designed to produce locally valid solutions, not universally
valid knowledge. PRICOR learned much of general interest
about how to resolve community financing problems, but results
should be applied cautiously because they were never intended
to be fully generalizable.
3
CHART 1: PRICOR-SUPPORTED COMMUNITY FINANCING STUDIES:
PROBLEMS, OBJECTIVES, PROPOSED SOLUTIONS, AND RESULTS
STUDY
SUBJECTS
ADDRESSED PROBLEM
OBJECTIVE
PROPOSED
SOLUTION
RESULTS
Schemes were instituted
Identify, implement, and Pricing and revenue
management system; fees in study villages and
monitor conmunity
to be charged by episode cover approximately one
financing methods for
third of direct and in
PHC
direct PHC costs
Benin/
Coit
RG
Need to develop CF
system for new PHC
services to complement
government and donor
financing
Bolivia/
Gonzalez
RG» CC
Very high CHW attrition Reduce high attritition
rate in Cochabamba, thus rate and thus increase
decrease in PHC coverage PHC coverage
in rural areas
Bolivia/
Hiller
RG, PS
Inadequate PHC services
and financing
To develop the PHC
activities of 12 agri
cultural cooperatives
Brazi1/
Baker
RG
See Comments
Analyze FSESP experience See Comments; job
with CHWs and community security, stable
financing of water supply salaries, system support
found to contribute to
CHW stability and
effectiveness
Brazi 1/
Lassner
RG, PS
Overdependence on exter Develop community cost- Variety of community and Solutions partly implenal resources for PHC and sharing mechanisms, while user financing mechanisms men ted; most successful
other social services
increasing service accescommunities covered 201
sibility and use; develop
of direct recurrent
replicable models for
costs
other parts of Brazil
COMMENTS
Non-residents charged
4 times the resident fee-an important revenue
generating technique
New, lower paid, CHW
cadre; periodic collec
tion of potatoes from
community residents to
finance CHW salaries and
medications
Six promotoras trained
and worked without pay
until first potato
harvest; then
accepted inkind payment
Due to Bolivian hyperinfla
tion, revenues from sale of
in-kind payments
changed into dollars.
See Comments
See Comments
Cooperatives decided not
to develop PHC services
CHW information system
found to be weak; revisions based on study
results implemented
This was a retrospective
study not directly intended
for problem analysis and
solution development.
Study conducted in urban
slums
CHART 1: PRICOR-SUPPORTED COMMUNITY FINANCING STUDIES:
PROBLEMS, OBJECTIVES, PROPOSED SOLUTIONS, AND RESULTS
STUDY
SUBJECTS
ADDRESSED PROBLEM
OBJECTIVE
PROPOSED
SOLUTION
RESULTS
COMMENTS
Healer participation in
solution development an im
portant factor in success;
study also produced cost
data
Brazil/
Nations
PS
Use of ORS is low, even
though caretaker aware
ness and knowledge are
high.
Train and incorporate
traditional healers into
the ORT delivery
system
Training curricula and
method for generating
interest about ORT among
traditional healers
Solution successfully
implemented, will be
partially replicated
in major regional
project
Dominica/
Cross,
McIntyre
CC
Government budget unable
to meet costs of all
drugs and supplies pre
scribed at NOH health
facilities
Establish capability of
Central Medical Stores
to successfully operate
national revolving drug
fund; implement and
evaluate consumer pay
ments for drugs and
supplies
First, improve central
and district pharma
ceutical management and
information system;
then, institute consumer
drug payments
Solution partially imple- Dominica model being applied
mented, but user charges in other Eastern Caribbean
not yet instituted due
countries under major AID
contract
to feared political
reaction. Fund was capi
talized; district health
staff participate, using
their MOH budgetary
allocations; central drug
procurement and distri
bution system has been
strengthened.
Dominican
Republic/
Udall
RG. PS
NOH wanted to launch
Design pricing structure Two-tiered pricing struc- See Comments
nationwide ORT campaign, and distribution system ture and an inventory and
distribution management
which ensures that ORS
but did not know how
is available and afford system
many ORS packets would
be required or bow to
able for poor and that
ensure distribution and pharmacists make suffi
availability for child cient profit from ORS
sales
ren under 5.
Haiti/
Augustin
(AEDC)
RG
Uncertainty about how to Develop ways to motivate Adaptation of traditional First groups of mothers Scheme only recently
learned interventions and implemented
tap local resources for CHWs to teach mothers to rotating credit clubs
participated
in schemes;
sustaining CTU preventive understand and use pre for mothers who demon
portion of funds gener
services
ventive services; estab strate competence in
ated used to pay CHW
lish supportive CF
child survival
salaries
mechanisms
interventions
(ji
Solutions not implemented
due to lack of support from
MOH and poor communication
between researchers and
MOH
CHART 1: PRICOR-SUPPORTED COMMUNITY FINANCING STUDIES:
PROBLEMS, OBJECTIVES, PROPOSED SOLUTIONS, AND RESULTS
STUDY
SUB3ECTS
ADDRESSED PROBLEM
OBJECTIVE
PROPOSED
SOLUTION
Identify alternative
Service fees, drug
financing mechanisms
charges, labor contribased on kinds of health butions, community
services for which people revolving funds
are able and willing
to pay, within equity
and quality of care
constraints
Honduras/
Hartman
RG, RR
Severe econonic crisis
of 1980 has limited
government's ability to
provide Health services.
India/
Elkins
RG, PS
Limited resources require Study financial, organ See Comments
voluntary agencies in
izational, and managerial
India to find most effec- aspects of 8 health
tive ways of financing, cooperatives; describe
various solutions that
organizing, and manag
cooperatives had already
ing conraunity-based
developed
PHC programs.
CC» RR
Low PHC team productivity Improve health center
in health centers due to: productivity
(a) inappropriate man
power allocation; (b)
health center schedules
not based on demand for
services
cr>
Jamaica/
Desai,
Zachariah
RESULTS
COMMENTS
PHC recommendations not
implemented due to
legal, political, and
attitudinal constraints;
charges authorized in
hospitals, however
Study contributed to
improved policy environment
for further OR.
National workshop held
This was a retrospective
to discuss findings with study not intended for
other agencies and vol solution development.
untary programs; arrange
ments made to institu
tionalize information
sharing. Policy deci
sions taken by Institute
of Rural Management
(Anand) to include train
ing in management of
rural health programs in
curriculum, place interns
in projects.
Microcomputer model to
Application being tested Model has shown that
optimize productivity by in Cornwall County
personnel costs can be re
reallocating clinical
duced, while at the same
personnel, restructuring
time actually increasing
clinics, and relocating
number of services delivered
health centers
and population covered with
essential services.
s
CHART 1: PRICOR-SUPPORTED COMMUNITY FINANCING STUDIES:
PROBLEMS. OBJECTIVES. PROPOSED SOLUTIONS, AND RESULTS
STUDY
SUBJECTS
ADDRESSED PROBLEM
OBJECTIVE
PROPOSED
SOLUTION
RESULTS
lorea/
tong
CC
Leaders trained but
Strengthen community
Training of CHWs and
Inadequate preventive
community interest still
understanding and support community leaders in
and promotive health
for preventive and promo- prevention and health
low.
activities in rural
areas; community health Live health activities; education; health educa
tion presentations at
practitioners are clinic- analyze cost
based and provide mainly effectiveness of various community meetings
curative care.
strategies for involving
non-health community
organizations in PHC
■Liberia/
Dale I
RG, PS
Revolving drug funds;
Health services and drugs Establish and finance
several methods for
relatively inaccessible community-based health
services and drug supply supporting CHWs
in rural communities
sources
Kali/
Traore
RG. RR
Overdependence on exter- Identify and implement
nal resources for PHC
community financing
methods for PHC
financing
Mexico/
de la
Macorra
PS
ORS, though widely
Determine if ORS tablets Developing ORS packets
would be acceptable and to be dissolved in
available in Mexican
marketable to public
8 oz. water
pharmacies, is sold
mainly in more expensive
liquid forms.
COMMENTS
Cost data available
Revolving drug funds
Solutions adopted by several
nearby communities not
implemented in study
communities and nearby;
formally involved in study.
CHW compensation schemes
not implemented
Villages to finance cer- Development council of
tain costs associated
1 village agreed to
with delivery of curative finance 1986 budget of
services and markup on
village health center;
drugs; government to
budget includes many
finance preventive and
financing strategies
supervisory activities
developed in study.
Pharmacists surveyed
preferred ORC packets
over tablets.
A National Policy Study
Based on findings, PROFAM
chose to produce packets of
ORS granules rather than
tablets.
CHART 1: PRICOR-SUPPORTED COMMUNITY FINANCING STUDIES:
PROBLEMS, OBJECTIVES, PROPOSED SOLUTIONS, AND RESULTS
STUDY
SUBJECTS
ADDRESSED PROBLEM
OBJECTIVE
PROPOSED
SOLUTION
Community-managed and
-financed revolving drug
funds; emergency hospi
talization fund; "lead
mothers" scheme
Philippines/ RG
Oster i a
Economic conditions pre Mobilize comnunity
resources for PHC,
clude scale of public
sector financial expan including prevention
sion needed to provide
health services to rural
areas
Senegal/
Gray
RG, RR
Overdependence on
Recommend measures to
Recommendations for
external sources for PHC wean Sine Saloum project future OR in areas of
financing
from AID support; examine supervision, CHW remuneration, drug resupply
possible (particularly
community) sources of
funding for PHC recurrent
costs
Somalia/
Lane
PS
Lack of adequate supplies Develop strategies to
of basic drugs in rural
improve PHC drug supply
areas
in rural areas
Swazi land/
Dunn,
Dlamini
RG
HOH cannot afford
Improve scope and quality Refresher courses
to pay rural health
of RHM services; identify (including home-based
most appropriate way of ORT) for RHMs, so their
motivators (RHM)
more than very
sustaining them through services are better ap
preciated by community
small monthly stipend;
conmunity financing
community unwilling to
and generate in-kind
support RHM preventive
(food or labor) payments
for RHM
services; RHM task speci
fication ambiguous.
oo
RESULTS
COMMENTS
Five community pharmacies Community involvement
still functioning in the considered essential for
black 9 months after
project success
study completion; little
support for "lead mother"
preventive activities
USAID/Dakar has requested
technical assistance in
developing supervision OR
agenda.
Use private sector as
Alternative solutions
well as public to deliver developed and placed be
fore decisionmakers,
drugs; prepare written
"Guide to PHC Drug Use" asking them to choose and
for villagers; review PHC implement
drug list and limit items
Chief assigned field to MOH area superior of RHMs
RHM & community to pre trying to expand approach
to second chieftianship
pare land for planting.
Community did so, but
due to external circum
stances, too late for
planting. Community
will repeat in time for
next planting season.
Small (n=64) evaluation
showed that 951 of res
pondents knew of ORT and
new
?ct
jla.
CHART I: PRICOR-SUPPORTED COMMUNITY FINANCING STUDIES:
PROBLEMS, OBJECTIVES, PROPOSED SOLUTIONS, AND RESULTS
STUDY
SUBJECTS
ADDRESSED PROBLEM
OBJECTIVE
PROPOSED
SOLUTION
RESULTS
COMMENTS
Thailand/
Orathip
RG
Inadequate PHC financial
resources
Multipurpose model more
Identify, test, and rep Combination of (1) revol- Thai government has
(1) adopted multipurpose cost-effective and sustain
ving fund mechanisms
licate cost-effective
model for National Rural able
models of conmunity
based on existing PHC
financing for PHC acti funds; (2) multipurpose Development Program;
models; and (3) multi
(2) encouraged single
vities
purpose funds to diver
purpose models with
sify; and (3) encouraged
health cards
establishment of multi
purpose funds.
Zaire/
Baer
RG
Inadequate resources to
sustain PHC services
Develop appropriate pay Fee-per-episode and perment schemes for self
consultation charging
financed health centers systems compared
that do not adversely
affect utilization
ABBREVIATIONS USED
Agency for International Development
Cost Containment
Community financing
Conmunity health worker
Fundaco Services de Saude Publica (Brazil)
Ministry of Health
Operations research
Oral rehydration salts
Oral rehydration therapy
Primary health care
Groupo Profam (Mexico)
Private Sector
Resource Generation
Rural health motivator
Resource Reallocation
AID
CC
CF
CHW
FSESP
MOH
OR
ORS
ORT
PHC
PROFAM
PS
RG
RHH
RR
i
6 health centers covered Some implementation problems
in fee-per-episode scheme
a median of 66.5% of
due to lack of cooperation
direct Preventive and
Curative costs. Effects from health center staff;
need further research in
on utilization mostly
reflected staff and com price-setting
munity reactions rather
than payment scheme
CHAPTER 2
INFLUENCING THE POLITICAL AND POLICY ENVIRONMENT
Many governments have policies that inhibit community
financing or user payments of any kind, while only a few (such
as Thailand’s and China’s) have policies that directly
encourage it. Discouraging policies encountered in one or
more PRICOR studies included:
a constitutional requirement that health care
be free (Somalia)
a requirement (formerly in Zaire) that all
receipts from user fees revert to the central
treasury
Lack of local control over such expensive
inputs as personnel (Haiti)
Policies, as in Swaziland, that prevent
community health workers from performing high
demand curative functions.
In addition to formal policies, certain political norms
and expectations also affected the potential for community
financing. In Honduras, researchers reported a widespread
perception among political leaders that the country is too
poor for community financing to be practical. Health care
providers in Rio de Janeiro, Brazil, reportedly felt that
health services should be accessible to all and thus resisted
charging fees for them. In countries where senior personnel
turn over quickly, people may hesitate to try new ideas or
change the status quo.
On the positive side, decentralization policies
explicitly encouraged community financing in Zaire and
Honduras by forcing local institutions to find their own
support or at least allowing them to keep whatever they could
get. The dramatic spread of community financing in Thailand
largely reflected efforts of the royal government.
Political and official policy environments played a
variety of roles in PRICOR-supported studies. Studies in
Mali, Somalia, Dominica, Senegal and Honduras, explicitly
tried to influence financing policies, looking both at the
adequacy of their response to health problems and at the
feasibility of change. Other studies attempted to change
policy by operationalizing new approaches and demonstrating
their political and practical feasibility. Virtually every
study produced findings on facilitating factors and
constraints that either encouraged or discouraged community
financing. Finally, political and policy environments
affected the ease with which researchers were able to carry
out their work.
10
STUDIES AIMED AT POLICY CHANGE
Studies in Dominica, Mali, Somalia, Senegal, and Honduras
analyzed current financing requirements and achievements and
made recommendations for change.
In Dominica, inadequate drug financing and
frequent stockouts led the government to seek
PRICOR and Management Sciences for Health
assistance to devise a national revolving drug
fund.
In Mali, the Director General of Planning for
Health and Social Services identified, costed,
and attempted to find viable financing for the
most essential PHC interventions. In the first
year following the study, one district adopted
several of the study’s recommendations.
In Somalia and Honduras, researchers studied
private health care expenditures in the hopes
of changing government policies. Policy
effects are unclear, though the Honduran
government did order MOH hospitals to begin
covering 30% of their expenses through user
fees.
In Senegal, USAID’s desire to find long term
recurrent cost financing for the Sine Saloum
project, especially for supervision, led to a
PRICOR-supported study of current government
spending patterns and possible options for
change.
Studies concerned with financing policy sought both to
provide new information and to suggest specific reforms.
Principal investigators believed that policy makers needed
information on:
(1) Current private expenditures, what they
are for, and who makes them; (2) Evidence that user charges
are politically acceptable, both to health care providers and
to influential leadership and population groups; and (3)
Guidance on how to operationalize new financing policies.
With a few exceptions, most studies found that user charges
were politically acceptable and that some users may prefer to
pay official fees if they replace variable and perhaps
excessive unofficial fees (Zaire). Users may also be quite
willing to pay if new revenue pays for more accessible or
higher quality services (Brazil). Most studies also
demonstrated the operational feasibility of community
financing, as described throughout this paper.
11
FACILITATING FACTORS AND CONSTRAINTS
While not attempting to change policy factors or the
political environment, many researchers found that political
considerations affected the range of community financing
options that they could propose.
In Honduras, for example, researchers found
that Ministry of Health personnel considered it
inappropriate to charge for basic services but
that they were willing to permit charges for
speciality care. Researchers proposed charges
for hospital care.
■
In Brazil, community residents were apparently
willing to pay for accessible and high quality
services, but health providers considered it
almost a breach of ethics to charge the poor
(even if this meant denying care to those
willing to pay). In retrospect, researchers
concluded that they could have charged more.
During the 1940s and 1950s in Zaire, the government had a
highly centralized policy, requiring that all receipts revert
to the central treasury. In the 1970s, some pilot PHC
projects were started, mostly by religious organizations, but
these survived mainly due to outside support. Finally in 1980,
the government authorized state hospitals to retain receipts
locally and adopted a laissez-faire policy toward religious
institutions. The government officially encouraged prepayment
arrangements, but allowed local institutions to find out what
worked best. This laissez-faire policy helped researchers by
allowing them greater latitude to investigate alternatives.
A significant question for further country-specific
research is whether it is more effective to address policy
issues head on and to delay community financing until they are
resolved, or whether community financing experimentation
should start beforehand as a way to demonstrate what can be
done. In Benin, Brazil (Lassner), and Bolivia (Gonzalez),
researchers and governments may have found it convenient to
conduct small scale community financing experiments as a
possible prelude to policy change. Studies may also have
influenced policy, even if that was not intended: senior
Philippine health care managers expressed interest in the
revolving drug funds created there, while the Thai government
now encourages multipurpose revolving funds because of study
findings. Researchers in Mali, Zaire, and India also saw
likely national ramifications of their studies.
12
EFFECTS ON RESEARCH PROCESSES
A number of PRICOR-supported researchers reported that
support of high level decisionmakers facilitated their work.
The support of the prime minister greatly helped PRICOR/MSH
researchers in Dominica, but political concerns dictated that
a smoothly functioning procurement and distribution system be
in place before fees were introduced. Within the Ministry of
Health, changes in the minister and principal secretary
delayed critical decisions.
CONCLUSIONS
Participants in a conference of PRICOR researchers
concluded that:
Support of high level decisionmakers is a
facilitating factor, possibly an essential
factor in the implementation of financing
schemes. This is much more difficult to
achieve in an unstable political environment
where there are many changes in key
decisionmakers.
It is essential to understand the values of the
decisionmakers at all levels - policy makers,
managers, community members - in addition to
existing practices. For example, in Honduras
it was acceptable to the Ministry of Health to
charge for special services but not for basic
care.
Government policies that allow regional or
local managers to try out new ideas encourage
community financing, whether decision making is
completely decentralized, as in Benin, or there
is just an unofficial, laissez faire attitude
from the central Ministry of Health, as in
Zaire. Depending on centralized, bureaucratic
systems greatly slows and restricts the
development of community financing.
Researchers need to recognize the time it takes
to develop support among decisionmakers and to
develop the financial and accounting systems
necessary to sustain financing. Studies which
tried to implement incremental phased-in
changes, as in Dominica, received support from
the central ministry. This allowed time for
the development of needed central systems and
also allowed for attitudinal changes that often
take some time to occur.
13
In several countries, including Honduras,
Somalia and Brazil, it was the key
decisionmakers, not the communities, who were
opposed to cost sharing by users.
Documentation of the substantial health
expenditures households were making and their
expressed willingness to pay for services were
used in an attempt to change decisionmakers’
basic perception of the necessity of free care
for all.
Economic crisis or the fear that external
funding will be reduced can stimulate interest
in community financing both by governments and
by users. Where providers and communities
believe that outside funding is certain, it is
very difficult to mobilize community resources
for health care.
14
CHAPTER 3
DEMAND FOR HEALTH CARE:
COMMUNITY INTERESTS AND RESOURCES
INTRODUCTION
Demand is the relationship between the amount of
something that a consumer will purchase and those factors that
determine that amount such as income, travel costs, price, and
perceived need. Demand is not a number but a relationship.
Demand is often confused with need, but they are not the same.
Someone may need medical care (because it would improve a
particular health condition such as tuberculosis) but not know
that they are sick or not believe that a treatment is
efficacious and therefore not seek out medical care.
Demand is not directly observable, but utilization is.
Utilization is the amount of health care that equates demand
and supply. Bartlett identified five categories of factors
that affect the demand for health care services: demographic,
biologic, cultural, service-related, and economic. Measuring
demand is further complicated by the context of the analysis
including seasonality, inflation, and external factors
affecting utilization such as natural disasters, disruption of
services, and education.
None of the PRICOR-supported studies completed complex
demand models as the primary purpose of their studies.
Rather, descriptive data about service utilization and service
preferences were collected in a number of the studies in order
to solve a specific operational problem in a community
financing system. The results are descriptive and apply to
specific community situations.
This chapter discusses ability and willingness to pay
viewed prospectively as a policy development and scheme design
issue. Chapter 10 again discusses these topics but viewed
retrospectively as community financing activities were
implemented.
HOW DEMAND WAS STUDIED
PRICOR-supported researchers used the following
techniques to study the public’s ability and willingness to
pay for health care:
National household expenditure surveys in
Honduras, Somalia, and Mali
Local or district household surveys in Liberia,
Brazil/Lassner, and the Philippines
15
■
Surveys of health providers in specific service
areas in Brazil/Lassner, Swaziland, and
Thailand
A survey of pharmacists in Mexico
Observation of families in Honduras and
Bolivia/Miller
Case studies of health care providers in
Honduras, Bolivia/Miller and Thailand
Meetings with an expert group in Brazil/Lassner
Studies of what people liked and disliked about
health care sources in Liberia, India, and
Bolivia/Gonzalez
Direct discussions with community leaders in
Benin, Bolivia, Brazil/Lassner, Swaziland,
Somalia, Liberia, the Philippines, Mali
Review of existing reports in Brazil/Baker and
Bolivia/Miller.
The PRICOR studies which collected information about
demand can be divided into two categories: those intended to
influence policy decisions and those measuring community
resources for the purpose of developing a particular community
financing scheme.
CURRENT HEALTH SPENDING AND UTILIZATION: POLICY STUDIES
Honduras
PRICOR-supported investigators used three data sources in
Honduras. First, 25 Honduran families were intensively
observed to gather data on illness behavior patterns. One
relevant finding was that individuals used multiple sources of
care for each illness episode.
Second, a survey team completed 1,017 household
interviews in 29 sampling sites in four Health Regions.
Approximately 26 percent of the sample population reported
having been sick in the past 15 days. Those who did
experience an illness episode reported the following:
Spending per illness episode, mean: $8.28
median: $1.00
Spending for home treatment, mean: $2.50
median: $0.50
Spending for private medical care, mean:
$34.45 median: $25.00
16
Monthly drug spending, mean:
median: $1.00
$4.50
Reported monthly household health spending had a mean of
$20.82 for all families and represented 11.4% of total monthly
expenses, ranging from 8.5% in urban areas to 12.9% in rural
areas. This was the third highest expense after food and
clothing. The investigators estimated that private health
expenditures consumed close to 11% of the 1983 GNP.
Drug expenditures were 50% of total health care
expenditures. Unfortunately, some drug expenditures were used
for inappropriate treatments. For example, those with
diarrhea paid an average of $3.00 for drugs even though 2
packets of ORS would have cost only $0.50. Drugs prescribed
by MOH personnel had to be purchased on the open market
because MOH had inadequate supplies, also adding to the cost
of treatment.
Ten case studies of clinics, health centers, and
hospitals that were charging for services or receiving
community support were the third source of demand data.
Direct observation and discussions with informed people at the
facilities were carried out as well as interviews with
community members. Results of this part of the study are
discussed under "willingness to pay" below.
Somalia
The purpose of the PRICOR-supported study in Somalia was
to analyze the use of modern drugs in rural areas. This was
an important public policy concern because all drugs sold in
Somalia had to pass through government inspection for quality
control, limiting the free flow of drugs into the private
sector.
First, problems associated with drug delivery in Somalia
were identified in discussions with PHC experts and through
direct observation of people in the field. Many people
working in Somalia assumed that peoples’ lack of disposable
income was one of the major factors limiting drug use at the
village level. However, on several occasions expatriates were
asked to purchase drugs on behalf of village committees,
suggesting that villagers might in fact be willing and able to
pay for drugs.
Researchers found very low private expenditures for
drugs, however, compared to other household expenditures, In
the 716 households contacted, the average drug expenditure in
the previous 6 months was $0.83 per household. For the 32.4%
who purchased drugs, this average was $2.55. Other average
expenditures in the comparable 6 month period were $20 for
tobacco, $16 for tea, $18 for celebrations, $335 for food.
Drugs represented 0.2% of total household expenditures during
17
the study period while food represented 71.8%. Expenditures
for treatment over the six month period averaged $1.63
including both drugs and consultation fee.
The researchers speculated that the very low average
expenditures for drugs was a function of the poverty of the
region, the high cost of food, and the unavailability of drugs
in the rural areas. The importance of the last factor (supply
of drugs) was underscored in interviews with drug dispensers.
Mexico
A market research study of ORS was carried out in Mexico
by PROFAM. Mexican health agencies have widely encouraged the
use of packets of of ORS granules for the treatment of
diarrhea. PROFAM was interested in producing, marketing, and
distributing an inexpensive, easy-to-use ORS tablet and wanted
to know if the public would find it acceptable. Since
drugstores are a popular source of ORS products, it was
decided that a survey of pharmacists would provide a
relatively good measure of supply and demand for ORS.
The survey included 116 private drugstores: 55 in Mexico
City, 30 in the hot region, and 31 in the temperate region.
The researchers asked pharmacists and other drugstore
personnel about the distribution and sales of ORS, their
knowledge of the purpose and correct use of ORS, demand for
ORS, and their opinions on the presentation of ORS products.
All but 1% of the drug stores reported that they sold
ORS. The price of the pre-mixed liquid ranged from $0.73 to
$0.94 per 500 ml bottle. Most sales were made to women.
Largest quantities were sold in the spring and summer. The
products with the highest demand were Pedialyte (75%), Solural
(21%), and Electrolite (3%).
The drugstore personnel felt that, in considering the
needs of the population, the prices of the ORS products were
generally too high. A few pharmacists felt that demand could
be increased by adding flavoring to existing ORS products.
There was no significant difference in opinion when asked
about the suitability of either ORS tablets or granules.
PROFAM decided not to proceed with the production of ORS
tablets for two major reason: suitability and cost. The
survey showed no particular consumer advantage of the ORS
tablets over granule packets, and the tablets also cost
slightly more than granule packets to produce.
Mali
A survey of 1,800 family and household heads in two
districts was conducted to obtain information about health
needs, current and suggested methods of paying for health
18
care, methods of paying CHWs and TBAs, household economic
data, and attitudes towards PHC programs.
Community financing of health care in a Malian village is
complicated by the economic structure of the family unit, and
by the fact that many farming famimlies are only able to farm
at subsistence levels, leaving little or no disposable income
for other necessities, The priority expenditures of a typical
family unit were food, health, and taxes, in order of
magnitude.
Eighty percent of the respondents in the Koro District
said that they consume all of what they grow, largely a result
of the continuing drought conditions in that part of the
countryi They are severely limited in their ability to pay
for health care. In Kita, on the other hand, farmers grow
peanuts as a cash crop, and only 4% live at subsistence
levels. Family members usually work collectively, but
individual health-related expenditures are most often paid for
by the family head (70%). Most of the respondents complained
of this system, and expressed interest in a more flexible
method which leaves the individuals more free to seek health
care.
CURRENT HEALTH SPENDING TkND UTILIZATION:
DESIGN STUDIES
COMMUNITY FINANCING
Brazil/Lassner
In Brazil, surveys of community residents and of CPAIMC
health service providers were conducted to collect data on use
of health services, on sources of health care for under 5s and
women 15-49, and on payment for this care. Community leaders
were also interviewed to get their opinions and suggestions
regarding community financing.
In the community survey, 93% reported at least one
household member was covered by public sector health
insurance. Government health care facilities were the major
source of care for women and children. Preventive health
services were sought for 33% of the children and 46% of the
women at their most recent health care visit.
Payment for health care was made for only 6% of the
children and 8% of the women, based on most recent visits.
Twenty-two percent of women whose most recent visit was for
dental care paid for that care, while 17% paid for injections
and 7% paid for gynecological care. Most people did not pay
for popular child services such as vaccinations, pediatric
care, well-child care, and emergency care, as they were
available free in public health facilities. Injections and
dental care were more likely to be paid for since these
services were not always available in public health centers.
19
Household members also paid for drugs and contraceptives from
pharmacies.
In generating alternative schemes, health providers were
very resistant to charging for services because they believed
that people were not able or willing to pay. Survey results
indicated that people were used to certain health care
payments and suggested what kinds of services people might
agree to pay for at CPAIMC facilities.
Liberia
The PRICOR-supported Liberian study examined ways
communities could use to generate funds to finance some or all
of their PHC services. Researchers collected data for their
study using a household survey of three rural villages and a
series of meetings with village leaders.
Seventy percent of households surveyed reported an annual
income of less than $200 and 41% reported incomes of less than
$100 a year, substantially less than national average of $280.
Though these data were not formally incorporated into the
development of alternatives, the survey itself proved to be an
effective entree into the comunities and generated support for
the project.
In each village the town council and other village
leaders met to consider four issues: what health care
services would be provided, who the health care provider would
be, who would participate, and how much the services would
cost. Further, the leaders discussed eight alternative
financing schemes for generating PHC funds weithin the
community. Each of the three study villages constructed a
preference matrix and, on each, the same four schemes ranked
the highest, although in different order: drug sales,
sales,
porduction-based prepayment, community and individual labor,
and donation and ad hoc assessments.
Benin
All
m,_l± the
cue aemana
demand aata
data collected in the Benin study came
from meetings with commune leaders and projections of health
service utilization based on estimated morbidity patterns.
The researchers contacted commune leaders directly in order to
identify which services and goods the communities would be
willing to pay for.
Leaders agreed that villages would support the costs of
drug supplies and VHW remuneration providing the government
paid health center salaries and infrastructure costs and
foreign donors covered investment costs.
20
Swaziland
A household survey was conducted to gather information
about use of health care, health expenditures, and willingness
of communities to support rural health motivators (RHMs).
Twenty percent of the respondents said they would be willing
to support the RHM financially, while 64% said they would
support them with in-kind contributions.
Interviews with 31 RHMs confirmed these findings: not one
felt that the community would be willing to support them
financially, mainly because they were already receiving a
stipend from the government (approximately $20 per month).
Half of the RHMs felt that communities would be willing to
give them in-kind support.
The researchers concluded that the communities were not
willing to pay for the services that the RHMs were currently
providing (health education), that they needed to enhance
their skills and increase their perceived value, and that inkind payments were most appropriate.
Philippines
Six participating barangays were studied to determine:
perceived health problems in the community; attitudes toward
and utilization of available health services, particularly the
Barangay Health Workers; current expenditures for health
services in the public and private sectors; and stated
willingness to pay for additional services not currently
available.
The major health problems were perceived to be
respiratory and gastro-intestinal illnesses. Annual health
expenditures per household were estimated to range from $29 to
$43. The unavailability of drugs and poor water supply
facilities were perceived as major health-related problems.
The majority of households expressed willingness to
participate in the community financing of health activities.
The purchase of drugs and the operation of a drug depot were
projects that community members were most willing to finance.
The study team charted the results of the baseline survey
pictorially and presented them to the barangay residents at
community assemblies.
Brazi1/Baker
The Brazil/Baker study was a retrospective analysis of
various aspects of primary health care activities managed by
the Fundacao Services Especiais de Saude Publica -(FSESP) and
included an analysis of community expenditures for the
development and operation of water systems.
21
Researchers discovered that small communities were able
to pay for both maintenance and operation of water supplies if
a suitable financing system was used. In the case of FSESP,
the regular payment of water bills by users was achieved
through the efficient management of the water system by the
municipal authorities and proved to be adequate to cover
operating costs. The researchers concluded that expanding
water supply coverage depended more on the method and
efficiency of billing and collection than on the wealth of the
commnity and the amount of charges for water.
Bolivia/Gonzalez
Researchers from the Instituto de Investigaciones Medico
Sociales (TIMS) conducted a household survey in the rural
areas to determine conditions under which communities would be
willing to support a health worker, how much they would pay,
and how payments would be made (i.e. in-kind, fee-for-service,
annual installments, etc.). Results indicated that most
people would be willing to support health workers if they had
established working hours and were selected by the community.
Smaller communities would be more willing to support a health
worker who resided locally and attended only to community
members.
Discussions with villagers indicated that, although most
of them clearly recognized the need for a local health worker,
they were reluctant to commit their own scarce resources until
they were assured that the worker would be competent,
reliable, selected by the community, and well supplied with
medications.
Haiti
Save the Children Canada (AEDC) examined the best ways to
motivate CHWs to provide preventive services and encourage
mothers to learn about, use, and maintain competence in child
survival interventions.
Using fee-for-services was explored, involving the
possiblity of raising money to support CHWs in one of two
ways: increasing curative care receipts and using the
increased revenue to pay for preventive services, or making
beneficiaries pay for preventive services. Examination of
financial records revealed that receipts from curative care
could not begin to pay for curative care expenses let alone
subsidize preventive care activities. To determine patients’
willingness to pay directly for preventive care, 900 mothers
from three study sites were interviewed about their socio
economic status and expenditures for health care. Results
showed that mothers paid a considerable’amount for curative
services but nothing for preventive care. People in rural
areas were not in the habit of paying for preventive care.
Women perceived curative care as their primary health need.
22
*
and CHWs affiliated with private health institutions do not
provide curative care.
The possiblity of supporting CHWs through community funds
was also explored. Leaders, treasurers, and members of
various community groups were interviewed and community
records examined when possible to answer the following
questions; 1) can communities raise sufficient sums to pay
CHWs? and 2) if they could, would they be willing to use the
collected funds to pay CHWs. In addition, a quantitative
survey of potential beneficiaries of health worker-related
services was completed as well as a qualitative survey of
selected mothers, group members, and group leaders, The
researchers found:
Communities can get together and raise
significant amounts of capital if they believe
the purpose is valid.
Purposes communities considered valid were:
capital for investment in group income
generation activities;
credit; and
insurance for catastrophic events (identified
at only one of the three sites).
India/Elkins
Data on health expenditures by community residents and
their preferences for services were collected in interviews
with health providers, community leaders and residents, focus
group discussions with staff and community members, and a
review of project records. Ability to pay for even curative
services was recognized as an important problem in several
projects by residents and project directors. Multi-purpose
organizations achieved wide participation by starting with a
strongly felt need in the community, even if that need was
only indirectly related to health.
Thailand
Researchers reported that in 1983, 18.9% of all health
sector expenditures in Thailand were made by the Minstry of
Public Health (MOPH), 12.2% by government agencies, and 68.9%
by private sources.
ABILITY TO PAY
The findings cited above indicate that study communities
had considerable ability to pay for health care, at least in
the aggregate. The great differences between means and
medians in Honduras, however, indicate considerable
maldistribution of resources.
23
Researchers in Brazil and India asked local residents
directly about ability to pay, using two different kinds of
questions: 1) Is cost an obstacle for you? and 2) Is cost an
obstacle for others? In Adayar, India, the most frequent
response to the second question was that others could not
afford to pay for the subscription; 85% cited inability to pay
as a reason why some did not join. The next most frequent
response, that the program did not provide the needed
services, was mentioned by only 10%. Also in Adayer, 12
former subscribers to a health care program were asked why
they dropped out of the program, and 9 responded that they
could not afford to continue. Thus, inability to pay was
identified as a significant constraint to service demand for
at least some of the local residents in one project in India.
Similar results were reported in Brazil. Nearly threefourths (71%) of those interviewed in urban Brazil said that
charges were not an obstacle for themselves, but 52% said that
cost was an obstacle for at least some of their neighbors.
Seasonality was identified as a variable affecting
ability to pay in India and Honduras. Investigators in India
felt that financing schemes must adjust to variation in
income, both seasonal cycles and longer periods of one or more
years. Cooperatives that meet recurrent expenses in normal
years may have problems in times of financial adversity. For
example, one health cooperative in Ambilikai had severe
financial difficulties when drought affected the ability of
members to make prepayments. Similarly, in Honduras, ability
to pay varied during different periods of the year, and
investigators saw a need for the MOH to develop limited credit
mechanisms.
WILLINGNESS TO PAY
In some PRICOR studies, researchers had to estimate
willingness to pay for services that people had not yet
experienced. They did this using surveys, community meetings,
focus groups, and previous knowledge and experience about
local use of similar services. Willingness to pay was easiest
to predict for drugs and services that people were already
obtaining outside the community. Given improved accessibility
and roughly comparable quality, some researchers assumed
continued willingness to pay. Brazi1/Lassner, for example,
concentrated on what people were already paying for (rather
than on how much). A subsequent survey found that low demand
was partly due to lack of public knowledge about facilities
and services that were available.
In the Philippines, surveys were conducted of community
leaders, health care practitioners, and program beneficiaries
as well as of comunity members. The majority (89%) of
household heads expressed their willingness to participate in
health financing schemes and had generally positive attitudes
24
toward participation. PRICOR-supported research in Bolivia
found similar attitudes.
In Honduras, the majority of surveyed heads of households
(93.5%) reported that they would be willing to pay an average
of $0.50 for MOH services. Most (92%) were also willing to
pay for medication in MOH facilities, although half said they
would do so only if drug supplies were improved. In all 10
institutional case studies, patients reported themselves
willing to support and/or pay for health care services,
provided they felt the services were of a high quality, The
most important factor in the decision to pay was the total
charge with the ceiling ranging from $4.00-5.00.
A survey
was also done to determine the level of satisfaction of the
community towards health services received. Willingness to
pay was found to be directly linked to how householders
perceived the quality of services provided.
In Benin, investigators asked villagers in communal
meetings about what health care services they would support.
Villagers indicated that they were willing to support CHWs and
drug costs. Curative services were chosen as the cornerstone
for generating funds as the community displayed a willingness
to pay for these. On the other hand, preventive care was
delivered free of charge as the community did not seem to
spontaneously seek this type of care.
Directors of health care facilities in India expressed
concern about the effect of poverty upon willingness to
participate in paid health care programs. One director
stressed that although poverty was a problem in an objective
sense, i.e., that health did compete with other necessities
for available funds, part of the problem was subjective. Some
community members felt that their poverty justified their
getting services free. This attitudinal problem was
particularly acute in Kangazha where many services were
originally free. Similarly, the majority of community members
surveyed in Bolivia/Gonzalez felt that the government should
provide health care services free of charge.
To reduce difficulties in predicting demand, many
researchers worked closely with community leaders, as noted
above. In Haiti, Swaziland, and the Philippines, residents
made it clear that they would not pay directly for preventive
care.
(See Chapter 9 for further discussion of generating
resources to support preventive care. Chapter 10 evaluates
ability and willingness to pay as experienced by researchers
during scheme implementation.)
CONCLUSIONS
A major operational question for many researchers was how
much to try to learn about PHC demand and how to go about
learning it. There was clearly a difference in this regard
25
between studies aimed at policy change and studies aimed at
the design of specific micro-level financing schemes. The
policy-level studies relied on quantitative results which
documented health expenditures or use of services. For policy
change, they needed to produce convincing statistics. Studies
in Honduras, Somalia, and Mexico collected data relevant to
policy issues. Data was collected using surveys and expert
opinion. In Honduras, using multiple methods of data
collection proved to be especially useful in confirming the
results to sceptical decision makers.
The micro-level studies used more heuristic techniques
for establishing service preferences and negotiating the
design and management of the community financing scheme, In
the real world, most managers cannot wait a year for the
analysis and results of a survey in order to make an
operational decision. Large baseline surveys are critically
important for establishing patterns of resource consumption in
order to shape policy, but they are less useful for design and
management decisions on a specific scheme. Qualitative
community-level data were generally found necessary for
predicting willingness to pay for new health care services.
Working at the community level on the development of
financing alternatives, several researchers found that data
collected on some of the larger macro questions, such as
actual health expenditures, proved less useful than relatively
simple questions on kinds of goods and services people are
used to paying for and their other specific preferences, For
example, in Bolivia and Liberia, researchers used simple
preference matrices to identify financing alternatives
acceptable to the communities. Though the principal
investigator of the Liberia study collected data on
expenditures, he did not use the data collected on annual
income for operational decisions. His knowledge of the
community and their basic preferences seemed to be enough to
initiate the financing scheme. In Brazil, the researchers
found that the communities preferences and values differed
from those of health professionals. This underlined the
importance of researchers/managers not assuming that they
understand the demand/preference for services without
consulting the clients.
Finally, some researchers used pre-design surveys at
least partly as entrees to generate community discussion and
participation. The process of involving the community in
aspects of the study proved to be as critical to the success
of the scheme as the findings of the data collection phase in
Liberia, Bolivia, Philippines, and Swaziland. Community
involvement resulted in greater acceptance of the scheme.
26
CHAPTER 4
PRIMARY HEALTH CARE COSTS
PRICOR-supported researchers collected cost data for
program planning, increasing program efficiency, setting
prices, and planning and evaluating cost recovery.
First, cost information was helpful in planning the
resource implications of alternative program strategies, This
was important especially with regard to the long term
viability of the program. For example, in Bolivia/Gonzalez,
the investigators identified a strategy to extend PHC services
into underserved communities by introducing a lower cost
health worker.
A second use of cost information was for planning more
efficient use of scarce resources. In Jamaica, a cost index
was calculated for health center services and from this a
model developed for more efficient allocation of the most
expensive resource: personnel.
Cost information was also useful in determining prices.
Cost analyses in Benin were used to calculate a single fee
that had to cover multiple components of the PHC system.
Finally, cost data were essential in order to plan and
evaluate the cost recovery of a specific financing system,
defining what costs were to be covered by the financing scheme
and at what level. It was an important outcome measure for
several studies. In Zaire, health centers were expected to
cover 100% of their direct costs. In Benin, the receipts from
the health system were to cover all costs except government
salaries, electricity, and start-up.
PRICOR-supported investigators used a variety of costing
protocols with varying degrees of detail. Methodologies
reflected how the cost data were to be used and the
sophistication of the information system. In some studies, as
in Mali, cost estimates were used instead of hard data because
the latter were not available and the results, in any case,
were to be used for policy-making rather than for detailed
implementation. In Benin and Jamaica, on the other hand,
rough estimates would not have been adequate for analyzing
cost-recovery and productivity. In both of these countries,
very detailed cost analyses were required to evaluate system
outputs.
Researchers in Benin, Brazil (Lassner) and Zaire carried
out cost analyses of full health services. Other researchers
analyzed costs of specific programs: in Brazi1/Nations, the
use of traditional healers to deliver ORT services; in
Dominica, the use of lower cost drug suppliers; in Jamaica,
personnel costs for curative’care; and in Korea, the use of
27
formal community groups to enhance PHC service delivery.
Retrospective cost analyses were carried out on a water and
CHW program in Brazil and for community-based health projects
in India. Several other studies collected partial cost data.
METHODS
In Brazil, systems for collecting, tabulating, reporting
and monitoring costs needed to be developed. Arrangements
were made so that the staff member responsible for cost
control at each unit and mini-post made monthly reports of
direct, indirect, and total costs.
Researchers in Zaire and Benin valued the resources
actually used in providing services, using shadow prices for
resources that were donated. Cost data were collected in the
following categories: time period (6 months); preventive and
curative functions; input line item; geographic location
(Zaire across regions, Benin across zones in the same health
center service area).
As can be seen in Chart 2, the methodologies for the
Zaire and Benin studies were similar: each valued all inputs
and applied the same assumptions for amortization. Results
are difficult to compare, nevertheless, because PHC services
were at different levels of development in the two countries.
Investment costs differed because Zaire’s was an on-going
program while Benin’s was new. Recurrent costs were
calculated in the same manner, though researchers in Benin
collected very detailed data about a small service area, while
in Zaire the analysis covered many health zones and was
necessarily less detailed. Researchers in Benin allocated
indirect costs through a complex functional analysis, while
researchers in Zaire simply divided regional office costs by
the number of health centers supervised. Zairian researchers
interviewed staff to estimate unrecorded resource use, while
researchers in Benin did a thorough time and motion study to
estimate personnel time needed for various functions. In the
case of Benin, PRICOR-supported cost analyses evolved into a
routine management information system so that cost monitoring
continued after research ended.
FINDINGS
Researchers in Benin, Brazil (Nations), and Zaire
analyzed and reported service delivery costs. In addition,
several researchers calculated the proportion of costs covered
through community financing. Findings on service delivery
costs and expenditures are reported here, while findings on
cost recovery are reported in Chapter 10.
Costs are often equated with expenditures, though this is
only correct if all resources consumed are paid for. Most
primary health care projects, however, receive ’’free”
28
CHARI 2:
COSTING METHODOLOGIES USED IN PRICOR-SUPPORTED STUDIES
DEFINITIONS
INVESTMENT COSTS
STUDY
TIME PERIODS
COST CATEGORIES
Benin
Every six
months
between July
1983 and
June 1985
Prevent i ve/curat i ve; All personnel, material,
transporation, and
input line items;
geographic locations infrastructural costs
incurred during the
development of project
services, amortized
over 5 years
BraziI
One 9-month
period
Input line items;
by service unit
Zaire
Six-month
Prevent i ve/curat i ve; Resources with a life
input line items;
expectancy of more
periods
than one year; resources
between
by health centers
consumed in start-up
October 1984 (10 centers in
and September 5 zones)
activities; all such
costs amortized over
1985
5 years
N>
Not analyzed
SOURCE OF DATA
RECURRENT COSTS
INDIRECT COSTS
Community-supported costs: Drugs,
VHW remuneration, gasoline and
maintenance costs for mobylettes
used for supervision and MCH
clinics, petrol for the cold chain,
the VHW information system, and
certain health center functioning
costs. Government supported
costs: salaries of health center
employees (government officials)
plus electricity.
Supervision, training, Health center
records; time/motion
administration/
studies; market
logistics,
i nformat i on/mon i tor i ng, value estimations
program development
Personnel, medicines, supplies,
and general expenses such as
electricity, maintenance,
transportation
Program management
and adninistration,
training, and
evaluation
Health center
records
Resources with a life expectancy
of less than one year or those
needed to be replaced repeatedly
in one year
Zonal office
administration
(supervision costs
were considered
direct costs)
Health center
records; estimates
based on interviews
with health center
personnel; market
value estimations
resources which are never recorded in budgets or expenditure
statements. In this case, costs are more than expenditures
though expenditures are often the best approximation available
for costs. In this chapter, costs will be used to mean actual
resources consumed (paid and unpaid) unless otherwise noted.
In Benin, recurrent costs for the period between July 84
and June 85 averaged $3.00 per person per year for a
population of 10,000 persons. This figure included all costs
incurred in the implementation of PHC for the entire commune,
including those for both health center and village level
activities. Curative care and logistics together represented
almost 25% of total recurrent costs. Personnel alone
accounted for 70% of recurrent costs. Researchers estimated
that if the Pahou Demonstration Project were replicated in an
area with a similar ecology, recurrent costs could be reduced
by as much as 50%, because of lower personnel, transport, and
infrastructure costs. Baseline surveys and the expense of
internships would also be eliminated.
In Zaire, researchers calculated three cost measures for
curative care, namely, costs per capita, costs per visit, and
costs per episode (defined as all visits needed for a single
course of treatment). At the health center level, the
greatest operating expenses were for medical supplies,
personnel, and transport, in that order. Calculations
included costs of associated supervision and mobile team
support from the zonal level.
All three cost measures varied enormously from one center
to another. Costs per capita in the period October 1984 to
September 1985 varied from 2 cents in Tshileo health center to
90 cents in Lukunga. Costs per visit were lowest in Kangoy at
4 cents and highest in Katanda at $3.46. Costs per episode
ranged from 27 cents in Kangoy to $7.46 in Katanda. Most
calculations involved ten health centers in five health zones
scattered throughout the country.
The enormous variations reported from Zaire may have been
partly due to data inconsistencies but differential
utilization, coverage, and distribution of resources appear to
have been more important. Tshileo reported extremely low
utilization per capita in addition to low per capita costs;
the center is clearly responsible for a larger population than
it has the resources to serve. The Kangoy center achieved its
extremely low per visit and per episode costs partly because
of its high utilization, since this spread fixed costs over
many visits. Katanda, on the other hand, reported very low
utilization and high fixed costs.
The average health center studied in Zaire incurred costs
of 95 cents per episode for health center salaries, zonal
supervisor and mobile team salaries for the time spent on
field work, all inservice training, administrative and
30
maintenance supplies, drugs, vaccines, kerosene, minor
building repairs, and transport for personnel and materials
including supervisors and the mobile team. Medical supplies
accounted for nearly half of all costs, though the proportions
ranged from a low of 8.8% in one center to a high of 70.1% in
another. Expenditures for preventive care (not included in
the figures above) accounted for 22 to 34% of health center
costs.
Zonal level operating costs other than direct support
costs (also not considered above) ranged from $3500 to $5500
per year in four of the five zones studied, even though the
number of health centers that the zonal offices supervised
ranged from 6 to 47. Certain zonal offices were obviously
better help to support subordinate health centers than were
others.
In Brazil, PRICOR-supported researchers led by Nations
mobilized, trained and supervised 46 lay healers and 4 village
health agents to deliver ORT; these healers delivered over
7400 liters of ORT in 12 months. Community residents also
constructed from scratch 5 fully equiped "curing rooms" or
popular rehydration clinics and upgraded and equipped 13
existing curing rooms. The total program cost for 15 months
was only $4027. Start-up costs including those for training,
supplies, materials to construct and equip the curing rooms,
and transportation were $2960. Training cost $870, while each
curing room cost $26 to construct and $43 to equip. Operating
expenses for a 15 month period were $1068, including costs for
supervision, incentives, salt, sugar, other necessary ORT
materials, and travel. Sugar cost an estimated 48 cents per
healer per month, enough for 36 liters of ORT.
31
CHAPTER 5
PROBLEM IDENTIFICATION AND OBJECTIVE SETTING
PRICOR researchers became interested in community
financing for a variety of reasons, most of them related to
the overall crisis in PHC funding. These were:
because managers feared excessive dependence on
external funding, or wanted to relieve existing
dependence;
because managers needed funds to expand PHC
services and goods within already served areas
or to extend them to currently unserved areas;
and
because of high CHW attrition rates or failures
in drug procurement and distribution systems.
Often the circumstances leading to community financing as
a possible solution were complex:
In Zaire, for example, program managers sought
community financing in order to implement a new
Ministry of Health decentralization policy.
Central government resources were not adequate
to permit service extension or new programs.
In Rio de Janeiro, CPAIMC researchers feared
their dependence on foreign funding sources,
while in Senegal, health planners sought
alternatives to use of USAID funds for
supervision and other recurrent costs.
Economic setbacks in Honduras,, and government
inability to buy sufficient drugs in Dominica,
motivated community financing studies in those
two countries. The Dominica study was also an
effort to improve drug procurement, inventory
control, and distribution.
The principal investigator in Liberia hoped to
help communities take on greater responsibility
for PHC.
■
Poor CHW morale and high attrition rates in
Bolivia and Swaziland led researchers to
identify financing shortfalls as key problems
requiring solution.
Chart 1 in Chapter 1 briefly lists study problems and
objectives as stated by the principal investigators.
32
SOLUTION OBJECTIVES
Most PRICOR-supported researchers stated their objectives
in terms of (a) goods and services to be supported or
strengthened; (b) financing, especially revenue generation and
cost reduction; (c) degree of community participation and
acceptance of responsibility being sought; and (d) quality,
accessibility, and utilization of services.
As suggested above, most PRICOR researchers had multi
dimensional goals, among which financing was not always the
highest priority. Dr. Gonzalez in Bolivia, for example,
sought to reduce high CHW attrition rates and ensure coverage
of vital PHC services in rural areas. CPAIMC in Brazil sought
to increase service accessibility and use in the urban slums
of Rio in addition to reduction of dependence on external
funding. Tony Augustin in Haiti wanted to give mothers
incentives for learning child survival interventions while at
the same time motivating health workers to teach these skills.
In Zaire, the study goal was to increase utilization rather
than to increase the proportion of costs covered; to do this,
program managers replaced arbitrary clinic fees, charged for
each visit, with fixed fees, covering complete illness
episodes. Dominican health care managers were just as
concerned to improve drug procurement and distribution as they
were to strengthen financing. In these cases and others,
community financing was seen as a means to an end, rather than
as an objective in its own right.
Achievement of more broadly stated objectives implicitly
required a systematic assessment of factors leading, say, to
increased service accessibility or CHW job stability. It also
required a more careful statement of financing objectives, In
some cases, the volume of resources generated, or even the
proportion of costs covered, was less critical than whether
the CHWs involved were satisfied. In Swaziland, for example,
(had they proceeded this far), researchers would have been
most concerned that the combination of morale-boosting and
tangible product resulting from contributed field labor was
adequate to keep the CHW on the job. Calculation of the
purely monetary value of the field work might have been
misleading. A similar situation occurred in Bolivia, where
communities attempted to generate inkind resources to
compensate CHWs; while collection targets were quantified (at
$7.00 per family per year), CHW stability was undoubtedly
affected by many other factors in addition to payment.
Stable, well-functioning revolving drug funds, as in the
Philippines, may be considered only partly successful if
researchers want them to cover preventive care and they are
unable to do so. Chart 1 indicates that PRICOR studies
differed in the extent to which objectives were stated in
terms of activities to be supported or inputs to be maintained
rather than in terms of specified monetary or non-monetary
resources.
33
This chapter, it should be noted, discusses solution
objectives, not knowledge objectives. Studies where
production of knowledge per se was the main objective are
mentioned only in passing.
GOODS 7KND SERVICES TO BE FINANCED
Researchers stated their objectives either in terms of
inputs (drugs, personnel, supply systems) to be paid for or as
outputs (MCH services, community health activities, etc.). In
Benin, Brazil (Lassner), Honduras, Mali, and Zaire, support
was sought for a general basket of PHC goods and services,
including drugs, health worker salaries, supervision, and
other expenses related to both preventive and curative care.
Researchers in Haiti and the Philippines sought community
financing specifically for preventive care.
(The Haitian
study succeeded in this, but the Philippine one did not.) In
Senegal the object was new financing for CHW supervision, In
terms of inputs, researchers in Bolivia, Swaziland, Liberia,
and the Philippines wanted to finance CHW salaries, and the
government of Dominica wanted user payments for drugs.
FINANCING OBJECTIVES: RESOURCE GENERATION
Some PRICOR studies sought community sharing of certain
cost burdens, while others identified specific goods or
services for which full community financing was required. The
distinction affected the kinds of cost monitoring systems that
had to be set up, the practical feasibility of cross-subsidies
within schemes, and the rigor with which community financing
targets had to be achieved.
The major example of a cost-sharing effort was the
Lassner study in Brazil. Researchers sought to reduce
dependence on external funds but not to eliminate it entirely.
When decisions were made about schemes to be supported,
assessment of cost recovery potential was defered to the
solution validation stage because no specific revenue
generation targets had been set.
In contrast, revolving drug funds (unless partially
subsidized, as planned in Dominica), had to fully cover
expenses in order to avoid decapitalization (that is,
reduction in the basic drug stock). RDF managers, therefore,
had to devise systems to monitor costs (especially drug
repurchase costs) so that prices could be set high enough to
cover expenses. While willingly extending credit to those
temporarily short of cash, managers of most PRICOR-supported
revolving funds had to insist on eventual payment because of
the practical difficulty of monitoring cross-subsidies (that
is, use of surpluses generated in one part of the program to
cover excess expenditures in other parts). A likely reason
for the success of community-level revolving drug funds in a
34
number of countries is that costs - provided they are stable are relatively easy to calculate.
FINANCING OBJECTIVES: COST REDUCTION
Cost reduction, and implementation of cost-effective
service delivery models, are essential elements in a
comprehensive financing strategy. This is especially true if
the resources saved can be devoted to more productive uses, or
if cost reduction makes services more affordable. PRICOR
studies that sought cost reduction or more efficient use of
resources include:
The Dominican study, where researchers and
managers insisted that more timely and costeffective drug procurement and logistics
systems be established before user charges were
instituted.
■
The Jamaican study and others aimed at staffing
patterns and time allocation
The Korean study, which identified costeffective means of mobilizing community
participation
In addition, many PRICOR-assisted decisionmakers
qualitatively considered cost and cost-effectiveness when
choosing among program alternatives.
COMMUNITY PARTICIPATION OBJECTIVES
Researchers in Liberia, the Philippines, Benin, and
elsewhere were most interested in strengthening the
community’s role in primary health care and saw locally
managed financing as a useful focus and facilitating factor.
The degree to which community participation was considered
necessary for achievement of the financing objectives of
various studies is discussed in Chapter 7.
QUALITY, ACCESSIBILITY, AND UTILIZATION
While service fees and drug charges are generally
believed to reduce utilization, the overall effect may be the
opposite when new revenues replace previous, inequitably
obtained resources, or are used to expand the service mix and
number of facilities. Some PRICOR researchers explicitly
sought this result. In Liberia, for example, a major study
goal was to bring basic drugs to the community and save
patients the time and travel costs of going to regional
centers. In Zaire, decentralization forced health zones to
rely on their own resources for maintaining and improving
services. Fixed service fees were promoted as a way to
increase service utilization by doing away with the
35
unpredictable fees sometimes arbitrarily set by clinic
personnel. The Dominican study was at least as much concerned
to improve drug distribution and reduce stockouts as it was to
develop user financing. New fees were also used to make
services more accessible in Brazil (Lassner).
36
CHAPTER 6
DECISIONMAKING FOR COMMUNITY FINANCING
Most PRICOR-supported researchers designed their studies
so as to guide decisionmaking, as outlined in Chart 3.
(The
chart omits a few studies not designed for decisionmaking.)
Major design and operational issues faced included:
Whether or not to institute user charges
What to charge for?
Whom to charge?
What schemes to introduce?
What prices to set?
What services to include in the PHC program and
how to deliver them?
Lots of operational details, regarding both
financing and service delivery.
A variety of information gathering and analytical
techniques were employed to resolve these issues, but most
common for community-based systems were such relatively simple
methods as preference matrices and consensus-building.
Many PRICOR-supported researchers conducted formal
household or service provider surveys. These surveys, though
time consuming and costly, helped researchers to:
gain entree to the community, as in Liberia,
even when results were presented very simply,
as in the Philippines
convince skeptical policy-makers (in Honduras)
and health care providers (in Brazil/Lassner)
that people were already paying significant
sums for drugs and services; and
study the effect of new fee systems on health
care utilization, as in Zaire, or coverage, as
in Brazil (Lassner).
At least one researcher was unable to fully analyze his
laboriously collected data, however, and another remarked
informally that focus groups would probably have yielded as
much or more information with considerably less effort and
delay.
As a prelude to Chapters 8 and 9, Chart 4 shows the kind
of decisions that were made regarding resource generation
methods and the application of these resources for specific
goods and activities.
37
CHART 3:
DECISIONMAKING IN COMMUNITY FINANCING DESIGN STUDIES
WHAT KINDS OF DECISIONS
WERE MADE?
WAT INFORMATION
WAS GATHERED?
HOW WAS INFORMATION
GATHERED?
HOW WERE DECISIONS MADE?
WHO DECIDED?
Benin
Scheme selection and
price setting
Opinions and preferences
of community leaders;
community leaders; costs
of main program inputs
Discussions with village
health committees and
assemblies
Cost and utilization
analyses; consensus
building techniques
Researchers; health
staff; community
leadership
Bolivia/
Gonzalez
Scheme selection, price
setting, service
selection and delivery;
lots of operational
details, regarding both
financing and service
delivery
Opinions and preferences
of community leaders and
intended service users;
costs of main program inputs
Formal surveys of households Consensus building
and health care providers;
focus groups, nominal
group process, village
health committee meetings,
community assemblies;
literature reviews
Health staff; community
leadership
Bolivia/
Miller
Scheme selection,
service selection and
delivery
Opinions and preferences
of health care providers,
intended service users, and
program managers; health
care sources, utilization,
and expenditures
Formal surveys of households Study terminated before
solution development phase
and health care providers
N/A
Brazil/
Lassner
What to charge for,
whom to charge, scheme
selection, price
setting, service
selection and delivery;
lots of operational
details, regarding both
financing and service
delivery
Opinions and preferences of Formal surveys of households Complicated weighted
preference matrices
and health care providers;
community leaders, health
nominal group process; key
care providers, intended
informant interviews
service users, researchers,
and program managers; costs
of main program inputs;
disease incidence and pre
valence; health care sources,
utilization and expenditures
STUDY
co
00
Group of 10, plus
providers; researchers
used survey results to
represent community
residents; community
leaders could veto scheme
CHART 3:
STUDY
WHAT KINDS OF DECISIONS
WERE MADE?
DECISIONHAKING IN COMMUNITY FINANCING DESIGN STUDIES
WHAT INFORMATION
WAS GATHERED?
HOW WAS INFORMATION
GATHERED?
HOW WERE DECISIONS MADE?
WHO DECIDED?
Dominica
Costs of main program
Whether or not to
inputs; disease incidence
institute user charges.
and prevalence
Whom to charge? What
prices to set? Lots of
operational details,
regarding both financing
and service delivery
Review of health system
financial and utilization
data; key informant
interviews
Quantitative analyses
MSH consultants and MOH
counterparts
Dominican
Republic/
Udall
What prices to set?
Costs of main program
inputs; disease incidence
and prevalence
Review of existing survey
data
Quantitative (econometric)
modeling and analysis
Researchers, largely in
isolation (but MOH rejected
recommendations)
Haiti/AEDC
Who to charge? Scheme
selection, prices
setting, service
selection and delivery;
lots of operational
details, regarding both
financing and service
delivery
Opinions and preferences of
community leaders, health
care providers, intended
service users
Anthropological study; group Investigator analysis
discussions
Researchers
Whether or not to
institute user charges.
What to charge for?
Scheme selection and
price setting
Health care sources,
utilization, and
expenditures
Formal surveys of households Investigator analysis
and health care providers
Researchers (but MOH has
largely not responded)
UJ
vD
Honduras
CHARI 3:
DECISIONMAKING IN COMMUNITY FINANCING DESIGN STUDIES
WHAT KINDS OF DECISIONS
WERE MADE?
WHAT INFORMATION
WAS GATHERED?
Liberia
Whether or not to
institute user charges;
what to charge for;
whom to charge; scheme
selection, price
setting; lots of
operational details,
regarding both financing
and service delivery
Opinions and preferences
Formal household survey
of community leaders; health (not analyzed); village
care sources, utilization,
health committee meetings
and expenditures (data not
analyzed)
Mali
Whether or not to
Health care sources,
institute user charges; utilization, and
what to charge for; whom expenditures
to charge; scheme
selection; service
selection delivery
STUDY
4^
O
HOW WAS INFORMATION
GATHERED?
HOW WERE DECISIONS MADE?
WHO DECIDED?
Preference matrices
Community leadership;
community residents
(male elders)
Formal surveys of households Investigator analysis
and health care providers;
key informant interviews
Mex i co/
Macorra
What to charge for;
scheme selection and
price setting
Opinions and preferences
Formal surveys of households Investigator analysis
and health care providers;
of health care providers;
costs of main program inputs key informant interviews
Philippine/
Osteria
Whether or not to
institute user charges;
what to charge for;
whom to charge; scheme
selection; price
setting; lots of
operational details,
regarding both
financing and service
delivery
Opinions and preferences
of canmunity leaders,
and intended service
users
Formal household survey;
village health committee
meetings, and assemblies
*5
Consensus building
techniques
Researcher (a senior MOH
official)
Researchers
Community leadership
and residents
CHART 3:
DECISIONMAKING IN COMMUNITY FINANCING DESIGN STUDIES
WHAT KINDS OF DECISIONS
WERE MADE?
WHAT INFORMATION
WAS GATHERED?
HOW WAS INFORMATION
GATHERED?
HOW WERE DECISIONS MADE?
WHO DECIDED?
Senegal
Whether or not to
institute user charges;
what to charge for;
scheme selection;
service selection
and delivery
Opinions and preferences
of health care providers
and program managers;
costs of main program
inputs
Key informant interviews
Investigator analysis
Researchers (but MOH
ignored recommendations)
Sanai ia
Whether or not to
institute user charges.
Disease incidence and
prevalence; health care
sources, utilization, and
expenditures
Formal surveys of households Investigator analysis
and health care providers
Researchers
Swaziland
Whether or not to
institute user charges;
what schemes to
introduce; service
selection and delivery
Opinions and preferences
of community leaders,
health care providers,
intended service users,
and program managers
Forma) surveys of households Consensus building
and health care providers,
techniques
village health committee
meetings and assemblies; key
informant interviews
Researchers, Ministry of
Health, traditional
community leader
Zaire
What schemes to
introduce?
Costs of main program
inputs; health care sources
and utilization
Formal surveys of households Quantitative analyses;
and health care providers
investigator analysis;
interaction matrices
Researchers and Ministry
of Health
STUDY
CHART 4:
CROSS TABULATION OF SOURCES AND USES OF COMMUNITY FINANCING
SOURCES
FINANCING
USES
SERVICE
FEES
DRUG
SALES
Drugs
Benin,
Brazil
(Lassner),
Zaire
Dominica
Liberia
Mexico,
Philippines,
Somalia,
Zaire,
Thailand
RAFFLES,
FESTIVALS,
AD HOC
ASSESSMENTS
COST SHARING,
VOLUNTEER LABOR
REVOLVING
FUNDS
Dominica
Liberia,
Philippines
Thai land
COOPERATIVE
SUMMARY
India
Benin, Brazil(Lassner),
Zaire, Dominica,
Liberia, Mexico
Philippines, Somalia,
Zaire, Thailand, India
Brazil (Lassner)
Brazil (Lassner)
Construction;
maintenance
India
Supervision
Benin, Zaire
Curative
Services
Benin, Brazil
(Lassner)
Zaire
Thailand
Preventive
activities
Benin, Brazil
(Lassner),
Zaire
Thailand
Benin, Zaire, India
India, Bolivia Benin, Brazil (Lassner),
(Miller)
Zaire, India, Bolivia
(Miller), Thailand
Haiti
Benin, Brazil (Lassner)
Zaire, Thailand, Haiti
CHARI 4:
CROSS TABULATION OF SOURCES AND USES OF COMMUNITY FINANCING
SOURCES
FINANCING
USES
SERVICE
FEES
Dominica
Philippines
Capitaliza
tion (or loan
CHWS
DRUG
SALES
RAFFLES,
FESTIVALS,
AD HOC
ASSESSMENTS
COST SHARING,
VOLUNTEER LABOR
REVOLVING
FUNDS
COOPERATIVE
Dominica, Liberia,
Liberia,
Philippines
Bolivia
(Gonzalez)
Liberia
(unsuccessful);
Philippines
(unsuccessful);
Swaziland
Benin, Zaire
SUMMARY
Haiti
India
Benin, Zaire, Bolivia
(Gonzalez) Liberia,
Philippines
Swaziland, Haiti, India
4^
Thailand
Sanitation/
Nutrition
Sunnury
Benin,
Brazil
(Lassner),
Zaire
Dominica,
I Liberia
Liberia,
I Philippines
Mexico,
Phi 1ippines,
Somalia, Zaire‘
Thai land
Haiti,
Philippines
Thailand, Bolivia (Gonzalez)
Brazil (Lassner)
Liberia Philip—
pines, Swaziland
Dominica,
Liberia,
Philippines,
Thailand,
Haiti
Thailand, Haiti,
Philippines
India, Bolivia Benin, Brazil(Lassner)
(Miller)
Dominica, Liberia,
Mexico, Philippines,
Somalia, Thailand,
Bolivia, Swaziland,
Haiti, India, Bolivia
(Miller), (Gonzalez)
Box 1:
Matching Needs With Resources in Bolivia
Researchers from the Institute de Investigaciones Medico
Sociales (IIMS) studied ways to reduce attrition rates of
community health workers, known as Sanitarios Natives (SNs),
in Cochabamba, Bolivia. The objectives of the study were to
identify both the conditions under which communities would be
willing to support a health worker, and financing schemes that
would be feasible within an impovrished, hyperinflationary
economy. Investigators worked with community groups and
individuals and held open community discussions. Discussions
established that villagers were willing to pay health workers
in kind provided that the workers were reliable and worked
exclusively in their own communities.
After several sessions, investigators proposed that an
in-kind payment (in the form of either wheat or potatoes) be
collected from each family after the harvest each year. The
quota was roughly equivalent to US$ 7.00 per family regardless
of the family’s land holdings. Since the total revenue from
these in-kind payments (US$ 7.00/household/year ) would cover
less than half the SN’s current salary, the researchers
concluded that stable, community-supported health workers
could be provided only by training and deploying a lower level
health promoter at a reduced salary level. This scheme would
also allow for the expansion of services into areas where no
formal services had previously been provided.
The organization and collection of the quotas relied on
two community organizations: the sindicato (a traditional and
highly visible organization in rural communities) and the
health committee. The health committee, composed of a
representative from the sindicato and an elected mother, was
directly responsible for the collection, storage and marketing
of the in-kind payments. This scheme has already been
implemented in seven communities.
44
CHAPTER 7
MOBILIZING COMMUNITY PARTICIPATION
Most PRICOR-supported researchers faced five community
organizational questions:
(1) how intensively to involve
individual communities in scheme design and management; (2)
how to initiate community contacts and gain acceptance; (3)
what types of organizations to support or create for financing
purposes; (4) what kinds of management training and
supervision to give community residents; and (5) what level of
effort to devote to community work? Few researchers resolved
these questions through formal data collection and
decisionmaking, but the success of some studies was almost
certainly due in part to effective community work. This in
turn may have depended more on local leadership and consensus
building than on technical precision. Whether "research
based” or not, community work was a vital part of the solution
development process for many PRICOR-supported comunity
financing studies.
HOW INTENSIVELY TO INVOLVE INDIVIDUAL COMMUNITIES
Community financing by definition involves work with
individual patients and communities at the peripheral level,
even if only for the collection of user fees and drug charges.
Earlier chapters have shown that this work required knowledge
of the interests and capacities of different population
segments, whether organized as self-conscious groups or
distributed with little sense of group identity. PRICORsupported researchers, with a few exceptions, spent
considerable time working with individual health care users
and their group representatives, and the success of several
studies was clearly due to this work.
The basic question here was whether to involve affected
communities one by one in an intensive scheme development
process, or rather to work only with a sample and then
generalize. Intensive community by community work appeared
necessary in studies aimed at:
development of cost sharing schemes, that is,
schemes that spread health care costs among all
local residents, regardless of health status
(as in Bolivia and Swaziland),
development of new services or facilities with
which community residents were previously
unfamiliar (as in Benin, Brazil, and Liberia)
nurturing of community self-reliance as an end
in itself (as in Liberia and many of the Indian
cooperatives).
45
In contrast, three types of PRICOR studies gave less
attention to community organization or postponed scheme design
until later. These included:
■
studies in Honduras, Somalia, and Mali, because
they were aimed at national policy change
rather than at detailed scheme design
the revolving drug fund study in Dominica
because it operated at the national level, and
fee for service and drug sales schemes in
Dominica and Zaire because they were based on
individual user payments and were to be managed
by health center staff rather than by local
residents.
Several of the community financing strategies in Brazil
also required little community education or management because
they were based on user fees and charges and were to be
managed by system personnel rather than by community
organizations.
Chart 5 provides additional information on individual
studies.
HOW TO INITIATE COMMUNITY CONTACTS
This is an important question for those starting
community work, but did not concern many PRICOR-supported
researchers because most who did community work had wellestablished ties with local leaders and institutions before
their studies began.
These prior contacts almost certainly colored scheme
design and implementation. In Liberia, the principal
investigator knew a great deal about local leadership and
decision patterns and was himself known and respected; this
enabled him to make reasonable judgements without extensive
survey data. Rio’s slum communities, often suspicious or even
hostile to outsiders, were probably more receptive to CPAIMC
because of its previous work history in the area. Philippine
researchers also had considerable knowledge of local
communities, though not of the specific ones in which they
were working.
Prior experience of negative outside contacts can cause
communities to resist participation, as in parts of Rio de
Janeiro where police actions had created suspicion and fear.
In Benin, mothers were reluctant to participate in a growth
monitoring program because a previous project had provided
free food and this one did not.
46
I
CHART 5:
COMMUNITY PARTICIPATION IN PRICOR-SUPPORTED
FINANCING STUDIES
FINANCING
OBJECTIVE
INITIAL RESEARCHER/
COMMUNITY CONTACTS
ORGANIZATIONS ON WHICH
COMMUNITY FINANCING WAS BASED
NATURE OF
PARTICIPATION
Benin
To develop community
financing schemes
for specific
communes
Previous project work in
the area (but not in these
communities)
Communes
Community leaders and
residents were actively
involved in scheme design
and management
Bolivia/
(Gonzalez)
To reduce high CHW
attrition and increase
support within specific
communities
Sindicatos and health committees;
sindicatos are traditional,
represent every household but only
include women (ineffectively) if
male has died; health committees
are new, have only two members,
but at least one a woman.
The two groups together had
major decisionmaking and
management responsibilities.
Extensive interaction was
needed between staff and
residents to design a
program that the community
could support. Health
committee collected and
managed assessments
(potatoes).
Bolivia/
Miller
To develop specific
cooperat i ve-based
health care schemes
Was to have been based on
agricultural and marketing
cooperatives
Study terminated before
schemes could be implemented
Brazil/
• Baker
This was a retro
spective study of the
financing of CHWs and
water supply systems
STUDY
After schemes were
implemented
Fundacao Services Especiais de
Saude Publics (FSESP)
EXTERNAL
MANAGEMENT SUPPORT
CHART 5:
COMMUNITY PARTICIPATION IN PRICOR-SUPPORTED
FINANCING STUDIES
ORGANIZATIONS ON WHICH
COMMUNITY FINANCING WAS BASED
STUDY
FINANCING
OBJECTIVE
INITIAL RESEARCHER/
COMMUNITY CONTACTS
Brazil/
Lassner
To reduce dependence
on external funds
Implementing organization had Each community had a neighborhood
association, but most organiza
10 year history of work in
the area
tional work was done by CPAIMC
co
Dominica
To establish a national Conmunities not involved
revolving drug fund
that will reduce costs
and increase user
financing
Centra] Ministry of Health, no
cannunity groups
NATURE OF
PARTICIPATION
EXTERNAL
MANAGEMENT SUPPORT
Residents responded to
CPAIMC actively managed
survey questions and paid
both the program and its
fees; in a few cases,
financing, with only
neighborhood associations
limited role for
organized labor contribu
neighborhood associa
tion and fund raising
tions
activities. Researchers
used survey results to
represent community
interests, but only leaders
participated directly (at
some stages) in decision
making. A few associations
undertook clinic cleaning
or payment of utility bills.
One sold “carnetsMost
associations had a limited
implementation role,
however. Some neighborhood
associations were dominated
by unpopular or disreputable
leaders.
None directly
Scheme was entirely run
by the Ministry of Health
and its district offices;
Management Sciences for
Health provided extensive
design and management
assistance
CHART 5:
STUDY
FINANCING
OBJECTIVE
Haiti/
(AEDC)
To develop financial
incentives to motivate
(1) health workers to
teach child survival
skills, and (2) mothers
to learn them
Honduras
To develop national
health care financing
policies
INITIAL RESEARCHER/
COMMUNITY CONTACTS
Limited community role in
this study
COMMUNITY PARTICIPATION IN PRICOR-SUPPORTED
FINANCING STUDIES
ORGANIZATIONS ON WHICH
COMMUNITY FINANCING WAS BASED
NATURE OF
PARTICIPATION
Women's credit clubs; though
traditional in form, specific
clubs were created for health
project
Researchers and consultants
made major design decisions.
Credit clubs decided on
their own membership and how
much members should contri
bute each month.
N/A
Limited
kD
India/
(Elkins)
To analyze the
After schemes were
successes and failures implemented
of 8 coop-based health
care projects
Women's clubs, youth groups, milk Variable
cooperatives, leather-working
cooperatives, church groups,
village councils, and students
enrolled in evening vocational
programs. New organizations,
several of them just for health
activities, were created in seven
of the eight communities. In one
study area, village health and
development promoters conducted
bi-weekly meetings of program users
in lieu of creating a forma)
organization.
EXTERNAL
MANAGEMENT SUPPORT
CHART 5:
STUDY
Liberia/
(Cole I)
FINANCING
OBJECTIVE
INITIAL RESEARCHER/
COMMUNITY CONTACTS
ORGANIZATIONS ON WHICH
COMMUNITY FINANCING WAS BASED
NATURE OF
PARTICIPATION
EXTERNAL
MANAGEMENT SUPPORT
To strengthen the role
of 5 specific
ccniDunities in PHC
financing and
management
Principal investigator was
the district health officer
and had been born and raised
in the area; these factors
clearly expedited design and
implementation.
Village health committees
Committee of 10 to 15 male
elders developed preference
matrices, made many of the
design decisions, collected
and managed revenue (except
in one convnunity), and
selected the CHW.
Outside supervision was
needed to resolve
conflicts between health
committees and town
councils regarding
appropriate use of funds.
Community chose health
services and financing
means: which drugs, who
would manage botikas,
prices. Residents
participated as botika
managers, fund collectors,
and lead mothers. Women
were well represented in
decisionmaking.
Researchers deliberately
limited their role to
providing impetus,
leading workshops and
various training
sessions, though
presence reportedly
helped study stay on
course. Researchers
also audited financial
records.
Philippines To develop self
financed PHC
activities
tn
in 6 barangays
o
(communities)
Swaziland
COMMUNITY PARTICIPATION IN PRICOR-SUPPORTED
FINANCING STUDIES
To develop community
support for rural
health motivators
Researchers did simple house- Barangay and purok associations,
hold survey and used results Core group of existing local
leaders and health workers.
as basis of discussion with
village assembly
Community as a whole (especially
traditional chief and runner)
Little needed except
Community preferences
influenced most aspects of to ensure that CHWs had
design. Researchers mainly appropriate skills,
worked with chief who then
consulted residents. Chief
donated land, community
donated labor. Chief was to
have managed community labor
but was away at the critical
time. Community preferences
led to some strengthening
of CHW skills.
CHARI 5:
STUDY
COMMUNITY PARTICIPATION IN PRICOR-SUPPORTED
FINANCING STUDIES
F1MARCIRG
OBJECTIVE
INITIAL RESEARCHER/
COMMUNITY CONTACTS
ORGANIZATIONS ON WHICH
COMMUNITY FINANCING WAS BASED
NATURE OF
PARTICIPATION
EXTERNAL
MANAGEMENT SUPPORT
To study the strengths
and weaknesses of a
sample of approx
imately 18,000
revolving drug,
sanitation, and
nutrition funds.
Researchers contacted retrospectively. Tarobon health
officers played a central
role in the initiation of
most local funds.
Community residents purchased
shares in revolving funds, newly
created in most cases. Study
found that multi-purpose funds
were more successful than
single purpose ones.
Fund managers and commit
tees handled virtually all
aspects of fund management,
including price and inter
est rate setting, approval
of loan applications,
collection and use of
revenue. Share ownership
was widespread for drug
revolving funds because of
low share price and wide
need for drugs. Sanitation
and nutrition funds had
more limited membership.
No management training
was given, but tambon
health officers were
active and made signi
ficant contribution to
fund management in
many cases.
To assist decentral
ized health zones to
develop self-financing
schemes.
Apparently little if any
contact with either the
health center staff or the
community prior to scheme
implementation.
Individual health center staff.
Communities were not directly
involved.
Virtually none provided.
Both health center staff
and community residents had
significant role in shaping
study results but in a
reactive sense only since
they were not consulted
beforehand.
Thailand
tn
Zaire
%
♦
!------OV
0 X '
o0) 3^
3 •
0
n
C
n
*\w
Q)
□
Q.
IS
Most researchers met first with local leaders assumed to
represent the community in which they wished to work. In some
countries, the form and process of this contact may have been
mandated by custom: in Liberia, for example, community
affairs are largely controlled by the traditional, all-male,
council of elders, and it was essential to meet with it. In
Swaziland, permission and preferably active support, had to be
obtained from the village chief prior to the holding of a
comnmunity assembly.
In Brazil, the original principal investigator developed
a first visit protocol indicating what was to be said. The
focus was on the study problem, namely, dependance on external
funds, and the community’s possible willingness to consider
self-financing. In the Philippines and Brazil, the first
contact secured leadership concurrence for a health care
utilization and expenditure survey. In Thailand, district
health officers made many of the first official contacts,
usually with a group of village leaders.
FINDING OR CREATING A SUITABLE COMMUNITY ORGANIZATION
Schemes intended to be community-managed obviously
require a local organizational base, whereas those that are to
be managed by health center staff (as in Dominica) may not.
For those seeking community management, identification or
creation of a suitable organization was an important
operational problem.
Two significant issues for some researchers were (1) what
sorts of community organizations are suitable for undertaking
community financing? and (2) should health activities be based
on existing organizations or should new ones be created?
Researchers in Bolivia, India, and Thailand studied the first
question retrospectively by analyzing existing activities with
community financing elements. The Indian analysis of coop
based health care found that organizations differed along
several dimensions:
History: some organizations predated the health
activity while others were created especially
to do health work.
Relationship to other local organizations: some
implementing organizations were closely related
to village councils or other local bodies while
others were relatively independent.
Purpose: some were single purpose while others
had multiple purposes.
Number of hierarchical levels: some
organizations had multiple levels,
corresponding to levels in the health delivery
52
system while others operated only at the
community level.
Breadth: some organizations represented both
genders and all significant socioeconomic
sectors while others had a narrower membership.
In addition, community organizations differed in the
depth of their participation in problem analysis, solution
development, and implementation.
Existing organizations studied in India included women’s
clubs, youth groups, milk cooperatives, leather-working
cooperatives, church groups, village councils, and students
enrolled in evening vocational programs. New organizations,
several of them just for health activities, were created in
seven of the eight communities. The Jamkhed and Ambilikkai
projects created multipurpose women’s clubs and youth groups,
while the Jaipur Trust planned to develop new, single-purpose,
village health committees to be associated with the local milk
cooperatives. In the Kangazha project area, village health
and development promoters conducted bi-weekly meetings of
program users in lieu of creating a formal organization. The
Tribhuvandas Foundation created new single purpose health
committees based on the established milk cooperative. Looking
at the eight projects together, the Indian study concluded
that "it is at least easier and faster to start with existing
organizations. . . . Several of the projects . . . have
achieved remarkable breadth and depth of community
participation through new organizations, but they have
achieved these levels of participation after long periods of
effort.”
A concern for some researchers was the domination of
existing groups by men and/or other elites. In Bolivia
(Gonzalez), managers decided to create new health committees
because the traditional sindicatos included only men. Study
participants in Swaziland suggested that health assemblies be
restricted to women to give them a stronger voice, but it was
eventually decided that men must participate as well because
of their control over family finances.
(Women were encouraged
to speak out, however.) Benin decisionmakers also made
special efforts to include women. In Rio, some slum
communities were dominated by criminal elements, and in the
Philippines, barangay and purok association were controlled by
socioeconomic elites; researchers in both cases nevertheless
decided that health activities had to respond to existing
power structures so they did not create new groups.
LEVEL OF EFFORT
Researchers in Benin, Bolivia (Gonzalez), Liberia, the
Philippines, and elsewhere made numerous visits and allowed
significant time to elapse in order to obtain design
53
decisions, or at least expressions of interest, from
participating communities. These efforts served the dual
functions of (1) improving the "fit" between scheme design and
community interest and capacity, and (2) educating and
motivating residents to support PHC activities later on.
Intensive community by community discussions also encouraged
local residents to ’’buy in” to PHC and financing, giving them
a sense that results were partly of their own design.
In Benin, researchers held a series of meetings with
community leaders to discuss financing alternatives and to
establish modes of payment, price levels, VHW remuneration
arrangements, and membership of commmunity committees for
financial management. These consultations were particularly
fruitful because they increased public understanding, hence
financial support, for preventive care and supervision. The
principal investigator in Liberia made two pre- design visits
to participating communities, the first to explain the study
and seek cooperation, and the second to obtain nominations for
the health committee and for health worker training. During
subsequent visits, health committee members constructed
preference matrices which were then used for detailed scheme
design. Working first through assemblies and then through
leadership groups, researchers and residents in the
Philippines chose the health services to be financed, the
specific drugs to be provided, and the individuals who would
manage boticas. CPAIMC staff in Brazil mainly designed
financing schemes themselves using survey results, but
community leaders were given veto rights.
TRAINING AND TECHNICAL SUPPORT
As researchers had intended, many commmunity groups took
an active role in scheme management. Roles performed by one
or more groups included:
consultation on CHW salary increases (Benin)
■
monitoring of CHW activities (Benin)
consulation on service changes (Benin)
revenue collection and management (Liberia,
Philippines, Benin, Bolivia/Gonzalez)
pressure on community residents to get them to
pay overdue clinic fees (Benin), drug charges
(Philippines), or assessments
inventory management and restocking (Thailand,
Philippines, Liberia)
54
provision of labor for CHW fields (Liberia,
Swaziland), or for clinic cleaning and
maintenance (Brazil/Lassner)
payment of utility bills, provision of office
and clinical supplies (Brazil/Lassner).
Local committees administering a revolving loan fund for
water supply in Brazil (Baker) managed loans, maintained
financial records, collected monthly payments from
participating families, and delivered these payments to the
program office. In the Philippines, community residents
managed botica inventories and restocking, collected funds,
and served as lead mothers; they did not, however, conduct
audits as planned becaused they preferred that the researchers
do this.
In Thailand, sanitation funds were managed by local
committees, whose members averaged 44 years old and had a
modest income. Committee members set share prices and
interest rates and in some cases decided who would receive
loans. Elected or appointed fund managers were responsible
for supervising construction, keeping records, controlling
inventory and collecting loans. Similar roles were played by
community managers and committees in other types of revolving
funds.
Mothers’ credit clubs established in Haiti managed their
own funds, and determined how much was to be collected and
loaned out each month.
Even though community organizations did play active
roles, few had the technical capacity or (in some cases)
motivation to develop community financing without signicant
outside support. In many study areas, help was particularly
needed for:
increasing awareness of primary health care and
the community’s role in it (called "cultural
preparation" in Thailand)
developing administrative and managerial
capacities
maintaining water supply and other physical
systems (Brazi1/Baker and Brazil/Lassner)
supervising the use of funds (Liberia) or
auditing (Philippines)
generally keeping work on course (Philippines).
55
Examples of outside technical support include:
■
training in maintenance of water supply systems
in Brazil (Baker)
training in management of revolving funds
(Brazil/Baker)
a series of training programs and workshops for
lead mothers and fund managers in the
Philippines.
Training was not provided in Thailand, but investigators
found that the most successful communities had had prior
experience in managing revolving funds, while communities
lacking such experience were notably less successful.
The management roles of researchers was often significant
in spite of their best efforts to provide only initial impetus
and limited technical support. In Liberia, the principal
investigator resolved a dispute between the health committee
and the town council about the proposed use of revolving fund
monies to pay for a funeral. This sort of use for community
funds was traditional in the study area but clearly would have
threatened RDF capital. In Bolivia, outside encouragement was
needed to ensure that potatoes were collected at harvest time
so that CHWs could be paid. Researchers in the Haiti/AEDC
study provided a great deal of external motivation and
education in order to create the new women’s credit groups.
Among other things, they conducted 16 3-day seminars in 11
villages with 801 participants. In the Philippines,
researchers reported that their continued presence helped to
keep scheme activities on course. ’’Research” activities of
this nature may continue to be necessary when study results
are replicated elsewhere.
56
CHAPTER 8
RESOURCE GENERATING METHODS
DRUG SALES
As detailed in Chart 6, PRICOR-funded researchers in 9
countries either documented existing drug utilization and
supply systems or assisted PHC managers to devise and pay for
new systems. Five of these studies were national in scope,
and five focussed on specific local schemes.
(The Thailand
study appears in both counts because the local schemes that it
analyzed existed nationwide.) Studies in Dominica, the
Dominican Republic, Liberia, the Philippines, and Thailand
specifically sought to design new drug financing methods; in
all but the Dominican Republic, these were revolving drug
funds (RDFs). This chapter concentrates on RDFs.
A revolving fund is essentially a financing scheme in
which, after the initial capital investment, supplies are
replenished with funds collected from the sale of the
supplies.
Capitalization :
A startup issue for RDF researchers in Thailand,
Dominica, Liberia, and the Philippines was how to obtain an
initial drug supply or to raise funds with which to purchase
it. Means employed included assessments of community
residents (Philippines, Liberia and Thailand), sale of shares
(Thailand), donations (Liberia), raffles and festivals
(Philippines), and loans (Dominica). In Dominica, the
national RDF was capitalized by an EC $500,000 (US$187,000)
loan from the Social Security fund. External agencies donated
initial drug supplies to many of the Indian and Bolivian
cooperatives studied by Elkins and Miller. In Thailand,
communities supplemented government-donated drug supplies, as
described in Box 2; the most successful communities generated
ten times as much as the government gave them. The Thai funds
also used RDF profits to capitalize further expansion.
Communities completely capitalized the Philippine and
Liberian drug funds through combinations of ad hoc
assessments, monthly payments, and donations. In Liberia,
village groups levied one-time assessments and collected
between $59 and $209 each. Between 75 and 90 percent of
village households participated. One missionary family donated
$300. Using similar means plus raffles and festivals,
Philippine communities generated 15 to 57 percent of their
capitalization targets, with participation rates ranging from
2 to 88 percent of households. Some funds were forced to
lower their ambitions because collection of assessments proved
to be difficult.
57
CHART 6:
STUDY
OBJECTIVE
SCOPE
Bolivia/
Miller
To develop
coop-based
health
services
National
Dominica
To increase drug
National
financing and
reduce costs
through centralized
procurement,
improved
distribution and
user payments
cn
oo
Dominican
Republic
To establish price
structure for ORS
Ecuador
To improve CHW
performance and
stability
PRICOR STUDIES THAT DOCUMENTED, OR HELPED TO ESTABLISH, DRUG SALES SYSTEMS
WHAT WAS
DOCUMENTED?
WHAT WAS SET UP?
CAPITALIZATION PRICING
The current health Study was terminated Mostly from
before results were external
activities and
financing of 12
implemented
donations
cooperatives
Current drug
utilization and
procurement
National ORS need and
utilization
Management
difficulties in
CHW-managed RDEs;
costs of RDFrelated storage,
transportation,
salaries and
travel allowances
SALES MANAGEMENT AND
REVENUE COLLECTION
Varied
from one
coop to
another
Variable
By health staff;
user charges not
to be introduced
until procurement and
distribution systems
are fully operational
RESULTS
National drug
procurement and
pricing system, with
drug costs charged
to district health
budgets
Loan from
social
security
fund
251 over
cost
Pricing strategy
recommended but not
implemented
N/A
Two-tier
N/A
pricing
recommended
Not implemented
Based on
drug costs
alone,
excluding
storage,
transport
ation, and
other
associated
costs
Subsidized associated
costs totaled 501 to
1001 of drug
wholesale costs;
deficits subsidized
by FODERUMA; it
might have been
cheaper for FODERUMA
to subsidize private
sector distribution
N/A
By CHWs
Dominica model
being applied in
other Eastern
Caribbean
countries under
major AID contract
CHART 6:
STUDY
OBJECTIVE
SCOPE
Liberia
To establish and
finance communitybased drug sources
Philippines To develop CF
schemes
PRICOR STUDIES THAT DOCUMENTED, OR HELPED TO ESTABLISH, DRUG SALES SYSTEMS
WHAT WAS
DOCUMENTED?
SALES MANAGEMENT AND
REVENUE COLLECTION
WHAT WAS SET UP?
CAPITALIZATION PRICING
5-10
Health care
villages utilization
patterns
5-10 RDFs
One-time
assessments
of community
residents,
plus
donations
5 com
munities
5 RDFs
Variable
15—25X
Variable (CHW, lead
(assessments, markup,
mother, village
donations,
based partly captain)
raffles,
on market
festivals)
retail
prices
5 RDFs still
functioning after
one year
N/A
Variable
markup,
determi ned
by village
health
commi ttees
Sold by committee
member or health
worker
Investigator recom
mended private sector
distribution methods
N/A
N/A
N/A
Senegal
To identify possible Regional Existing revolving N/A
funding sources
drug funds
to cover recurrent
costs
Somalia
To develop strat
egies to improve
PHC drug supply in
rural areas
National
Household drug
expenditures and
sources of supply
N/A
25X above
wholesale
CHW or committee
treasurer
RESULTS
RDFs started in 5
communities but
spread to nearby
areas; researchers
considered that
their most important
achievement was
creation of community
awareness and self
confidence
CHART 6:
PRICOR STUDIES THAT DOCUMENTED, OR HELPED TO ESTABLISH, DRUG SALES SYSTEMS
WHAT WAS
DOCUMENTED?
WHAT WAS SET UP?
STUDY
OBJECTIVE
SCOPE
Thailand
To identify, eval
uate, and replicate
effective financing
alternatives
National; Prevalence of RDFs, A number of proto
22 funds range of types and type multipurpose
studied functions; factors revolving funds
in depth associated with
success and failure
O1
o
CAPITALIZATION PRICING
MOPH, sale
of shares,
donations
and assess
ments
30X over
wholesale
SALES MANAGEMENT AND
REVENUE COLLECTION
Variable (CHW, head
man, member of RDF
management committee)
RESULTS
Numerous profitable
funds; profits
distributed to share
holders or used for
compensation of RDF
manager or investment
in additional drug
supplies; government
is encouraging
extension into
nutrition and
sanitation activities
Box 2:
Sale of Shares for Capitalization in Thailand.
In Thailand, capital was raised by sale of shares in the
revolving fund. Share prices were set by the management
committee. In case study funds, prices ranged from 10 to 50
Baht a share, and shares were owned by 30 to 100 percent of
village households. Households could own more than one share
but an upper limit (for example: of 50 shares per household or
25 percent of all shares) was sometimes set by the committee
to prevent a single family’s taking over the fund. Poor
families unable to purchase shares were sometimes able to buy
shares on credit or were provided shares in exchange for labor
contributed to construction or improvement of the drug fund’s
stores. When share ownership was widespread and/or multiple
shares were owned by households, the amount of capital raised
was impressive. The initial capital raised in each community
exceeded the amount provided by the Ministry and in the most
successful funds was ten times as much.
Differences in village size, wealth, proportion of
households purchasing shares, and share price accounted for
some of the inter-village variation in capital raised. The
MOPH offered an incentive of an additional 1,000 Baht for any
drug fund with share ownership by 80 percent or more of
households. Individual reasons for buying shares included
wanting to help the village have the benefits of a drug fund
(71.5%), and respect for the individuals setting up the fund
(11.0%). Both responses were measures of social obligation.
A few shareholders also reported low share price as a reason
for purchase.
Statistically, villages with the highest percentages of
households owning drug fund shares were small, with relatively
equitable distribution of land ownership and household income.
Their drug funds had low share prices and were managed by
persons of relatively modest income. These villages also had
fewer funds supporting other activities and making social
obligation claims on households.
61
Pricing:
For funds to ” revolve", monies recovered from drug sales
must cover the cost of replenishing drug stocks and, in some
cases, such operating costs as transportation and
administration as well. Most PRICOR-funded researchers priced
drugs at a percentage markup over initial wholesale costs:
25%, for example, in the cases of Dominica and Liberia, and
30% in Thailand. Some RDF managers in the Philippines applied
a flat markup, while others adjusted prices so as to be near
or slightly below private sector prices. Program managers in
Benin discovered that their prices for malaria treatment
exceeded the private sector’s, so they reduced their prices
accordingly. Most,percentage markups reflected original
purchase costs rather than restocking costs; this worked
adequately in most countries with two significant exceptions:
in Ecuador, prices did not reflect storage, transportation,
and other associated costs, estimated to total 50 to 100% of
drug wholesale costs; because of the resulting deficits, it
might have been cheaper for FODERUMA to subsidize private
sector distribution. Problems were reported in Bolivia
(Miller) as well because of hyperinflation; at least some
funds failed because drug costs rose more rapidly than prices.
None of the revolving funds reported here adopted sliding
scales or offered exemptions for certain persons or drugs.
The Dominica Ministry of Health briefly considered exempting
the indigent, those under 15, those over 65, and those with
chronic illnesses, but a brief study showed that the young and
the old alone constituted 45% of health center utilization.
Researchers in the Dominican Republic recommended a two-tier
ORS pricing strategy, but no action was taken.
Activity and Financial Management
Drug Procurement
Two issues were common in drug procurement, namely, which
drugs to buy, and where to get them. A related issue was
whether to buy drugs by brand name or in generic form. In
Dominica, RDF managers selected items from the national drug
formulary and sought generic suppliers as a way to reduce
costs. In Liberia, drugs were purchased wholesale from the
Kolahun Health Cooperative, which in turn obtained them from
the Christian Health Association of Liberia; selection was
limited to essential generics. Researchers in the
Philippines, on the other hand, found that people strongly
prefered brand name drugs and attributed part of their success
to sale of familiar items; these were purchased from private
sector drugstores in the nearest town. Thai RDFs were
resupplied by the Government Pharmaceutical Organization which
maintained stocks at each provincial health office.
62
Drug Sales and Inventory Maintenance
Most PRICOR-developed revolving funds, except the one in
Dominica, were managed by community residents. In the
Philippines each depot had an appointed manager, who was, in
different villages, the barangay health worker, a ’’lead
mother”, the barangay captain, a council member or a core
group member. Managers were responsible for day to day
transactions. In Liberia, sale and restocking was the
responsibility of the village health worker. In Liberia, four
of the funds were managed by the village health worker and one
by the village health committee’s elected treasurer.
Management of individual PDFs in Thailand was overseen by
committees of 10-12 shareholders. Committee members
represented leading families and included monks, school
teachers, headmen, health and community development
volunteers, and ’’communicators”. Successful funds, in
addition, tended to have near fulltime managers, serving
either without pay or with minor compensation. The Village
Health Volunteer was responsible for sales and inventory
management, while daily receipts and working capital were
sometimes entrusted to a monk. The most common problems were
incomplete records and inadequate accounts. The most
successful PDFs were managed by persons experienced in other
community funds, particularly ones which sold goods on a
revolving fund basis. Tambon Health Officers acted as
consultants to most of the PDFs, and some funds received
management assistance from nearby district hospitals.
Success of PDFs depends in large part on good financial
and material management. In Dominica, a computerized system
of inventory control was designed and bin cards were used to
record disbursements, replacements and the cost of each item.
The Central Medical Stores (CMS) maintained drug stocks, while
the Ministry of Health accounts office managed financial
affairs. This separation of material and financial management
required close coordination and a well developed information
system.
In the Philippines, fund managers were trained to manage
and keep records, but many were still weak even after a year
of operation. A formal system was devised for monitoring
stocks, but most depot managers simply watched their inventory
and restocked when it was low.
Credit
Managers of all five Philippine PDFs offered credit to
those unable to pay at a time of need, though one fund limited
credit to persons who had paid membership dues. Three of the
funds reported that debts were generally repaid within 15-30
days. In one village the village captain sometimes paid for
those who were unable to pay back their credit. In another
63
village, credit purchases ran rampant. Credit occupied 30-50
percent of the sales with an average of one to two months
required for collection. In Liberia, researchers decided that
it was too difficult to decide who could and could not pay and
that relatives must pay if the patient could not.
Supervision and Audit
PRICOR experience suggests that some community or outside
oversight is needed both for routine supervision and for
auditing. In Liberia, elders in one village wanted to use RDF
proceeds to cover funeral costs for a prominent chief. The
VHW refused, and a research assistant had to step in to
prevent decapitalization. A similar problem appeared in
Dominica where interest earned on RDF monies was spent on
items that had not been budgeted for. Researchers in Dominica
worked with the government to develop procedures for annual
closing. In the Philippines, most villages appointed an
auditor, but several ended up entrusting this function to the
PRICOR researchers.
SERVICE FEES
PRICOR studies in Benin, Brazil, and Zaire designed or
refined service fee systems, while the India/Elkins study
documented existing practices in 8 cooperatives. Data on
service fees are also available from a PRICOR-supported study
in Mali. This chapter concentrates on the decisions made in
the solution development studies in Benin, Brazil, and Zaire.
Chart 7 provides details.
Pricing
What to Charge For?
In each case, researchers/managers had to decide what
specific services to charge for and how to "package” them.
Fee arrangements varied from one community to another in
Brazil. Some clinics charged for selected types of visits
(all physician visits, all or annual family planning visits),
others for such selected services as pap smears, injections,
blood pressure checks, first aid, pregnancy tests, and IUD
insertion. Still others charged registration fees. Various
combinations of these service fee schemes were implemented charging for some preventive and curative services while
offering others free of charge.
In Benin and parts of Zaire, curative care was charged
for as episodes rather than as discrete visits or services.
Patients paid a single fee at their first visit for a given
condition and then received all related care free of
additional charge. In Zaire, five health centers continued
existing fee per visit systems and five collected fees per
episode; analysts hoped to study effects of payment scheme on
64
CHART 7:
STUDY
REVENUE
GENERATION
OBJECTIVE
SCOPE
To cover all 4
cojinunes
recurrent
health center
costs,
including
those
attributable
to super
vision and
logistics
(JI
Brazi 1/ An unspe9 urban
lassner cified degree slums
of cost
sharing
(rather than
full coverage
of specified
costs)
i
PRICOR STUDIES THAT DOCUMENTED, OR HELPED TO ESTABLISH, FEE FOR SERVICE SYSTEMS
WHAT WAS
DOCUMENT- WHAT WAS
ED?
SET UP?
Extensive Fee for
cost and episode
and util- schemes in
ization 4 communes
data
HOW WERE
PRICES SET?
WHAT SERVICE
UNITS WERE
CHARGED FOR?
WHAT DID
FEES COVER?
SLIDING
SCALES,
AND
EXEMPTIONS
REVENUE
COLLECTION
(CREDIT,
RECORD
KEEPING)
USE
OF
FUNDS
RESULTS
To pay for
Non-residents, Cash box
Detailed cost Set individual- All visits
transactions
drugs, VHW
and
those
who
ly
for
each
and
drug
use
analysis
supervised remuneration,
bypassed
major condition connected
based on
by financial personal pre
expected drug and age group with a single required
management
ventive care,
use and number
condition
referral
subcommittee
health center
channels
of visits per
maintenance
paid four
of the
episode; see
Commune
costs, fuel
Box 3 for
times the
for supervision
normal fee
Health
details
and mobile MCH
Commi ttee
team, and village
level information
system; followup
visits
Varied from
Variable
Health
Comnunity Intuitive
care
financing judgements,
community to
community (set
util
strategies based on
costs, market per visit or
ization tailored
service rather
patterns, to
prices, and
types of individual users’ ability than per
service communities and willing episode)
ness to pay;
that
reconsidered
people
are used
every 4 months
because of
to paying
for; in
inflation
volvement
with com
munity
organ
izations
Fees adjusted
for 0.41 of
visits, based
on employment
status,
conditions at
home,
financial
factors
To defray overall All CF
Credit
schemes
extended if CPAIMC costs
together
needed; fees
covered 201
for 6.11 of
of direct
visits never
costs; in
collected
conclusive
effect on
utilization
CHART J:
STUDY
REVENUE
GENERATION
OBJECTIVE
India/ Varied From
Elkins one coop to
another
SCOPE
PR1C0R STUDIES THAT DOCUMENTED, OR HELPED TO ESTABLISH, FEE FOR SERVICE SYSTEMS
WHAT WAS
DOCUMENT- WHAT WAS
SET UP?
ED?
8 coop- Existing N/A
eratives service
delivery
and
financing
efforts
2 pro
vinces
HOW WERE
PRICES SET?
WHAT SERVICE
UNITS WERE
CHARGED FOR?
WHAT DID
FEES COVER?
Variable
Variable
Variable
To develop
PHC support
mechanisms,
especially
conuunitybased
Zaire
Full coverage 10 health Extensive Fee for
of specified centers cost and episode
schemes in
health center (5 zones) utilrecurrent
i zat i on
4 centers
costs,
including
those for
logistics,
supervision,
and drugs
No standard
method though
costs usually
considered
In 5 centers,
fixed fees per
episode; other
5, fee per
visit
REVENUE
COLLECTION
(CREDIT,
RECORD
KEEPING)
All fees
Variable
were adjusted
for family
income
Variable
Variable
Existing N/A
service
delivery
and
financing
efforts
Mali
SLIDING
SCALES,
AND
EXEMPTIONS
Episode fees
covered all
associated
visits and
drugs; visit
fees covered
only a single
service
USE
OF
FUNDS
RESULTS
Variable
In one "cercie",
revenue covered
costs of drugs,
petrol, and
furniture
Strong com
munity and
staff
reactions
to changed
fee systems;
mixed
effects on
utilization
and costs
PHC costs and utilization. In Benin, a fee for episode scheme
was implemented in health units in four areas. The Zaire and
Benin results suggest consumer preference for episode-based
fees, provided they are affordable, though disgruntled staff
in some Zaire centers obstructed their full implementation.
Confounding factors unfortunately prevented an objective
comparison of visit and episode fees in Zaire.
Program managers charged for some preventive care in both
Zaire and Brazil. Even though preventive interventions were
free of charge in Benin, those costs were foreseen and added
to the cost of curative care.
How Much to Charge?
In setting fee levels, PRICOR researchers considered: (1)
the magnitude of costs that had to be covered; (2) the ability
and willingness of users to pay, and, (3) the fee levels of
other local health care providers.
(The second and third
factors were obviously closely related.) As noted in the
chart, PRICOR researchers in Benin and Zaire sought to fully
cover certain PHC costs, while Brazilian researchers sought
cost-sharing only. The former, in other words, needed good
cost data on which to base prices, while the latter needed
only a general understanding of magnitudes, Only in Benin was
a technical method used for price setting.
Prices in Brazil reflected results of a survey of prices
charged for similar services in neighboring communities and an
analysis of CPAIMC’s costs. Researchers and managers agreed
that prices should not be lower than actual service costs nor
higher than local market prices; that low income families
should be able to afford them, both individually and in common
combinations; and that prices should be adjusted periodically
to reflect increasing costs and the government-mandated
minimum wage. Though communities varied in the services
charged for, they did adopt uniform prices for those fees that
they did collect.
In Zaire, fees were to reflect drug and service delivery
costs and the population’s economic status. The mode of
setting them varied from one health center to another, in some
cases giving greatest weight to the community’s preferences,
in others to the staff’s. Disputes about fee levels led to at
least one staff resignation. Some per visit fees may have
been negotiated by the worker and patient directly.
In Benin, prices were based on drug costs for specific
illnesses and age groups, plus a percentage of the recurrent
costs of petrol, VHW remuneration, and certain minor items.
The percentage mark-up was calculated by estimating the number
of cases of the disease and the proportion of personnel time
devoted to its treatment. The price of curative care also
included a proportion of preventive care costs not directly
67
charged for. Prices were calculated by the same method for
each disease, but varied with patient age since this affected
drug utilization. Sophisticated cost and utilization
information systems were required for these estimations, but
the end result was that most treatment fees were set at
approximately three times the cost of the basic drugs that
were used. See Box 3 for further details on price
calculations in Benin.
Activity and Financial Management
Fee Introduction
In Brazil, new fees were introduced for previously free
services, while in Zaire, the most important changes were from
fees for service to fees for episode. In Benin, service fees
applied to new services in previously unserved areas. Staff
and community reactions naturally varied from one place to
another.
In Zaire, reactions to the shift in payment scheme were
strong and unexpected. Several health personnel preferred the
old fee for service system because it gave them some control
over price levels. In one health center, the auxiliary nurse
was dismissed after showing open hostility; he then set up a
private dispensary and drew away most of the center’s clients.
In another, open conflict arose between members of the village
development committee and the auxiliary nurse. One health
post was closed for a period of time, and in some posts staff
worked less productively or with a lower standard of care.
Community members, on the other hand, preferred the fee for
episode system because prices were publicized and not subject
to bargaining; utilization of the one health center that
changed from episode to visit fees dropped significantly
because of the public’s reaction.
Fee Adjustment
In each of the seven Brazilian communities charging
service fees in, allowances were made for those too poor to
pay. Free samples of commonly used drugs and contraceptives
were retained for distribution to the indigent. A sliding
scale was established, and every client unable to pay was
asked about conditions at home, employment status, family
income and expenses, and financial difficulties. Depending on
circumstances, discounts ranged from 20 to 100 percent.
During the nine month implementation stage, only 0.4 percent
of visits were affected by fee reductions or exemptions.
Clients were also allowed to postpone payments if they did not
qualify for exemptions; debts were not repaid for 6.5 percent
of visits. Analysis showed that actual revenue over the nine
months was 10% less than potential revenue without fee
exemptions, discounts and credits, a difference that
researchers considered insignificant.
68
Box 3.
Factors Considered in Setting Fee Levels in Benin
Costs Included:
VHW renumeration
Equipment depreciation
Gas for the mopeds
VHW and community health center supplies
Electricity and other general operating expenses
Costs of preventive services and health education
Factors Considered in Distributing Costs:
Incidence and treatment of frequently seen illnesses
Percentage of personnel time and travel used in the
treatment of each disease
Desire coverage - 100%
Population size - 10,000
Adjustments:
Adjustment for patient age based on likely drug use
and number of repeat visits for the same condition
An across-the-board increase of 50% to provide incentive
for people to choose a prepayment scheme
Quadrupling of charges for persons residing outside the
project area
69
All eight projects in India adjusted service fees
according to the ability of people to pay. Some project
hospitals charged up to ten times more for wealthy persons.
In Benin, fees were increased by a factor of four for
people who wished to use project services but lived outside
the project area. This was done to limit the influx of
outsiders and to prevent them from obtaining and selling the
project’s low cost drugs; it also had a significant effect on
project revenue. ’’Outsider” treatment fees were also applied
to project area residents seeking care at the health center
without a proper referral; the objectives in these cases were
to increase the stature of health workers and encourage people
to follow the established hierarchy.
Revenue Collection and Management
In Benin, receipts from the health center and VHWs were
placed in a cash box under the control of a financial
management sub-committee of the Commune Health Committee; this
sub-committee included three community leaders and the health
center’s head nurse. In Brazil the personal prepayment,
personal service fees, and drug sales were managed by the
CPAIMC service providers and managers who normally operated
the health units and miniposts.
When new charges were instituted for previously free
services, problems occurred with collection from some people
who either could not or did not want to pay. As mentioned
earlier, some programs extended credit to those temporarily
unable to pay, but those who did not want to pay presented
different problems. In one of the centers in Brazil, the
staff at first felt uncomfortable charging patients with whom
they had developed a close relationship and to whom they had
previously given free services. The research team indirectly
supervised by scrutinizing reports from health center staff.
In Zaire at the village level, collection of payments
sometimes caused problems because villagers were not used to
paying for services in the manner that the project had
planned. Re-trained village midwives, for example, had to
begin charging for childbirth services that they had
previously provided at little or no cost. Several "important
persons" thought they should not have to pay - that the VHWs
should extend them credit - which put the VHW in a difficult
position. Community pressure helped ensure payments by elite
community residents in Benin.
70
Recordkeeping
Researchers in Benin developed forms for systematicly
recording diagnoses, treatments, and revenues. These proved
useful for both financial management and routine project
monitoring.
In Brazil, CPAIMC developed a system consisting of
client-by-client notation and daily summation of all revenue
generated by type of service or commodity. New instruments
were designed including a daily service statistics reporting
form, a monthly service statistics reporting form, a revenue
collection form, a stock control form, a summary of health
care visits, a revenue and cost monitoring form, a monthly
supervisory report, a contribution "carnet", a blood pressure
monitoring chart, and a road-to-health card. The greatest
problem encountered during the first two months was the
difficulty the auxiliary and technical nurses had in correctly
completing the statistical and revenue collection forms, but
with help, staff soon learned to fill these forms without
difficulty.
After treating a patient in Benin, the village health
worker recorded the diagnosis, treatment, and payment
received.
Drug usage and total revenues were calculated
daily and monthly for comparison with drug stocks and expected
revenues. Whenever receipts were lower than expected, the
missing amount was deducted from the responsible person’s
salary.
Lack of standard records was identified as a crucial
problem in the Zaire study.
Use of Funds
The revenues generated from payments for curative care in
Benin were used to pay village health workers and replenish
drug supplies. As the study progressed, revenue from curative
care was used to finance personal preventive interventions,
maintenance costs of the health center, petrol used for
supervision and mobile MCH clinics, and the village level
information system. In one "cercle" in Mali, service fee
revenue was used to pay for drugs, petrol, and clinic
furniture.
COSTSHARING SCHEMES
Charging for specific goods and services is not the only
means of having community residents contribute to PHC
financing. Potential users can also pay for "coverage", that
is, eligibility to receive treatment when needed at little or
no additional cost. A number of PRICOR-supported researchers
documented or developed cost sharing schemes, that is, schemes
in which all or most community residents contributed to
71
financing regardless of personal use levels,
were of several types:
These schemes
sale of health cards entitling purchasers to
defined benefits as needed (Thailand)
contributions of community labor, for
cultivation of health workers’ fields
(Swaziland, Liberia) or cleaning and
maintenance of health centers (Brazil)
financing of certain health center costs by
community organizations (Brazil)
flat rate assessments for payment of community
health workers’ salaries (Bolivia/Gonzalez)
financing by agricultural or other cooperative
(Bolivia/Miller, India).
The diversity of these studies limits comparative
analysis, but further details are given in Chart 8 and in
three case studies in appendix 2.
72
CHART 8:
PRICOR STUDIES THAT DOCUMENTED, OR HELPED TO ESTABLISH, COST-SHARING SCHEMES
REVENUE GENERATION
OBJECTIVE
NATURE OF
COSTSHARING
Benin
To finance recurrent direct
and support costs
Prepayment scheme proposed
but rejected because people
did not have confidence
that services would be
delivered.
Bolivia/
Gonzalez
To generate enough income to
motivate CHWs to stay on the
job
All community residents made Payments were made, and
a flat annual payment to
health workers have stayed
on the job.
health workers in the form
of potatoes.
Bolivia/
Miller
To establish coop-based
health care
Cooperatives with existing
health activities were
studied with a view to
expanded PHC financing.
Brazi1
Cost-sharing
Schemes were generally not
Coununity labor and supply
donations to clean and
carried out and, in any
maintain clinics and pay
case, produced little
utility bills; sale of
practical support.
"carnets’' to local residents.
India
Variable
Cooperatives financed health Several highly successful
care through sale of milk or cooperatives were studied,
other products.
STUDY
A)
RESULTS
COMMENTS
Eleven health staff
eventually created their
own prepayment scheme.
Effort collapsed because
coops could not expand
their activities.
Considerable community work needed to generate
payments. Residents required that promoter
work in only one community.
See case study in Appendix 2 for details.
See case study in Appendix 2 for details.
CHART 8:
PRICOR STUDIES THAT DOCUMENTED, OR HELPED TO ESTABLISH, COST-SHARING SCHEMES
STUDY
REVENUE GENERATION
OBJECTIVE
NATURE OF
COSTSHARING
Liberia
To compensate CHW
Swazi land
RESULTS
COMMENTS
Coumunal rice farm and
in-kind contributions to
compensate health worker
Scheme failed because
contributions were
generally not made and
because health workers did
not accept the arrangement.
Methods used for compensating traditional
healers did not work well for CHWs. See case
study in Appendix 2 for details.
To compensate health worker
Conmunity plowed CHW’s
fields to supplement
government salary.
Plowing delayed by chief's
absence but eventually
occurred. Field not sown
due to delays.
Chief's role in mobilizing community was
clearly vital. Health staff in neighboring
communities want to replicate the scheme. See
case study in Appendix 2 for details.
Thai land
To support basic PHC services.
MOPH now encouraging wide
Residents purchased health
replication.
cards entitling them to
treatment of up to 8 illness
episodes and expeditious
care at health centers.
Exact costs and benefits
varied from one community
to another.
Zaire
To support decentralized
health zones.
Prepayment scheme proposed
but rejected.
i
See case study in Appendix 2 for details.
CHAPTER 9
INPUTS AND ACTIVITIES FINANCED
This section deals with alternative methods used by
PRICOR researchers to (1) provide effective incentives for
community health workers, and to pay (2) for preventive/
promotive activities.
COMMUNITY HEALTH WORKERS
A major problem for many PHC programs is how to pay for
frontline health personnel, commonly known as community health
workers. PRICOR-supported researchers studied alternative
monetary and non-monetary methods for attracting qualified CHW
candidates and reducing attrition, for motivating effective
performance, and for covering the costs of supervision and
logistical support. A general finding was that both money and
non-material factors such as community prestige and selfrespect were needed to motivate CHWs, and that to a certain
extent one could be traded off against the other. Studies
discussed in this section are briefly described in Chart 9.
Recruitment and Retention
Several rsearchers either reported or helped develop CHW
recruitment and retention measures. In Thailand, CHWs could
easily be recruited and retained as managers of revolving drug
funds because the position was highly respected, community
service was a social obligation, and revolving drug funds were
structured to give managers financial incentives. In
recruiting CHWs, Indian program directors emphasized
commitment to social service rather than to personal income;
subsequent attrition rates were generally low. Lower caste
members were especially drawn to CHW positions by the
increased social status they offered. Unpaid "lead mothers"
in the Philippines (Osteria) kept at work partly to earn
social prestige.
Baker in Brazil attributed strong interest in new CHW
positions to high salaries (four times the minimum salary paid
to teachers) and the lack of other employment opportunities in
rural areas. The salary structure rewarded long service,
since increases of five percent were given for each additional
five years of service. CHWs with more than 25 years of
service were paid 20 percent more than those with less
experience.
More commonly, however, PRICOR-supported researchers
reported problems in CHW recruitment and retention, either
because of low social prestige or inadequate funding. Several
of the cooperative-based health programs in India attempted to
recruit relatively educated persons but could not because of
low pay. In Swaziland, delayed government salary payments and
weak community support for preventive care contributed to poor
75
CHART 9:
PRICOR STUDIES THAT DOCUMENTED, OR HELPED TO ESTABLISH, METHODS OF COMPENSATING CHWs
STUDY
TYPE OF WORKER
MATERIAL REWARDS
NON4iATERIAL REWARDS
COMMENTS
Benin
Community health worker
Salary, paid by the community,
through service fees
Community prestige; perceived
importance of CHWs was
increased by quadrupling
fees for patients who went
to the health center without
a CHW referral.
Fees charged only for curative care but
covered preventive costs as well.
Bolivia/
Gonzalez
Health promoter
Small salary; residents paid
fixed annual assessment in the
form of potatoes or wheat.
Attrition was high before
study due to inadequate or
irregular PVO or MOH salaries.
Scope of promoter’s functions
and coverage had to be reduced
so that community residents
would be willing and able to
pay for them.
Brazil/Baker
Basic worker responsible for
health education, maternal
and child health, and
communicable diseases
Salary paid by Federal
government at four times the
minimum wage
3ob security
Brazil/Nations Traditional healer, trained
for ORT
Gifts from patients but no
regular salaries
Social events with other
trainees; increased confidence
in healing skills; community
love and respect
CHW with basic promotive and
curative skills
Stipends and materials from
conmunity organizations and
residents
Ecuador
Regularly paid salary with increases for
long service; low attrition.
CHART 9:
PRICOR STUDIES THAT DOCUMENTED, OR HELPED TO ESTABLISH, METHODS OF COMPENSATING CHWs
STUDY
TYPE OF WORKER
MATERIAL REWARDS
NON-HATERIAL REWARDS
Haiti/
AEDC
CHW teaching child survival
skills
Cash payments based on number
of mothers acquiring a
specified set of child
survival skills, small
salary; funds derived from
small membership fees for
revolving credit clubs, from
the government, and from
drug sale profits
Prestige, increased knowledge
of health issues, sense of
good for the community,
personal satisfaction
Haiti/
Cayemi ties
ORT teachers and distributors
(traditional birth attendants,
ORS vendors, conmunity and
religious leaders, teachers
and mothers
Training per diem (paid by
government), but otherwise
none
New knowledge and skills
Honduras
Four types of volunteer
CHWs providing basic curative
services and health education,
midwivery services,
organizational support,
and malaria control. Paid
CHWs provide various services
Food, salary (some
workers only); resources
derived from community
gardens, government, and
drug sales
India/
Elkins
Various, but most had
curative, preventive, and
promotive functions
Small salaries, in-kind
contributions; resources
derived from government,
village milk and other
cooperatives, clubs, drug
sales, donor agencies
Encouragement of initiative,
job security, support when
there is opposition,
recognition for good service,
power to influence village
committee, job satisfaction
COMMENTS
CHW attrition was low in all 8 projects
studied
CHART 9:
PRICOft STUDIES THAT DOCUMENTED, OR HELPED TO ESTABLISH, METHODS OF COMPENSATING CHWs
STUDY
TYPE OF WORKER
MATERIAL REWARDS
NON-MATERIAL REWARDS
COMMENTS
Korea
Two types: Government-employed
CHWs who provided mainly
curative care with some
prevent i ve/promot i ve serv i ces;
Volunteer CHWs who provided
health education and minimal
curative care and supported
paid CHWs
Government salary for some
CHWs
"Meaningful and valuable"
health education programs
Volunteer CHWs discouraged by lack of
community support
Liberia/
Cole I
Mainly curative CHWs, also
responsible for health
education, birth and death
registration, and
coordination of immunizations
Labor in CHW fields and inkind payments were
recommended but not carried
out
Liberia/
Hall
School children, trained to
teach ORT, hygiene, home
sanitation, malaria prevention,
and referral
None
Nigeria/
Gray
CHWs responsible for diagnosis
and treatment of simple
diseases, health education,
and immunizations
Small salaries, derived from
patient fees managed by
village health committee
Advanced training for veteran
CHWs leading to higher job
classification, better
salary, and greater coninunity
prestige
High attrition, especially among the most
poorly paid workers. Researchers solved a
"financial" problem by finding no-cost
methods of increasing non-material
incentives.
Philippines/
Oster i a
Lead mothers, responsible for
environmental sanitation,
home gardening, drug sales,
recordkeeping, fund raising
None
Supervision and encouragement
from community leaders and
formal health care providers
Attrition was high in communities without
strong supervision
00
Compensation through labor donation used
for traditional healers but was
unacceptable to CHWs and community
residents. High attrition resulted.
CT
C
I
8 z0>
3
"0
o
0>
•o
0
c
a>» >
CT >»
O **
L. •*“
$
£
un
z
£
c
0)
fQ o
m ■♦-»
t.
a.
'TJ
5
I
3fl3 J3
*O
>
Q.
0>
>»
>»
i
hg
0)
2 2a.
Ql
i2
>»
C
fQ
g §
0 **
Q
01
a2
3€
ft
§
in
Q
£
c
o
<T3
Q
S
t—I
3
zuus
s
Q.
£
>»
>»
<U
<D
<
LM
c
!
o
a
£
£
0
8
0
UJ
§
s
I
£
3
s
£
§
•u
c
JK
C
3 3
§
c 8
<«
c o
**
zo 0
"O ■d
a3
2
Si
3
2 * 0i
o
f
'■a
•
Q
'TJ
Ii
** z 1
Oj
5
ft
S!
o
•• 0
0 4-»
0
CT
C
"Q
S 3 2
'O
0
TJ
*n 0
8 8
Q.
o
Q
o
0)
CT
>» -
g
<n i
a3
1
i!
<sJ
c
0
j 'n
u JZ
~~
(J) -Q
3
U 'TJ
O **
l
—
0.0
U J2
3
*9
c?
TJ
O
0
0
1
c
-■o
N
§
<0
C
2*
2 c
5
■Q 5
C
T3
O
JZ
C
HQ
2 S.c
o
3
&
*
CT
O TJ
-X C
U
0)
TJ
o
3 x:
8
Q. C
i-h-
□
CT
C
CT
C
o
£ g 8 2 ii
Q. ■•■*
C
AS
N1
?
40
0>
x r
u>
>»
Q
rO
■o
O u.
3
x: —
”'2
0
c
0
CT 0)
Q U
0>
c
3
0
8
9s
■sf
8a. in _C OiU 8
^3 •*•
a
O
x
Q
a
c
Q.
*D
9
S
0
a 3 -1
§ 8
0
-
2 =
O
0i
3 §
Hs
•— in
.«
8,3 CT Q 3
U
3c CT
2
CT)**
8 — •••0 8 3
c u -a
I Iu 88
•Q
C
° 2 0- 08— 0 3 2
i*.
U
0)
I
£
c
o
8
TJ
c
.2 0 <j ”*
<✓>
5
0>
79
2
2
h- O
8
"O
G
C
3
O
CHW morale and low retention. In Thailand, recruitment was
more difficult for managers of sanitation and nutrition funds
than for drug funds because the former were less often
profitable and could not be easily structured to provide cash
incentives for managers. High CHW turnover was also
experienced in Nigeria (Gray), one barangay in the Philippines
(Osteria), Liberia (Cole I), and Bolivia (Gonzalez).
In Gongola State, Nigeria, many CHWs left employment
within one to three years because of low salaries and lack of
advancement opportunity, supervision and community support.
Even though all communities paid minimal wages, a survey found
that CHWs at the higher end of the pay scale stayed on the job
about 50 percent longer than lower paid ones. Fourteen of 29
terminated CHWs had sought further training and were working
in higher level health care positions. Unable to increase CHW
salaries, program managers offered veteran CHWs additional
training as Community Health Aides so that they could qualify
for the greater pay and social prestige derived from
promotion.
To reduce turnover in Swaziland, researchers, managers,
and a local chief organized community labor to plow and sow
the CHWs’ fields so that CHWs could spend more time on health
work. The general reaction to the scheme was positive even
though labor contributions were delayed. A similar
arrangement in Liberia (Cole I), though based on traditional
modes of paying healers, was not acceptable to CHWs and
attracted little community support.
In Cochabamba, Bolivia, attrition rates as high as 70
percent were attributed to inadequate or unstable salaries.
CHWs also felt that the community was largely unsupportive or
unaware of their activities. They saw themselves as divorced
from important decision-making organizations and felt that
institutional support from a highly visible group of community
leaders would increase their credibility with the villages and
improve their motivation. PRICOR-supported researchers
organized contributions of wheat and potatoes to increase CHW
incentives, including perceived community support, and to
reduce turnover.
In Korea, both volunteer and paid CHWs were unable to
deliver effective preventive services where community support
was lacking. PRICOR-supported researchers attempted to use
members of informal community organizations as volunteers to
supplement government-paid CHWs. The latter trained
volunteers to conduct health education, to provide minimal
health care, and to support their own activities. The
volunteers told researchers that they found the training to be
"meaningful and valuable," but subsequent surveys of trainees
and community members found that efforts of the volunteer CHWs
had been largely unsuccessful and that their role was not
recognized in the community. In the Korean case, lack of
80
community support, rather than inadequate finance, appeared to
be the major obstacle to preventive services.
In some cases, careful role definition and task selection
were essential for health workers to gain either social
prestige or community financing. In Bolivia, villagers were
reluctant to support a CHW without assurances that the worker
would be competent, reliable and well-supplied with
medications. In Swaziland, about half of the CHWs and a third
of the community members interviewed told researchers that the
tasks performed by CHWs would have to be modified before the
community would be willing to support them. In conjunction
with the Ministry of Health, researchers determined that
immunizations, ORT and growth monitoring should be added to
the tasks performed by CHWs. Eight CHWs received additional
training to perform these tasks in one community. The
response was positive. Six later reported that the number of
villagers who came to them for advice on children’s health had
increased since their training. Community members were then
willing to contribute labor for the CHWs’ field work.
Role of Volunteers
An important question for PRICOR-supported researchers
and for primary health care generally concerns the potential
role of volunteers. Alternatives to direct cash incentives
were successfully used in Haiti (Cayemittes), Brazil (Nations)
and Liberia (Wall). In Haiti (Cayemittes), inability to pay
CHWs restricted effective ORT delivery, so community and
church leaders, traditional birth attendants (TBAs), teachers,
market sellers and others were invited to attend ORT training
sessions. TBAs were found to be particularly effective in
promoting ORT. Another study (by Augustin) in Haiti, on the
other hand, concluded that volunteers could be used for
discrete activities of short duration but not for sustained
community health work.
A key to successful mobilization of volunteers may be
linkage with tasks traditionally performed by school children,
healers, and others. In Liberia (Wall), researchers and
community leaders developed preventive health modules for
adolescents based on such common household activities as
cooking, cleaning, washing clothes and carrying water. School
children were then trained to teach others about ORT, hygiene,
home sanitation, malaria prevention and hospital referral.
Study results showed that the children had learned a good deal
and were carrying out their lessons. They received no
monetary incentives but did gain the respect and appreciation
of family and friends. The major program costs were for
training which was funded publicly through the regular
education budget.
Traditional healers working without formal compensation
were found to be effective in ORT programs in Brazil. In many
81
rural areas, traditional healers were more accessible,
affordable and acceptable than other health workers, and a
household survey found that a large majority (76.9%) of
mothers sought help first from a traditional healer when their
children had diarrhea. Almost all worked without cash
payment, only 15.8 percent telling surveyors that they should
receive any payment. Non-monetary incentives for traditional
healers included community love and respect plus nominal gifts
from patients. To mobilize these resources, PRICOR-supported
researchers worked with the healers to develop ORT training.
In the first 12 months afterwards, over 7,400 liters of ORT
were delivered at a cost to the program of approximately
US$4,000 for training (plus unspecified costs for
supervision). Additional non-monetary incentives for
participating healers included social events with other
trainees and increased confidence in health skills.
A study in the Philippines (Osteria) found that frequent
supervision, by health professionals or by local leaders,
increased chances that volunteers would continue working, In
one barangay where no support was given. 7 of 10 lead mothers
recruited for environmental sanitation tasks dropped out,
citing lack of time and pressures of farm and other incomerelated activities. Similar effects occurred in two other
barangays, but in three areas where local officials and health
personnel closely supervised and supported the lead mothers,
significant environmental sanitation improvements occurred.
Program directors interviewed in India stressed the
following non-material incentives to compensate CHWs for their
relatively low wages:
1.
Encouragement of initiative within the project
design;
2.
Job security to the extent possible;
3.
Full support when there is unfair opposition;
4.
Special public recognition for good service;
5.
Opportunities for further training in health
skills, including training programs with
attractive features, such as trips;
6.
Opportunities to share experiences with
coworkers who see self-fulfillment in their
service.
7.
Community respect.
Staff, program directors and CHWs themselves felt that
increased community respect was a major incentive.
82
Performance Incentives
It is not enough that workers stay on the job if their
performance is substandard or emphasizes services of little
longterm value. Program effectiveness may also be improved
through behavioral incentives to clients.
CHWs in Thailand were motivated to effectively manage
revolving drug funds because their salaries depended on the
existence and magnitude of profits. Fund managers were
generally paid only if the fund was profitable and growing.
Payment could take the form of wages, but was more often a
share of the profits ranging from 5 to 20 percent. Focus on
incentives derived from drug sales may have distracted CHWs
from preventive care and community activities in some cases.
In a PRICOR-supported study in Haiti, CHWs were paid on
the basis of the number of mothers who acquired a specified
set of skills in four child survival areas. Mothers were
motivated to learn these skills by the prospect of
participation in a credit scheme. Profits from the credit
scheme were then used in turn to pay the CHWs. In this
manner, CHWs’ income and mothers’ access to credit were both
directly related to the mothers’ mastery of child survival
skills.
In Benin, persons presenting at the local health unit
without a referral from a CHW were charged four times as much
as those who had been properly referred. In Thailand, health
card holders who obtained a hospital referral from a CHW were
able to receive faster attention.
In Thailand, women, especially mothers of malnourished
children, were willing to contribute labor or raw materials to
nutrition funds in exchange for food service and reduced
prices as well as for such non-material incentives as
opportunities for socializing and interesting demonstrations.
PREVENTIVE/PROMOTIVE ACTIVITIES
Though weak community demand for preventive/promotive
activities was a problem nearly everywhere, researchers
developed or reported a number of innovations to cover costs
of water and sanitation systems, growth monitoring, nutrition,
family planning, and oral rehydration therapy (ORT).
In Thailand, for example, communities
established and managed over 1000 nutrition and
sanitation revolving funds. Nutrition funds
support growth monitoring and remedial
nutrition activities, largely through the
voluntary labor of mothers, while sanitation
83
funds made loans for construction of privies
and water storage jars.
Researchers in Haiti (see Appendix 3) created
innovative women’s credit clubs that generated
funds for CHWs’ preventive activities and
rewarded mothers for learning child survival
skills.
In Rio de Janeiro, Brazil (Lassner), family
planning services and Pap smears, both
preventive in nature, were the two largest
revenuegenerators.
Fees for curative services in Benin and Zaire
were set high enough to generate surplus
revenue for preventive care.
These and other study results are summarized in Chart 10
and in Appendix 3.
84
CHART 10:
PRICOR-SUPPORTED STUDIES THAT DOCUMENTED, OR HELPED TO ESTABLISH,
METHODS OF FINANCING PREVENTIVE/PR0M0T1VE ACTIVITIES
PREVENTIVE/PROMOTIVE
ACTIVITIES FINANCED
STUDY
FINANCING OBJECTIVE
Benin
To cover costs of both
preventive and
curative coumunity
health services,
including supervision
and other support
systems.
Brazi1/
Baker
To analyze FSESP
experience in the
ccnmunity financing
of water supply
systems and CHWs
Brazil/
Lassner
To develop community
Family planning and
cost-sharing mechanisms Pap smears
while increasing
service accessibility
and use
Brazi1/
Nations
To incorporate ORT
within self-financed
traditional healing
activities
00
Ln
Water supply systems
ORT
FINANCING METHODS
RESULTS
Service fees
Most community and
district-level costs
were covered.
User fees
Service fees
Family planning
and Pap smears were
the two biggest
sources of revenue
Voluntary labor of
traditional healers
plus occasional gifts
from beneficiaries
Considerable ORT
provided; scheme
replicated in larger
project
CHARI 10:
PRICOR-SUPPORTED STUDIES THAT DOCUMENTED, OR HELPED TO ESTABLISH,
METHODS OF FINANCING PREVENTIVE/PROMOTIVE ACTIVITIES
STUDY
FINANCING OBJECTIVE
Dominican
Republic
To develop ORS pricing
strategy that the poor
could afford and that
would nevertheless
give distributors
adequate profit
incentives
Haiti
To develop ways to
motivate CHWs to teach
mothers to understand
and use preventive
services; to establish
supportive financing
methods
oo
cn
PREVENTIVE/PROMOTIVE
ACTIVITIES FINANCED
FINANCING METHODS
RESULTS
ORT
Two-tiered pricing
strategy
Ministry did not
accept
recommendations
ORT, growth monitoring.
family planning, and
immunizations
Elaborate revolving
credit scheme that
rewarded teaching
and learning of child
survival skills
Mothers learned
interventions and
participated in
schemes; portion of
funds generated used
to pay CHW salaries.
(See Appendix 3 for
additional
information)
Volunteer "lead
mothers"
Communities were
to develop tangible
means of supporting
"lead mothers", but
residents showed
little interest in
their activities
Community to plow
and sow the CHWs'
fields
Community supported
new CHW activities;
fields were plowed
but somewhat late
Environmental san
Philippines/ To mobilize community
Osteria
resources for PHC,
itation, health
education
including preventive
activities and services
Swazi land
To improve the scope
and quality of CHW
services, and to
supplement government
salaries
ORT
CHARI 10:
PRICORSUPPORTED STUDIES THAT DOCUMENTED, OR HELPED TO ESTABLISH,
METHODS OF FINANCING PREVENTIVE/PROMOTIVE ACTIVITIES
PREVENTIVE/PROMOTIVE
ACTIVITIES FINANCED
FINANCING METHODS
STUDY
FINANCING OBJECTIVE
Thailand/
Orathip
To identify> test, and
replicate costeffective models of
coununity financing
Sanitation and
nutrition
Zaire
To develop appropriate
payment schemes for
self-financed health
centers that do not
adversely affect
uti1ization
MCH, some immunizations Service fees, mostly
for curative care
(with some cross
subsidization)
oo
RESULTS
Multipurpose revolving Government now
funds based on drug
encourages rapid
sales generated
expansion of multi
surpluses for
purpose funds
nutrition and
sanitation. Many
communities also
capitalized and
managed funds devoted
specifically to
sanitation and
nutrition
Little collected on
preventive care, but
3 of 10 centers were
able to cross
subsidize preventive
care from curative
fees
CHAPTER 10
EVALUATION
Though successful implementation was the major goal of
most researchers, many were concerned as well to estimate the
effects of user charges on utilization/coverage and cost
recovery.
MAINTAINING AND ENHANCING COMMUNITY SUPPORT
Community groups and individuals asked to contribute for
a new service (or for an existing one that was previously
free) need incentives for doing so. The most commonly cited
incentive was improved accessibility and/or quality of
services, often because new services cost users less in time
and travel expense than previously available ones. In Rio de
Janeiro, women proved willing to pay for a range of services
when they were provided within the community, even though they
were available without charge at less convenient locations.
In the Indian study, 76% of participants cited "services at
hand" as one of their reasons for support. Increased
accessibility undoubtedly contributed to the success of
schemes in Liberia and the Philippines as well.
Results from India, Swaziland, and elsewhere indicate
that users’ perception of improved quality was a significant
factor in motivating community participation. In India, 42%
of those surveyed mentioned "good treatment" and 19% "easy
communication with staff" as reasons for joining. One project
director cited the "humaneness of the health workers expressed
through frequent home visits . . . and the personableness and
sincerity of the visiting nurse and doctor." The refresher
training given to Rural Health Motivators enhanced the
public’s willingness to pay in Swaziland.
Three other reasons cited for community support in India
were availability of supplementary food (in two locations
only), liberation/opportunity for self-development (cited by
many project directors and staff members), and ease of
hospital referral. Improved hospital access undoubtedly
facilitated the sale of health cards in Thailand as well.
Household surveys were used to identify obstacles to the
participation of residents in community health programs in
Brazil and India. Researchers in Brazil concluded that weak
participation was largely due to a lack of knowledge about the
location of health units, their operating hours, the services
available, and their costs. Although all community members
surveyed in Prazeres were aware of the local health unit, only
68 percent of those interviewed in Senador Camara had heard of
the unit there. Throughout the study area, the majority of
respondents did not know about recently introduced services,
depending upon the community, 40 to 86 percent did not know
that fees were charged for some services.
88
In the areas served by the Tribhuvandas Foundation and
the Adayar Project in India, the overwhelming majority (85
percent) of community members cited inability to pay as a
reason for not joining a local prepaid health program. A few
respondents had previously participated and dropped out due to
lack of funds. In contrast, only 10 percent of those surveyed
in Brazil (Lassner), for all communities together, responded
that charging fees weas an obstacle to using services.
Other major reasons for non-participation cited in India
were that the program did not provide needed services or drugs
or that personnel were not available, not competent, or
favored one group. Some staff felt that lack of awareness of
the program might have inhibited participation because
illiteracy was widespread and promotions relied heavily on
written material.
Affluence was found to be a barrier to broad
participation in India. The affluent did not seek service
because they could afford private doctors and did not find the
quality of service offerred to be acceptable. Furthermore,
community leaders and members of local organizations tended to
be from the more affluent classes who did not perceive the
program as benefiting them and consequently did not support
it. Several community and organizational leaders opposed
project objectives, especially in the area of water
management. The elites in TiIonia, for example, wanted to
prevent lower castes from using wells.
Several methods were considered to overcome these
obstacles. Service providers in Brazil recommended that
promotional activities emphasize new services, operating
hours, and general awareness of service availability. Staff
in India preferred house to house visits to promote the
program as well as community meetings, and enlistment of the
help of local institutions and influential persons.
UTILIZATION/COVERAGE
Researchers in Zaire and Brazil (Lassner) attempted (but
with limited success) to study the effect of community
financing on service coverage and utilization.
In Zaire, clinic utilization was affected more by local
management factors than by either the mode of fee collection
(per visit or per episode) or prices. Reactions of community
members and service providers to the shift from visit to
episode fees were strong and unexpected. Several personnel
preferred visit fees because they were generally not fixed and
could be ’’negotiated” with the client. In the Katanda Health
Center, the auxilliary nurse was dismissed after showing open
hostility to the shift. He subsequently opened a private
dispensary in the same service area and drew many patients
89
away from the SANRU center. An open conflict developed in
another center between members of the Village Development
Committee and the auxilliary nurse because the latter felt
that fees had been set too low.
Communities generally preferred episode fees rather than
visit fees because of greater:
health care continuity (once payment was made,
there were fewer financial reasons to
interrupt treatment);
ease in health center administration (since the
fee level was known to everyone to facilitate
monitoring of receipts)
reinforcement of community solidarity (since
health care prices were equally shared by all
users).
Utilization of some services decreased in the one center
that shifted from episode to visit fees.
Curative visits per capita varied enormously from one
health center to another between October 1984 and September
1985, from a low of 3 visits per year per 100 population in
Tshileo and Katanda to a high of 245 visits per 100 persons in
Lukunga.
(See Table 1.) One reason for this was that people
seeking health care were far more likely to.go to the SANRU
health center in some areas than they were in others. The
fifth column of Table 1 shows that the proportion of persons
seeking health care who went to the health center during the
1985 two week recall period ranged from 1.4% in Katanda (where
the nurse set up a competing private practice) to 83.2% in
Lukunga (a small, well-defined, and isolated service area), A
second, less interpretable, explanation was that people in
some areas reported very infrequent use of any health care
provider (only 10 visits per 100 persons per year in Tshileo,
for example), while in other areas (Kangoy, for example)
visits were up to 35 times more frequent. Results in one area
were affected by the large number of followers of an apostolic
faith who reported their source of care as "other”. Visits
per episode of illness also varied greatly, from scarcely more
than 1.0 in Lukunga, Muadi Kayembe, and Kaniama to over 6.0 in
Kangoy.
Although researchers anticipated increased use of centers
charging fixed fees per episode (rather than variable fees per
visit) , before/after changes were inconsistent and difficult
to interpret. No general conclusions can be drawn from the
data collected in the two-week recall survey and the review of
health center records. However, much of the variation in
utilization can be explained when these data are complemented
90
TABLE 1
Zaire:
Health center visits for curative care
October 1984 - September 1985
Health Center
Population
1) Kabambaie
2) Mbau
3) Katanda
4) Tshileo
5) Kindamba
6) Lukunga
7) Muadi Kayembe
8) Kaniama
9) Kangoy
10) Keba
8403
33455
43870
1908
1972
12924
10337
16941
13839
6257
Percent of
Conditions
Treated at
Clinic **
Visits per
Capita to
All Care
Providers
0.39
NA
0.03
0.03
1.22
2.45
0.19
0.56
1.72
0.58
47.1%
45.5%
1.4%
31.2%
50.7%
83.2%
18.5%
15.1%
48.8%
17.1%
0.83
NA
1.80
0.10
2.41
2.94
1.04
3.73
3.52
3.38
0.56
38.4%
2.41
Visits per
Capita to
No. of Health
Visits* Center
2436
NA
843
1374
2332
4824
2489
5823
29086
7992
Medians
*
From Health Center records
** Household survey, two week recall data
*** Estimated by combining Health Center and two week recall data
91
with a qualitative analysis of the context of events and
behavior in each zone.
To evaluate utilization in Brazil, researchers used a
quasi-experimental design and time-series analysis. Data from
the existing CPAIMC micro-computer-based information system
were used to identify pre-implementation trends for 13
utilization variables. These were projected into the trial
period by fitting a linear regression line to monthly values.
The two-tailed T-test was used to compare the monthly number
of observed visits for each variable with the number of
projected visits. Variable by variable analysis helped
determine whether utilization remained stable, decreased or
increased following implementation.
In six of eight service sites, the trend in total number
of visits remained unaltered, while in two of the sites it
decreased significantly. For all variables combined, apart
from the total number of visits, significant utilization
decreases occurred more often than no change, Significant
increases in utilization were few.
Only one significant association was found between an
individual financing scheme and changes in use. In all five
units where registration and Pap smear fees were collected,
the number of initial visits for women declined.
Coverage - defined as the proportions of women aged 15 to
49 and children aged 0 to 4 who received care in the previous
six months - was estimated before and after implementation
using community surveys. In the four communities where valid
comparisons were possible, coverage for women increased in two
and declined in the others. For children, coverage rose in
all four communities.
COST RECOVERY: REVOLVING DRUG FUNDS
Of the 22 Thai revolving drug funds studied in detail, 19
were profitable and 10 earned profits over 30 percent.
Profitable funds were well managed, well stocked, and able to
compete successfully with both private and public sector
alternatives. Most importantly, some funds had diversified
sources of income by selling other goods in addition to drugs
and supplies. One fund had a gross profit of 172 percent, of
which half went to shareholders, 30 percent to fund capital,
and 20 percent to fund managers. This fund had expanded to
cover a total of ten villages. In another case, the profit
distribution was 20 percent to shareholders, 65 percent to
investment, and 15 percent to managers.
Even though social obligation may have motivated initial
share purchase, high profits were the main reason for
subsequent share purchases and for the dramatic variation in
the size of drug funds.
92
In one village, the initial share purchase at the
beginning of 1981 was 219 shares with total shares of 10,950
baht. In January of 1982, shareholders received profit
distributions of 43 baht per share - an annual return of 86
percent. In January of 1983 they received 25.49 baht per
share - an annual return of 50.8 percent. Since January of
1982, this fund has sold 321 additional shares with a total
value of 16,050 baht, more than doubling the initial capital.
Of the 22 funds, only one had decapitalized, Problems
occurred in other funds, however, vl
j
when the ;tambon
health
officer was unsupportive, when management committee members
lacked commitment and energy, and, most commonly, when health
volunteers played a limited role in sales, record-keeping, and
inventory control.In the Philippines, all funds enjoyed
profits ranging from 17 to 29 percent. Profits were added to
capital for purchase of additional drug stocks. In Liberia as
well, none of the funds decapitalized during the first year of
operation. Dominica’s fund had operated successfully for two
years at the national level, although user charges had not
been instituted.
COST RECOVERY: OTHER
Researchers in Benin, Brazil (Lassner), and Zaire
collected data on the proportion of health care costs covered
through community financing.
In Benin, researchers sought to cover all essential
recurrent costs at both the health center and village levels.
Since the government paid salaries and electricity costs,
community financing was sought for restocking of drugs and
supplies, for VHW remuneration, for the transportation costs
incurred in supervision and peripheral MCH clinics, for VHW
treatment and home visit forms, for the cold chain, and for
health center maintenance. Researchers sought to cover these
costs with receipts from curative care delivered at both the
health center and village levels.
° Receipts collected at the village level in Benin covered
23-6 of total recurrent costs and were not sufficient to cover
even village-level expenses (VHW remuneration, drug costs, and
transport for MCH clinics and supervision). The financing
scheme was designed, however, so that non-residents using
project health centers were charged four times the rate for
residents; and by combining these receipts with those from the
village, researchers were help to cover both village and
health center costs. Initial costs were partially subsidized,
and use of curative care was low. As utilization increased,
profit margins also rose until income and revenue reached the
breakeven point two to three years after project startup.
93
In Brazil (Lassner), "proportion of total PHC costs
covered by the revenue generated" was selected as the
principal indicator of performance. Costs of providing care
at the units and miniposts were defined as "direct" (all costs
for personnel, drugs, and supplies, as well as other general
expenses such as water, electricity, maintenance, and
transportation); "indirect" (costs related to program
management and administration, training, and evaluation); and
"total" (sum of direct and indirect).
The mean monthly revenue generated by the schemes varied
significantly among the nine communities. The strategy that
generated the most revenue involved charging for selected
services, selected types of visits, and drugs. The strategy
that generated the least revenue involved charging only for
selected contraceptives. The most successful units and
miniposts were able to cover 20 percent of direct recurrent
costs (11 percent of total recurrent costs) during the nine
month study period.
(See Table 2 for additional details.)
The scheme components that generated the most revenue were
charges for Pap smears, contraceptives, physician visits, and
drugs, in that order. The proportion of total revenue
generated by Pap smear fees ranged from 20 to 72 percent, In
general, most experimental communities covered a greater
proportion of PHC costs than did control communities, partly
because experimental communities used a variety of financing
means while control communities were limited to fees for
registration, Pap smears, and contraceptives.
In Zaire, fees were set in diverse ways and varied from
one health center to another. No technical formula for setting
prices was agreed upon, yet the six centers with usable
revenue data reported that they covered a median of 107.3% of
their curative costs, including the costs of zonal supervision
and mobile teams.
(See Table 3.) Revenues generated from
preventive care covered between 0.4% and 10.6% of associated
costs only, but three centers generated revenue from curative
care in excess of costs permitting cross-subsidization of
preventive care. Overall, the six clinics reported median
total cost recovery of 66.5%.
While Zaire appears to have been most successful in cost
recovery, it is not clear whether this should be attributed to
revenue generation or to cost (ie., expenditure) reduction.
Clinics in both Brazil and Benin had supplemental funding to
ensure certain quality and coverage standards, even though
researchers in both instances sought the maximum feasible
self-financing. In Zaire, on the other hand, health zones
were largely on their own to generate income as best they
could from PVOs, external agencies, and users. Most of the
shortfall that occurred had to be taken from drugs and
services. As noted in Chapter 4, PHC costs per capita and per
visit varied enormously in the Zairian centers studied, partly
reflecting a very uneven distribution of resources.
94
TABLE 2 (BRAZIL/LASSNER)
Proportion of Total and Direct PHC Costs Covered by
Revenue Generated from Comnunity Financing Strategies
April-December 1985
Mini-Posts
PHC Revenue &
Costs in US$
Senador
Camara
Total Revenue
Women's Health Care Units
Vila
Alianga
Vila
Kennedy
Barrelra
do Vasoo
Bispo
Prazeres
249
812
801
1,087
321
200
3,745
2,702
4,086
5,116
5,432
5,645
6,414
4,110
6,390
4,999
7,477
9,318
9,685
9,537
10,888
20.0
1.6
4.5
9.2
19.9
15.7
20.0
5.7
3.1
10.9
.9
2.6
5.0
10.9
8.6
11.2
3.4
1.8
Borel
Dende
Parada
de Lucas
478
37
168
2,385
2,383
4,370
Direct Costs (%)
Total Costs (%)
Total Direct
Qosts^
Total Costs2
MCH/FP Units
Proportion of
Costs Covered
by Revenue3
(JI
1
2
3
Direct costs include personnel, medicines and supplies and general administrative expenses, such as utilities, rent.
maintenance, transportation, food.
~
etc.
Total costs include direct costs plus indirect costs.
In this and other tables proportions were <calculated
’ '
utilizing revenue and cost data in cruzeiros and may not equal exact
proportions calculated in US dollars due to currency conversion and rounding.
TABLE 3
Zaire:
Percentage of health center operating costs covered
October 1984 - March 1985*
Health Center
Curative
Preventive
Total
1) Kabambaie
2) Mbau
3) Katanda
4) Tshileo
5) Kindamba
6) Lukunga
7) Muadi Kayembe
8) Kaniama
9) Kangoy
10) Keba
197.1%
139.8%
33.0%
119.3%
NA
NA
95.3%
94.1%
NA
NA
10.6%
1.6%
0.4%
4.7%
NA
NA
8.3%
2.7%
NA
NA
136.6%
120.0%
7.6%
35.7%
NA
NA
63.2%
69.8%
NA
NA
Medians
107.3%
3.7%
66.5%
* Cost data are for the time period indicated,
for September 1984 - February 1985.
96
Revenue data are
FACILITATING FACTORS AND CONSTRAINTS
Participants in a June 1986 conference of PRICORsupported researchers identified a number of incidental
factors that either facilitated or inhibited the success of
their community financing schemes. National policy factors
were summarized in Chapter 2, but there were others of a more
personal or cultural nature as well. Other facilitating
factors included:
the local origin of the principal investigator,
as in Liberia, Bolivia (Gonzalez), and Benin
a widespread public conviction, as in Mali and
Liberia, that outside help for primary health
care would not be available and that self-help
was essential
■
a tradition of loyalty to royal directives, as
in Thailand
a high health services utilization rate to
spread fixed costs, as in parts of Zaire
■
prior community experience in managing
revolving funds, as in parts of Thailand.
Constraints, on the other hand, included:
the apparent availability of outside funds, as
in Brazil (Lassner) and the Philippines
a history of receiving certain goods and
services without charge, as in Benin and Brazil
(Lassner)
■
prior experience of paying and getting poor
results, as in Liberia and Benin
the periodic scarcity of cash and even inkind
resources in Bolivia (Gonzalez), Zaire, and
Benin
weak community management skills in many locations.
Several,researchers described major problems in
convincing people that primary health care, especially
prevention, was good health care and worth paying for.
Researchers in a number of sites, including Zaire, Brazil
(Lassner), and Benin, encountered difficulties in orienting
health center staff to the new fee structures and the need to
enforce them impartially.
97
THE AGENDA FOR THE FUTURE
PRICOR-supported studies successfully resolved a number
of PHC financing issues, particularly ones relating to
curative care, drugs, and strictly local costs. Progress was
reported on devising methods of paying for supervision and
preventive/promotive activities and on strengthening CHW
compensation and incentives. A great deal, previously
discussed only in anecdotes, was learned about the process of
establishing self-managed PHC financing schemes within
specific communities. The American and developing country
researchers that PRICOR supported deserve much of the credit
for these successes.
Research should continue in each of these areas, however,
especially on:
how to pay for supervision and other support
activities
how to generate community demand for
preventive/promotive activities or to find ways
of cross-subsidizing them
simple price and fee-setting techniques,
especially for hyperinflationary conditions.
A great deal of work is also needed on methods of
replicating study results, with lower research inputs as
population and geographic coverage expand. Operations
research was a very useful tool at the micro level and should
be equally useful during replication. Clearly, though, many
researchers were active in community mobilization, training,
and routine management - roles that someone else will have to
fill as projects expand. Replication in Thailand, Zaire,
Liberia, and Brazil (Lassner and Nations) occurred with
relatively little research input, but other settings may
require further community by community groundwork and data
collection. Future operations research should consider how
best to conduct this expansion.
98
BIBLIOGRAPHY
Operations Research Methods: A General
Approach in Primary Health Care. Center for Human
Services, Chevy Chase, MD. (1985).
S.N. Blumenfeld.
A. Goldsmith, B. Pillsbury, D. Nicholas. Operations
Research Issues: Community Organization, Center
for Human Services, Chevy Chase, MD. (1985).
L. Rice Marquez, A.T. Brownlee, J.P. Molzan, J. Reynolds.
Community Health Workers: A Comparative Analysis of
PRICOR-Funded Studies. Center for Human Services,
Chevy Chase, MD. (1987).
J.S. Newman, P.S. Reyes, K.E. Johnson.
Reaching Mothers
with ORT: A Comparative Analysis of PRICOR-Funded
Studies. Center for Human Services, Chevy Chase,
MD. (1987).
J. Reynolds, K. Celeste Gaspari.
Operations Research
Methods: Cost-Effectiveness Analysis. Center for
Human Services, Chevy Chase, MD. (1985).
S.
Stanton Russell, J. Reynolds. Operations Research
Issues: Community Financing. Center for Human
Services, Chevy Chase, MD. (1985).
M.
Schaefer, J. Reynolds. Operations Research Issues:
Community Health Workers, Center for Human
Services, Chevy Chase, MD. (1985).
99
APPENDIX 1
BRIEF SUMMARIES OF PRICOR-SUPPORTED
COMMUNITY FINANCING STUDIES
100
PRISSR
CF
Benin •
Primary Health Care Operations Research
Study Abstract
COMMUNITY FINANCING OF PHC SERVICES IN THE
PAHOU HEALTH DEVELOPMENT PROJECT, BENIN
An operations research study was conducted by the Unitarian Universalist
Service Committee in the People’s Republic of Benin during 1983-85 to develop
a community financing strategy to support the Pahou Health Development
Project's primary health care (PHC) system. After considering several viable
alternatives, the project managers proposed two financing strategies to the
families living in the 15 villages that constitute the communes of Pahou and
Avlekete: fee per episode for curative treatments, with prices marked up to
cover other costs, and a prepayment insurance-type scheme that would give them
free access to PHC services.
Even though it had been expected that both
options would be selected, the families all chose to pay the fee per episode.
Families explained that even though they recognized that they would benefit
from a prepayment scheme, they could not afford to make the annual (or even
semiannual) payment in advance, Only the 11 members of the health staff chose
the prepayment scheme for their families.
With the agreement of the community leaders, the revenue generated from the
payment of curative care was used to cover the costs of drug supplies and
village health worker (VHW) remuneration. The investment costs were supported
by foreign donors with the understanding that the operating costs; would be
financed by the Beninese government (health center staff salaries and
infrastructural costs) and by the communities.
Standing orders guide VHWs and health center staff in diagnosis and treatment
and in determining what fee to charge the patient. The treatment fees charged
The. base
by health centers and VHWs are standard throughout the project,
treatment charge is the cost of the drugs involved in the treatment plus a
This markup was calculated by
percentage to help cover operating costs,
estimating the number of cases of the disease and thus the proportion of
This calculation was then used to
personnel time devoted to the treatment,
the
operating
costs to be added to the drug
of
determine the percentage
total
preventive care costs (e.g.
proportion
of
costs.
In addition, a
home
distribution
of Oralyte) are
and
pre/postnatal care
care,, vaccinations,
The
reduction
in
drug costs as a
included in the price of curative care,
possible
to
mark
up
the treatment
result of using generic drugs makes it
prices so that they cover other costs (besides drugs and VHW remuneration) and
yet remain affordable and acceptable to the population.
People seeking curative care pay for the entire treatment at their first
Curative care and
visit, regardless of the! number of followup visits.
interventions are administered free to school children and
preventive
Treatment prices have been quadrupled for those people residing
indigents.
In addition to
outside the project. area who wish to use project services.
increasing project revenue,
revenue, this helps limit the influx of "outsiders” and
prevents them from coming
coining to obtain project drugs to resell for a profit
elsewhere.
The "outsider” treatment prices are also applied at the health ’ c
03
center to people living in the project area but whose VHW has not referred >-
fa
s
5530 Wisconsin Ave. • Chevy Chase, MD 20815 USA • [301)654-2550 • Cable: URCINTER • Telex: 64693
x
ip
cn 2 £
I c TO
r.j J® CH
XQ £
CD /
PR1COR is operated by the Center for Human Services for the .Agency for International Development under Cooperatrive Agreement No. AID/DSPE-5920-A-00-1048-00. \k'§\
101
,I
pHCAUO
/,
them to the health center,. supporting the VHWs and increasing
credibility by ensuring that people follow the established procedures.
their
A considerable effort went into managing and sustaining the community
financing system, Both the health center staff and the VHWs received training
The head nurse of the health center
in resource management and logistics.
also received inservice training in bookkeeping by an accountant who regularly
goes over the health center and VHW accounts of the cash box. Control of cash
box transactions (collection of receipts and of incurring expenses) is the
responsibility of a sub—committee on financial management of the Commune
Health Committee comprised of three community leaders and the health center's
head nurse. In addition, biannual evaluations of coverage levels achieved are
carried out by the health center staff.
These results, discussed with the
Commune Health Committee, have led to readjustments in the prices of certain
treatments. Supervisors also use the receipts to compare the amount of drugs
used and revenues collected with drug stocks and expected revenues
revenues.. When
nurses' or VHWs' receipts are lower than they should be, this amount is
deducted from the responsible person's remuneration.
The
revenue
from
this
financing
system has
assured 90-100 percent
accessibility to services for the entire population (defined as living within
2 kilometers of service delivery locations) and a VHW attrition rate of less
than 7 percent in 3 years.
Between July 1983 and June 1985, receipts from
community members registered in the project were able to cover 40-50 percent
of the following recurrent costs:
replenishing of drugs, remuneration of
VHWs, petrol for supervision and mobile maternal and child health clinics,
maintenance of the health center, and the information system at the village
level (VHW treatment forms, home visit forms, pre/postnatal forms).
When
’’outsider” receipts are included,
included^ 60-80 percent of these costs were covered,
•In the second half of 1985, total revenues generated from curative treatments
were able to cover all of these expenses.
The cost analysis was based on a series of studies of utilization, personnel
time (work-logging), and resource utilization (logistics records).
The
preliminary results (July 1983 through June 1985) showed an average cost of
$4.75/capita per year, including all recurrent costs, annual amortized capital
costs, and a yearly portion of costs of startup activities amortized over ten
years. These costs represent real project costs, including consultants, extra
training for project staff, time for research, and training of student
midwives by the health center staff. Estimated recurrent service costs come
to $2/capita per year.
In conclusion, to become self-sufficient, a community financing system needs
at least 2 years of good management (routine monitoring of coverage and strict
accounting).
Further, a community financing system operating only at the
village level should not expect to cover the costs of drugs, VHW remuneration,
and such support services as supervision.
The study showed that only 25
percent of -project revenues came from village-level treatments (including
"outsider” receipts).
However, further study might reveal other ways of
generating revenues and organizing service delivery to allow self-sufficiency
at the village level.
*
*
★
This study was conducted by the Unitarian Universalist Service Committee
(UUSC) from April 1983 to December 1985.
Further information is available
from the principal investigators. Dr. Eusebe Alihonou, B.P. 1822, Cotonou,
Benin, or Ms. Elizabeth Coit, UUSC, 78 Beacon Street, Boston, Massachusetts
02108, or from Ms. Marty Pipp, PRICOR study monitor (Chevy Chase).
102
PRI<S®R
CHW/CF
Bolivia
Primary Health Care Operations Research
Study Abstract
COMMUNITY FINANCING TO REDUCE ATTRITION OF
COMMUNITY HEALTH WORKERS AND INCREASE
HEALTH SERVICE COVERAGE IN RURAL COMMUNITIES
Researchers from the Institute de Investigaciones Medico Sociales (IIMS)
undertook an operations’ reseach study to find ways of reducing attrition rates
of community health workers, or Sanitarios Natives (SNs), in Cochabamba,
Bolivia.
The objectives of the study were to identify both the conditions
under which a community would be willing to support a health worker, and
feasible financing schemes that would provide stable salaries for the SNs.
Preliminary results show that villagers are willing to pay in kind for health
worker salaries provided the workers are reliable and work exclusively in
their own communities.
Since the total revenue from these in-kind payments
would cover only about half the SN's current salary, the researchers concluded
that stable, community-supported health workers could be provided only by
training and deployingr a lower level health promoter at a reduced salary level
to be determined by the
1
community.
This scheme would also allow for the
expansion of services into areas where no formal services had previously been
provided.
In order to determine» how best to manage these in-kind payments, project
investigators worked with community groups and individuals and held open
community discussions,
After several sessions the project investigators were
able to propose the following scheme: the quotas (in the form of either wheat
or potatoes) will be collected in May after the harvest. The quota is roughly
equivalent to USS 7.00 per family regardless of the family's land holdings.
The organization and collection of the quotas rely on two community
organizations:
the sindicato and the health committee.
The sindicato, a
traditional and highly visible organization in rural communities, is an ideal
collaborating organization because it exists in each community and attendance
is obligatory for each family head. Each sindicato director in the study area
supports the project and will be responsible for ensuring that each family
pays its quota. Furthermore, each sidicato has elected a member to serve on
the second collaborating organization for this scheme, the health committee.
This group, composed of the representative from the sindicato and an elected
mother, will be directly responsible for the collection, storage, and
marketing of the in-kind payments. The overall scheme will be supervised by
two Bolivian physicians who work with INEDER, a Dutch private voluntary
organization (PVO).
The scheme, already implemented in seven communities, involves retraining the
existing SN to provide better services and to supervise six promoters located
in satellite villages.
Promoters differ from the SN in that they are
generally older mothers, with little or no education, who will receive half
the salary of the SN for the same time worked. The SN's longer training, and
his previous year's work experience in the health post, partially explain the
pay differential.
5530 Wisconsin Ave. • Chevy Chase. MD 20815 USA • (301)654-2550 • Cable: URCINTER • Telex: 64693
PR1COR is operated by the Center for Human Services for the .Agency for International Development under Cooperative Agreement No. .AID/DSPE-592O-.A-OO-1048-00.
103
Training and supervision of the promoters is conducted by the research staff
and the two Bolivian physicians who supervise the overall scheme.
Bimonthly
training focuses on specific health topics such as ORT, first aid.
aid, or
sanitation.
The worker is provided with pamphlets that include illustrations
to reinforce important points.
Supervisors visit once a month and review
workers' records and meet with the community organizations responsible for the
management of the health workers' salaries.
A signed contract between the Ministry of Health (MOH) and INEDER guarantees
continuity of this project's personnel, and institutional support, for at
least 3 more years.
Dissemination of the project's results has been
accomplished through several meetings with the MOH and PVOs working in the
Cochabamba area. Two PVOs (IDEPO and Project Tirague) hope to use the results
from the PRICOR study to expand the model.
*
This study was conducted from October 1984 through March 1986.
Further
information is available from the principal investigator. Dr. Rene Gonzalez,
Director, UMS, Casilla 4444, Cochabamba, Bolivia, or from Ms. Karen Evalyn
Johnson, PRICOR study monitor (Chevy Chase).
104
PRIS®R
CF
Bolivia
Primary Health Care Operations Research
Study Abstract
DETERMINING ALTERNATIVE COMMUNITY
FINANCING MECHANISMS FOR SUPPORTING
PRIMARY HEALTH CARE SERVICES IN BOLIVIA
The Bolivian Mission of the U.S. Agency for International Development
conducted a feasibility study in the early 1980's to investigate the potential
to develop privately financed Primary Health Care services in the Department
of Santa Cruz, Bolivia.
In 1984 PRICOR funded the Fundacign Integral de
Desarrollo (FIDES) of Santa Cruz to design PHC service packages and test
alternative financing mechanisms.
During the preliminary stages of this
project, three regional cooperatives had been identified to jointly administer
the project.
It was hoped that through these cooperatives, a PHC Delivery
System could be established which would provide essential primary health care
services to unserved or underserved, high-risk populations in selected rural
and marginal urban areas of Santa Cruz, and that these services would be
supported entirely by community-generated funds at the end of the project's
third year.
The coops are particularly suited for this purpose because they
are client-oriented, privately administered and are often organized in part to
provide limited health services to their members.
Based on the belief that the new PHC services would be administered and
financed through these cooperatives, FIDES/PRICOR designed the field study
around the idea of cooperative-based PHC for low income households. Data were
collected through a variety of studies to describe the environment in which
(1) a literature review, (2) key informant
the PHC would operate including:
interviews in the Montero area, (3) case studies of innovative health care
financing approaches in Bolivia, (4) household surveys, (5) an inventory of
institutional resources in the Montero area, (6) an analysis of communities in
the Mineros and San Julia area, (7) area mapping, (8) family notebooks, (9)
longitudinal data collection/analysis, and (10) an analysis of the program
environment. FIDES initiated this effort by conducting a household survey of
three different populations in the Department of Santa Cruz: peri-urban
neighborhoods in the Departmental capital, Santa Cruz; the town of Mineros and
its rural dependencies; and the remote rural colonization zone of San Julian.
For several reasons beyond the scope of this report, the PHC services will not
be administered through the original three cooperatives, The data presented
here, however, are still appropriate for any entity concerned with serious
health planning in the Santa Cruz region and should be so considered.
The survey was administered to 545 households in Santa Cruz, 470 households in
the Mineros area and 267 in the San Julian area, all households within the
coop health projects' target populations.
Information gathered included:
basic socio-demographic characteristics of the household; a two-week recall of
illness episodes; a one-month recall of hospital utilization; fertility data
for the last year for all women 15 - 45 years old; data on health care
utilization and expenditures during last two weeks; the education and economic
status of the household; informants'
informants' community participation and health
5530 Wisconsin Ave. • Chevy Chase, MD 20815 USA • (301)654-2550 • Cable: URCINTER • Telex: 64693
PR1COR is operated by the Center for Human Services for the .Agency for International Development under Cooperative .Agreement No. .A1D/DSPE-5920-.A-00-1048-00.
105
<
attitudes; and types of health services that informants would like to have
access to and for which they would be willing to pay.
The results of the household survey in the three locations can be separated
into three distinct areas:
What people consider to be the preferred or ideal delivery system
What financing mechanisms people would be willing to support
What are the current health expenditures
The interviewer asked the household respondent (head of household) whom he
felt was responsible for providing health services in the community, In each
area, over half (51.1-63.5 percent) of the respondents felt that the
government should provide health services. In Santa Cruz, 25 percent felt
that the community/families should provide services, while that figure was 17
percent in Mineros and only 2.1 percent in San Julian.
In strong contrast to the ideal provider, the responses to the ideal payment
mechanism indicated that a majority of households felt it was the family's or
the community's responsibility to pay for health services: 51.3 percent in
San Julian; 57.6 percent in Mineros; and 61.0 percent in Santa Cruz.
While it is informative to understand the preferred, or ideal, delivery scheme
(provider and payment mechanism) in each area, it is perhaps more useful to
examine what financing schemes people in these areas are willing to support.
In San Julian, the most popular financing mechanisms were crops, annual fees.
and installments, In Mineros, they were installments, annual fees and
payments-in-kind. In Santa Cruz, long term credit was overwhelmingly the most
popular, followed by annual fees and installments. The percentage of
respondents unwilling to support any of the schemes was very low in each
area: 3 percent in San Julian, 1.5 percent in Mineros, and 1.3 percent in the
city of Santa Cruz.
While it seems clear that people in each of the survey areas would be willing
to contribute to the financing of health services, it is recommended that a
combination of financing strategies be implemented in each of the areas since
no one mechanism is supported by a majority of residents. Fee for services
and drug sales are two common ways to finance curative services. The
strongest rationale in support of fees and drug sales is that it is possible
to encourage cost sharing while at the same time reducing potential
overutilization. However, these mechanisms may exclude the poorer members of
communities who cannot afford to pay, and a graduated scale might be developed
in accordance with ability to pay.
However, due to Bolivia's current economic crisis, the inflation rate
seriously hinders the continuation of revolving drug funds or set
fee-for-services . The authors argue that regardless of the particular
combination of financing schemes in each area, all goods and monies should be
converted into durable goods immediately. Because of Bolivia's unprecedented
inflation rates, this represents a challenge for any community financing plan.
*
*
A
This study was conducted from April 1984 to August 1985. Further information
may be obtained from the principal investigator, Mr. Martin Miller, Fundacion
Integral de Desarrollo, Casilla 1911, Santa Cruz, Bolivia, or from Dr. Jack
Reynolds, PRICOR study monitor (Chevy Chase).
106
PRI€®R
CHW/CF
Brazil
Primary Health Gare Operations Research
istudy Abstract
THE USE OF VISITADORAS AND
FINANCING OF COMMUNITY WATER SUPPLY
Researchers from the Fundagao Servigos Especiais de Saude Publica (FSESP) and
Johns Hopkins University conducted a retrospective case study of FSESP in
The FSESP is a public foundation started by AID and the
rural Brazil.
It is linked to the Ministry of Health and
Brazilian Government in 1942.
provides primary health care and water supply services to over 13 million
In particular, the
people (approximately 10 percent of Brazil’s population),
study focused on FSESP's work with community health workers (visitadoras) and
community financing of water supplies.
This study presents the findings of many earlier operational studies and the
resultant changes.
The PRICOR researchers believed that primary health care
(PHC) planners and workers all over the world would benefit from the
documentation of experiences of a highly successful and enduring program of
PHC delivery and backup secondary care.
COMMUNITY HEALTH WORKERS:
THE VISITADORAS
In their study of the visitadora (CHW) program, the PRICOR researchers looked
at the role, selection, training, supervision, compensation, total costs,
quality of care, coverage, and productivity of the visitadoras.
The
investigators found that the FSESP visitadora program was directed from the
central level and was highly organized. Visitadoras perform a variety of PHC
tasks in such areas as maternal health, infant and child health, school
health, control of communicable diseases, curative health care, and community
education.
Selection of young women for the job of visitadora was rigorous
and highly competitive and a great deal of effort and resources went into
their training and supervision.
The visitadoras received regular salaries
that were four times as much as the standard minimum salary in Brazil.
The
monetary incentives and job security made the job of visitadora an attractive
employment opportunity for young women in rural areas of Brazil.
Using the data from a 1982 expenditures survey of the regional directorates,
Total
the study team calculated annual costs of each FSESP health post.
serves
an
average
population
of
annual costs per health post, each of which
The
training
costs
for
each
1,250 people, were estimated at US$ 18,208.
visitadora were only 0.5 percent of this total, The combined salaries of the
visitadora and a sanitary auxiliary. including their social security benefits,
accounted for 70 percent of the total costs of the health post.
The costs of direct supervision from the support unit were estimated to be 1.2
percent of the total annual costs of the health post. All of these costs were
covered by FSESP with funds from the Government of Brazil.
FSESP authorities
recognize that the cost of the high quality health services provided by the
5530 Wisconsin Ave. • Chevy Chase MD 20815 USA • (301)654-2550 • Cable: URCINTER • Telex: 64693
PR1COR is operated by the Center for Human Services for the Agency for International Development under Cooperative Agreement No. AID/DSPE-5920-A-00-1048-00.
107
visitadora could not be paid by the rural communities served and that
government subsidization is required to meet the needs of the "poorest of the
poor.”
While utilizing the service statistics available for visitadoras,
the
researchers discovered inefficiencies in FSESP’s health information system.
Although data collected by the visitadoras and their supervisors ‘ were. sent to
the central level and put on a computer, these data were not analyzed,
condensed, or appropriately displayed. Therefore, they were of little use for
evaluation or supervision.
As a result of the research, changes in the
information system were proposed to FSESP officials.
COMMUNITY FINANCING OF WATER SUPPLIES
A review of the development of FSESP’s system of water supply financing and
operation revealed that FSESP has assisted over 1,000 Brazilian communities to
FSESP has created municipal
install water supply and sewage systems.
systems,
authorities (SAAEs), with paid employees, to manage the water supply and
Where necessary, FSESP provides technical and economic
sewage systems.
This
approach
has been very successful and many SAAEs have generated
support.
surplus cash which' has been used to expand the systems.
The methods used to generate capital varied from community to community, The
most common methods were subsidies, external financing, community contributions
of labor and money, and some combination of these. Most smaller communities'
water systems are capitalized through loans from development banks, while loan
interest payback, operation, and maintenance costs were met through user
charges. Rates for utilization of water and sewage systems were based on the
minimum salary in Brazil and thus adjusted for inflation. Rates were set to
favor poor users rather than large commercial users.
The PRICOR team discovered that small communities in Brazil are able to pay
for both maintenance and operation of water supplies if a suitable financing
system is used. In the case of FSESP, the regular payment of water bills by
the community was achieved through the efficient management of the water
system by the SAAEs and proved to be an adequate financing mechanism for
operating costs.
The researchers think that expanding water supply coverage
depends more on the method and efficiency of billing and collection than on
the wealth of the community and the amount of charges for water.
The methods and results of the PRICOR study on community financing of water
supplies are being used to evaluate and revise the new national scheme to
develop water systems (PLANASA) in Brazil. A survey of the State Water Supply
Agencies is being conducted using the same approach used in the PRICOR study.
Results of this survey will be compared with those obtained in the PRICOR
study.
*
*
*
This study was conducted from January 1983 to April 1984 by the Johns Hopkins
University School of Hygiene and Public Health and the Fundagao Servigos
from
Especiais
de
Saude
Publica.
Further
information
is
available
Dr. Timothy Baker, School of Public Health and Hygiene, the John Hopkins
University, 615 North Wolfe Street, Baltimore, Maryland 21205, or from Dr.
David Nicholas, PRICOR study monitor (Chevy Chase).
108
PRI<S®R
CF
Brazil
Primary Health Care Operations Research
No. 23, November 1986
Study Abstract
COMMUNITY FINANCING OF PRIMARY HEALTH CARE IN RIO DE JANEIRO
K
An operations research study was undertaken by the Centro de Pesquisas de
Assistencia Integrada a Mulher e a Crianga (CPAIMC) of Brazil, a private
nonprofit organization, to identify and test alternative schemes for community
financing of primary health care (PHC) at selected CPAIMC health units and
miniposts.
Ten low-income communities of Rio de Janeiro served by CPAIMC
participated in the study.
The objective was to identify the PHC community
financing strategies (i.e., mixture of several community financing schemes)
for CPAIMC units and miniposts that maximize coverage and utilization of
services, cover the largest proportion of recurrent costs of service delivery,
and decrease CPAIMC's dependence on external financing sources.
To begin, CPAIMC identified four groups to share in decisionmaking regarding
the community financing
(CF) projects:
community residents, community
leaders, CPAIMC service providers, and a representative group of CPAIMC
managers, supervisors, and researchers called the "Group of Ten." The meeting
participants also generated a ranked list of alternative community financing
schemes and constraints using Nominal Group and Delphi techniques. From this
list the group constructed a preliminary matrix and identified major
information gaps.
To fill these gaps, CPAIMC researchers conducted two surveys of 10 low-income
Rio de Janerio communities:
one of the community and one of CPAIMC health
service providers.
Community leaders were also interviewed to get their
opinions and suggestions regarding community financing of PHC.
The surveys provided, useful information on community utilization of and
payment for health services.
Most households (93 percent) reported that at
least one household member is covered by public sector health insurance.
Government facilities provided the major source of health care for women age
Payment for health care was made for only 6
15 to 49 and childen under 5.
percent of the children and 8 percent of the women, based on most recent
visits.
Some women did pay for dental care (22 percent), injections (17
Most vaccinations, pediatric
percent), and gynecological care (7 percent),
care, well-child care, and emergency care, were attained free at public health
facilities.
Household members also paid for drugs and contraceptives from
pharmacies.
Solutions to the PHC financing problem were developed by the relevant decision
groups using preference and impact interaction matrices.
These matrices were
completed by all members of the Group of Ten and by all CPAIMC supervisors and
service providers working in nine communities. Consumers were represented by
the community survey data from which the research staff made inferences.
The
Group of Ten incorporated the opinions of community leaders into their matrix
entries.
5530 Wisconsin Ave. • Chevy Chase. MD 20815 USA • (301)654-2550 • Cable: URCINTER • Telex: 64693
PR1COR is operated by the Center for Human Services for the Agency for International Development under Cooperative Agreement No AID'DSPE-5920-A-00-104S-<X1
109
The final ranking of the financing schemes in each community was determined
from tabulation of these matrices.
These rankings were used to construct
financing strategies
(mixtures of several schemes) specific to each
. community. For example in Vila Alianca, the strategy consisted of payment for
selected services and visits, sale of prescribed drugs, collection of
registration fees, and contributions of funds and labor by a community
organization for the maintenance of a women's health care unit. For schemes
involving payment, group debates identified which services, visits, and drugs
would be sold and their prices.
The community and health care provider
surveys provided important input to this decisionmaking process.
Once final details of each strategy were defined, they were reviewed in a
meeting between community leaders and Group of Ten representatives. Schemes
or elements of schemes not considered acceptable to the leaders following
these discussions were deleted. Community leaders were asked to sign written
agreements detailing the responsibilities of both CPAIMC and the community in
implementing the CF strategy.
The individual community financing strategies were implemented in nine study
communities over a period of nine months and then evaluated, The evaluation
of each community financing strategy was based on three indicators of
success:
(1) its ability to generate revenue to cover PHC costs, (2) its
effect on service utilization, and (3) its effect on service coverage.
The mean monthly revenue generated by the schemes during the 9-month test
varied significantly among the nine communities, ranging from USS 121 to
US$ 4. The three most successful strategies covered approximately 11 percent
of total costs (20 percent of direct costs) of PHC service provision. The CF
schemes that generated the largest proportion of revenue in all communities
combined were payment for selected services and sale of prescribed drugs and
contraceptives.
The scheme components that generated the most revenue were
charges for Pap smears, contraceptive sales, physician visits, and drugs, in
that order.
In six of the eight service sites, the trend in total number of visits
remained unaltered following implementation of the CF strategies, while in two
of the sites it decreased significantly.
Only one association appears to
exist between individual financing schemes and changes in utilization trends:
in all five units where registration and Pap smear fees were charged there was
a significant decrease in the number of initial visits for women. The lack of
control communities in the evaluation of PHC service coverage precluded any
conclusions as to the effect of CF strategy implementation on coverage.
During a final community survey, information was collected from all households
regarding their knowledge of CPAIMC services and CF schemes, and their
opinions on PHC community financing. Nearly three-fourths of the respondents
stated that charging for services does not impede service utilization. Fully
65 percent of the respondents who knew of the CPAIMC service site in their
respective communities stated that community residents should help maintain
the CPAIMC unit/minipost.
*
i*r
*
This study was conducted from July 1983 through March 1986 by the Centro de
Pesquisas de Assistencia Integrada a Mulher e a Crianga (CPAIMC). Further
information is available from the principal investigator, Ms. Karen Johnson
Lassner, CPAIMC, Avenida Presidente Vargas 2863, 20210 Rio de Janeiro, RJ,
Brazil, or from Dr. Wayne Stinson, PRICOR study monitor (Chevy Chase).
110
1
PRISSR
ORT/CHW
Brazil
Primary Health Care Operations Research
Study Abstract
MOBILIZING TRADITIONAL HEALERS TO DELIVER ORT
In northeastern Brazil diarrhea is a major source of morbidity and mortality among
infants and small children.
In rural areas, traditional healers have long been the
first source of medical care for children suffering from diarrhea and other
illnesses. The healers are available to the community 24 hours a day and work out
of a desire to serve their community, without monetary incentives. A PRICOR study
showed that these healers can be effective in preventing and treating dehydration
and in reversing mothers’ harmful health practices, at a very low cost.
The
objective of the study, conducted by faculty from the Federal University of Ceara
and the University of Virginia, was to determine how best to mobilize and integrate
traditional healers into the official health system to clinically manage diarrheal
illnesses and to deliver oral rehydration therapy (ORT).
The study was conducted
i n-Pacatuba r a rural community of about 7,000 near Fortaleza, the capital of Ceara.
In the problem analysis, the researchers sought to understand the social, cultural,
and medical systems in which the traditional healers work. Data was collected from
surveys on the knowledge, attitudes, and practices of the community regarding the
treatment of diarrhea/dehydration and child morbidity and mortality due to
diarrheal diseases.
Ethnographic analyses were undertaken of the health care
delivery system and utilization patterns, and in-depth interviews were conducted to
reconstruct patterns of household response to a diarrhea illness episode.
I
These analyses revealed some interesting findings on the incidence of diarrhea in
the region and the possibility of using traditional healers to deliver ORT. First,
the infant mortality rate was high, at almost 150/1,000, with over half the deaths
due to diarrhea/dehydration.
Mothers widely perceived diarrhea as a "fright
disease”, or other supernatural malady which requires the intervention of the
traditional healer.
Seventy-seven percent of
mothers
representing
all
socioeconomic strata - first sought a traditional healer in cases of diarrhea.
Knowledge of oral rehydration therapy in the community was high, but lack of
service providers resulted in low utilization.
Finally, mothers experienced
serious problems with the modern health care system, including long waits, rationed
appointments, extensive travel, and expensive and improperly prescribed drugs.
During Phase II of the study, the researchers worked with traditional healers to
develop a strategy to involve them in the promotion and use of ORT. Group meetings
were held during which the traditional healers had their first opportunity to share
ideas with one another and participate in the formulation of a strategy for
incorporating ORT into their healing rituals. They also participated in choosing
an ORT recipe that was most acceptable to the target community.
Forty-six popular
healers in Pacatuba were trained in how to correctly prepare and administer ORS,
and to teach mothers how to give the solution to their children at home.
The healers were provided with the basic equipment needed to prepare ORS, including
measuring utensils, containers, and water filters.
In several cases the community
contributed by helping to build ’’curing rooms," simple mud and thatch room
additions on the healers' homes where they could treat patients. A manual for
5530 Wisconsin Ave. • Chevy Chase, MD 20815 USA • (301)654-2550 • Cable: URCINTER • Telex: 64693
PRICOR is operated by the Center for Human Services for the .Agency for International Development under Cooperative Agreement No. AID/DSPE-5920-A-00-1048-00.
111
instructing traditional healers was produced with substantial input from the
healers, and guidelines were formulated to assess their clinical competency in the
identification and treatment of diarrhea and dehydration. Healers were also taught
to identify and refer severe cases that do not respond adequately to ORT.
The testing of the strategy began in October 1984, and after 12 months of
activity, the impact of the traditional healers' efforts has been dramatic, A
comparison of the responses of 204 mothers with children less than 5 years old
before the PRICOR study with the responses of 226 mothers after the study showed a
highly significant increase in mothers who know about homemade ORS, from 3% to 72%
(p less than .001). Over half of the mothers surveyed had used the traditional
healers' ORS, with the greatest number among the poorest families. Moreover, the
traditional healers' promotion of ORT positively influenced mothers' feeding and
medication use behaviours during diarrheal episodes. After the intervention, the
number of mothers who believe they should continue breastfeeding during diarrheal
episodes increased by 20.5 percentage points (p less than .001) and the number who
believe that feeding should continue (not be withheld) increased by 18.0 percentage
points (p less than .01). A significant decrease of 25.5 percentage points (p less
than .0001) was shown in the number of mothers using expensive, commercially
prepared ORS packets. The use of pre-diluted ORS dropped by 11.6 percentage points
(p less than .01). The greatest percentage drops in the use of these expensive
methods occurred in the poorest neighborhoods, where the traditional healers'
homemade solutions enjoyed the greatest increase in popularity. A significant
decrease (from 93% to 63%) was also shown in the number of poorest mothers who
believe they must give phamaceuticaIs to a child with diarrhea/dehydration.
Traditional healers have demonstrated that they are capable of preparing safe salt
and sugar solutions and are effective at conveying the value of ORT to mothers.
The introduction of ORT through traditional healers did not change villagers'
medical beliefs about the causes of diarrhea, but rather strengthened the healers' •
role in the community by the incorporation of ORT skills.
A cost analysis of the intervention showed that the costs of incorporating
traditional healers into ORT delivery is quite low since the healers work without
salaries and because the community supplied much of the materials needed to
construct the curing rooms. The average cost of constructing a curing room was
USS 26.22, and equipping it for ORT, USS 43.15. The operating expenses for the
program, including biweekly supervision, salt, sugar and replacement supplies for
preparing ORS, averaged US$ 71.18 per month. The cost of sugar needed per month
per healer was only USS 0.48, suggesting that the costs of providing salt and sugar
for the traditional healers could be borne by the community.
Based on the successful experience with traditional healers in Pacatuba, the
researchers are planning for the incorporation of traditional healers into a new
large-scale child survival project that the Federal University of Ceara is
implementing in 33 municipios (counties) in Ceara with funding from Project HOPE
and AID.
*
*
*
This study was conducted from March 1984 to February 1986 jointly by researchers
from the Federal University of Ceara, Brazil and the Division of Geographic
Medicine in the Department of Medicine of the University of Virginia. Further
information is available from the principal investigators. Dr. Marilyn K. Nations,
Box 485, School of Medicine, University of Virginia Medical School,
Charlottesville, VA 22908, and Dr. Maria Auxiliadora de Souza, Caixa Postal 1674,
Aldeota 60.000 Fortaleza, Ceara, Brazil, or from Dr. David Nicholas, PRICOR study
monitor (Chevy Chase).
112
■PRI€®R
CF
Dominica
Primary Health Gate Operations Research
|Study Abstract
REVOLVING DRUG FUNDS IN DOMINICA
To increase the availability of essential drugs in its primary health care
(PHC) program and at the same time decrease the government costs of supplying
those drugs, the Dominica Ministry of Health (MOH) proposed the establishment
of a national revolving drug fund (RDF). To aid development of the planned
RDF, Management Sciences for Health (MSH) worked with the MOH using operations
research to identify key operational problems and issues related to design and
implementation of the RDF and to develop and test solutions for these problems.
The study began with a systematic analysis of Dominica's pharmaceutical supply
system.
This analysis identified the problems that impede effective and
efficient operation of the system. A preliminary systems model for the RDF
was available from the start of the study and was used as the basis of study
The model consisted of eight
planning and all subsequent activities.
finance;
management information system
(MIS);
selection;
components:
distribution;
warehouse/inventory
management;
organizational
procurement;
and public
knowledge,
attitudes,
attitudes,
and practices.
While
development;
I
issues
were
identified
and
solutions
developed
within
each
operational
component, the issues are interdependent and interacting, It is the composite
result—the sum of solution development in the various components—that
emerges from this study.
During, the course of the study it was decided that the RDF should be
implemented in two phases. During Phase One, the districts and independent
government health facilities were the clients, purchasing drugs and medical
supplies from the central medical stores (CMSs) with funds provided in their
budget. During Phase Two, the patients became the clients, reimbursing the
RDF through payments at government health facilities for drugs and medical
supplies consumed.
The MOH, through the CMSs, operated the RDF which was
initially capitalized with a $500,000 loan from the Social Security Fund.
The goal of Phase One was to develop and establish the capability of the CMSs
to successfully operate the national RDF. The objectives of this phase were:
to decrease the unit costs paid for drugs and supplies; to improve CMS
inventory and distribution management; to increase the availability of drugs
and medical supplies to health districts and independent government health
facilities; to increase cost consciousness on the part of the users (defined
in Phase One as health districts and independent government health
facilities); and to develop the financial management system at the district
level necessary to institute consumer payment. The objectives in Phase Two
were: to implement and evaluate consumer payment for drugs and supplies at the
district and independent government health facility level; and to increase
cost consciousness on the part of the consumer.
One of the major factors
:
Phase One has; achieved remarkable success.
!
the
decision
to implement it in two
influencing the success of the RDF was
as
researchers
to focus on central
the MOH as well
phases.
This allowed
;
5530 Wisconsin Ave. • Chevy Chase MD 20815 USA • (301)654-2550 • Cable: URCINTER • Telex: 64693
PR1COR is operated bv the Center for Human Services for the Agency for International Development under Cooperative Agreement No. AID/DSPE-5920-A-00-1048-00.
113
managerial
systems
for
both
financial
and materials
management
and
implementation of the MIS accounting system. The systems development efforts
required to introduce consumer charges for drugs on a national level seem, in
retrospect, nearly unattainable without the support of these central level
management systems.
This may be one of the clearest and simplest, yet most
important, findings from the study.
Introducing this preliminary phase has
the effect of stretching out the timeline for all of the management systems
development work required for an RDF, allowing the necessary attention to be
focused on central systems development rather than dispersed across a wider
range of issues.
The results of Phase One identified the following four issues as critical to
the implementation of an effective and efficient RDF.
Adequate Capitalization. The level of capitalization required is the product
of the monthly usage rate and the length of the pipeline, i.e. time required
for flow of drugs from supplier to user and flow of funds from user to
supplier.
If accurate data are not available at the outset, the information
system should be designed to collect the necessary data for continually
monitoring both usage and pipeline length, in order to continually refine the
capitalization estimate and make the necessary adjustments.
Adequate funds
for capitalization must be available if the RDF is to succeed.
Assurance of RDF Reimbursement.
Because all distribution of drugs and
supplies from the RDF must be reimbursed in order for the fund to revolve, two
variables - distributions and reimbursements - are important and must be kept
in equilibrium.
This means that if distributions are to increase, as is
happening in Dominica, additional funds must be found to reimburse the RDF.
Authority and responsibility for RDF
Coordinated Leadership and Management.
assets should be vested in either a single person or in a committee that meets
regularly to review financial reports, address operational problems, and make
managerial decisions.
Adequate Staff.
Initial design of the RDF, and in particular of the MIS
accounting system, requires the advice of an expert accountant.
RDF
maintenance requires all the staff normally required at a CMS to maintain the
supply system, plus an RDF accountant.
This accountant will maintain the
accounting books and will ensure maintenance of other aspects of the MIS that
provide information for RDF management; he or she should report to the RDF
manager or RDF management committee.
Phase Two has not been implemented, although some Phase Two operational issues
in some components have been addressed.
District physicians and pharmacists
are monitoring the value of the usage of drugs and supplies against their
budgetary allocations and are beginning to manage their inventories by
monitoring patient usage.
The physicians and pharmacists are also beginning
to discuss the value of drugs with patients.
Usage monitoring and patient
education are prerequisites to a successful RDF with consumer drug sales.
*
★
*
This study was conducted from April 1983 to March 1986 by Management Sciences
for Health (MSH) and the Dominica Ministry of Health. Further information is
available from Mr. Peter Cross, MSH, 165 Allendale Road, Boston, MA 02130;
Dr. Desmond McIntyre, Health Services Coordinator, Ministry of Health, Roseau,
Dominica; or Dr. David Nicholas, PRICOR study monitor (Chevy Chase).
114
PRI®®R
ORT/CF
Dominican
Republic
Primary Health Care Operations Research
Study Abstract
IMPROVING DISTRIBUTION OF ORS IN THE DOMINICAN REPUBLIC
Diarrheal diseases arB a leading cause of death among infants in the Dominican
Republic, where approximately 550,000 children are under 3 years old.
A
nationwide government campaign to promote the use of oral rehydration therapy
(ORT) was being planned for 1984.
The purpose of the PRICOR study was to
assist the government to ensure the supply and distribution of oral
rehydration salts (ORS) nationally and to determine an appropriate price for
the packets. The study was thus concerned with developing solutions to a
number of problems in the supply and distribution of ORS.
The study began with an inventory of the available health statistics to
determine the incidence of diarrhea.
However, the quality of the existing
data was not high enough to support sound planning. It was therefore decided
to conduct a survey to collect basic data on the incidence of diarrheal
diseases and on the degree to which ORT was currently being used.
This
information was crucial for the development of solutions and to the government
in planning a distribution program.
The survey, which examined records of 27 hospitals, 21 regional subcenters,
and 91 rural clinics, gathered data on mortality, morbidity, and use of ORT.
In the rural areas socioeconomic data and information on health practices was
In the end, the survey data proved unreliable and, therefore,
also collected,
the solution development as originally hoped.
However, it
not as useful to
•
was the only data available, so some tentative conclusions were drawn for
general planning purposes.
A quantitative inventory model was formulated to determine the best way to
'
■
1 F store, and distribute ORS.
The application of this model
order,
transport,
required both information from the government on how it proposed to distribute
The information
information from
from the
the
ORS and data from the incidence and coverage survey. The
government was not available, and the model was applied using only the general
conclusions drawn from the survey.
The resulting recommendations emphasized
the importance of developing nongovernmental supply channels and using as many
kinds of organizations as possible, including private commercial distribution.
The study team developed a theoretical pricing model to determine the best
pricing strategy for ORS.. The principal operational problem was to determine
a price that was 1high enough to cover most costs and a small margin for the
Due to the unreliable survey
retailers and yet low enough to be affordable,
data, this model was applied using income and expenditure studies from the
5530 Wisconsin Ave. • Chevy Chase, MD 20815 USA •'(301)654-2550 • Cable: URCINTER • Telex: 64693
PRICOR is operated by the Center for Human Services for the .Agency for International Development under Cooperative Agreement No. AID/DSPE-5920-A-00-1048-00.
115
central bank and time series data on prices and sales of essential medicines.
The study assumed a two-tier pricing system.
It recommended a price per
packet of US$ .08 for the low and moderate-income market and a price of
US$ .30 for the high-income market. Approximately 6.5 million packets would
be made available each year.
This would be enough to provide universal
coverage in the D.R. assuming the use of four ORS packets per episode of
diarrhea.
The private pharmacies would provide about 25 percent of the
packets and the government 75 percent.
The design of this study depended heavily on input from the Government of the
Dominican Republic. However, since the resources and political support to
move ahead in planning the national ORT program were not available, the
government was unable to supply the necessary information needed to complete
the PRICOR study. Nonetheless,
Nonetheless, recommendations based on the results of the
technical analyses were made to the government in the areas of management
information systems, inventory and distribution management. and pricing,
These recommendations might serve as a framework of basic ideas and
information for developing a complete distribution plan in a future ORT
project.
*
*
★
This study was conducted from December 1983 through April 1985 by the San
Clapp and Mayne, Inc., in cooperation
Juan, Puerto Rico consulting firm of
the Dominican
with the Secretariat of Health andSocial Assistance in
Further
information
is
available
from
Dr.
Jose
A.
Herrero,
Km. 25,
Republic.
the
Duarte,
Santo
Domingo,
Dominican
Republic;
from
the
principal
Autopista
Alan Udall, Clapp and Mayne, Inc., 1606 Ponce de Leon
investigator. Dr. Alan
Avenue, San Juan, P.R. 00909, or from Dr. Jack Reynolds, PRICOR study monitor
(Chevy Chase).
116
PRI<S®R
CHW/CF
Haiti
Primary Health Core Operations Research
Study Abstract
ALTERNATIVE METHODS OF MOTIVATING COMMUNITY HEALTH WORKERS
The Ministry of Health of Haiti has adopted a strategy of primary health care
(PHC) to achieve the goal of "Health for all by the Year 2000." The success
of this strategy depends largely on the ability to recruit and support
government health workers, or non-government community health collaborators,
who will provide community-based preventive PHC services. The recurrent cost
of financing these peripheral-level workers is too high for either the
government or private institutions to absorb.
Save the Children Canada
(AEDC), in collaboration with institutions associated with the Haitian
Association of Voluntary Health Institutions (AOPS), conducted a study during
the period 1984 - 1986 to examine the best ways of motivating the CHWs to
provide preventive services that would encourage mothers to learn about, use,
and maintain their competence in child survival interventions.
Given that subsidies from institutions managing the, CHW programs were not an
acceptable option, the researchers began by identifying alternative community
financing mechanisms. These included: (1) funds from the community (fee-forservice at
rally posts, contributions from existing community groups,
(2)
volunteers;
and
(3)
revenue-generating
prepayment for
services);
activities.
Data were collected in community surveys and intensive case studies to better
understand the motivation problem and the feasibility of the proposed
alternatives.
Results from the surveys, carried out in three rural areas,
showed that people are not willing to pay for preventive
services.
Communities perceive curative care as their primary health need and CHWs in
the rural areas do not provide curative care.
In fact, people perceive health
care as a "service" delivered by an outside agency. While people understand
the value of health promotion, they expressed no willingness to finance a
health care delivery system.
Given these constraints, the researchers and
consultants then systematically evaluated the options using a multiple
criteria utility assessment.
With regard to community funds, increasing revenues from fee-for-service
activities at the rally posts (where health services are delivered from mobile
units), was not acceptable to local institutions, This option was eliminated
because the doctor-oriented, "curative"
’’curative" focus would detract from the four
child survival interventions, and it would not generate <enough money to
Local community
regularly pay the CHWs (as demonstrated in Mirebalais).
However, most community
groups often pool
oool resources for special projects.
groups agreed that although they appreciate the work of the CHWs and would
like to encourage it, they did not normally have enough funds or community
The prepayment
support to use existing funds to regularly finance a CHW.
scheme was not feasible because people were n'ot willing to pay for preventive
health services.
5530 Wisconsin Ave. • Chevy Chase, MD 20815 USA • (301)654-2550 • Cable: URCINTER • Telex: 64693
PR1COR is operated by the Center for Human Services for the Agencv for International Development under Cooperative Agreement Na AID/DSPE-5920-A-00-1048-00.
117
The second alternative for providing preventive services was volunteers.
Existing volunteer projects were examined and it was found that they shared
the common characteristic of being discrete activities of short duration,
None were ongoing activities.
Volunteers did not seem a likely source of
energy for the preventive health work, The third major alternative considered
was revenue-generating activities that would produce enough profit to
The major problem with these activities
reimburse the CHWs for their work,
was that they usually did- not provide incentives to the CHW to do preventive
services.
The solution the group found most likely to motivate the CHWs to do preventive
tasks was a combination of the prepaid scheme, the existing community groups.
It is based on traditional Haitian credit
and a revenue-generating activity.
associations called "cengle” or •"solde."
’soldo. ” In these traditional rotating credit
schemes, friends contribute a fixed amount of money each month to a general
fund and take turns receiving the entire pool.
In the proposed health
financing, scheme, groups of mothers who can demonstrate competence in the four
child survival interventions and whose chldren are fully immunized and
These
participating in growth monitoring will be eligible to participate,
women, organized in small groups by their natural friendship networks, each
pay an annual fee for her health card (which is used to support the CHW) . The
group decides the monthly contribution each person must make depending on how
much they know that person can pay. The monthly contributions are used by the
women as in a traditional cengle. However, the real attraction of the health
card is not the cengle but the access that the affinity group then has to
low-interest loans from the Bureau de Credit Agricole (BCA) for income
generating activities. The pooled monthly member contributions are matched by
a one-time grant from the institution sponsoring the CHW program.
This
matching grant is used as capital for the loan, kept at BCA. BCA will lend
four times that amount to the group for income-generating activities.
A
counselor from the BCA will help the affinity groups develop feasible projects
and a payback schedule, The affinity group is an essential component of the
scheme as its cohesion is the reason people will be motivated to continue
paying into the fund.
This project appears to have produced favorable results for a number of
reasons. Credit schemes such as these are attractive to rural residents
because there is great demand for credit, and private-source interest rates
are very high.
Because low-interest credit is so desirable, there is an
economic incentive for the mothers to learn about the health interventions in
order to have access to the pooled funds. As the CHW salary is based on the
number of mothers who qualify for the health cards, the financing scheme
cleverly links the promotion of preventive health interventions with revenue
Some of the groups have developed successful income-generating
generation,
One
group bought a mature mango tree from which they will harvest
projects.
mangos.
Another group bought a goat with their loan and have
and sell
recently acquired another with their monthly contributions.
*
*
*
This study was conducted from December 1984 through March 1986 by the Alliance
pour 1'Enfance et le Developpement Communautaire, in collaboration with
institutions associated with the Association des Oeuvres Privees de Sante.
Further information is available from the principal investigator. Dr. Antoine
Augustin, Alliance pour 1'Enfance et le Developpement Communautaire, 3 Ruelle
Duncombe, Port-au-Prince, Haiti, or from Ms. Marty Pipp< PRICOR study monitor
(Chevy Chase).
118
fRI<S®R
Honduras
Primary Health Gore Operations Research
|Study Abstract
FINANCIAL ALTERNATIVES TO SUPPORT EXTENSION OF
BASIC HEALTH SERVICES IN HONDURAS
Seeking alternative means of financing basic health services, researchers from
the Honduran Ministry of Health (MOH) and Management Sciences for Health (MSH)
conducted an operations research (OR) study in several regions of Honduras.
This study focused on identifying alternatives for financing primary health
care services and on assessing the viability of selected alternatives at the
communi ty 1eve1.
The researchers used three separate methods to determine how primary health
care (PHC) services were being financed: (1) intensive observation of a small
number of Honduran families; (2) a household survey in four Health Regions;
In
and (3) case studies of existing community-financed health care programs,
addition, interviews were held with community leaders, cooperatives and
syndicates, and key government officials to identify the range of alternatives
considered feasible.
Observation of Families. A locally-trained observer visited 25 households in
urban slums of Tegucigalpa during October and November 1982.
Direct
observations of these families revealed that they regularly use multiple
sources of medical treatment for illness, often at the same time. Observation
of specific illness episodes was particularly useful in identifying this
phenomenon.
Household Survey. The survey team completed 1,017 household interviews in 29
sampling sites in four Health Regions of Honduras during the first 3 months of
1983.
Within these households, 1,648 illnesses were reported for the 15-day
recall period and 910 interviews were conducted in households that had
experienced illness.
Approximately 26.1 percent of the sample population
reported having been sick in the past 15 days.
Nearly 46 percent of these
illness episodes had been treated at home.
The mean expense for illness
episodes treated at home was approximately US$ 2.50. Most of this amount was
for the purchase of medications.
Reported monthly household health care
expenditures averaged US$ 20.82 (11.4 percent of total reported monthly
expenses) and were the third highest category after food and clothing.
Ninety-four percent of the respondents reported a willingness to pay for MOH
services.
Case Studies.
The study team completed 10 case studies of clinics, health
centers, or hospitals in Honduras that were either charging for services or
collection,
the
Following
receiving
active
community
support.
data
financial
scheme
researchers developed concise analytical descriptions of the
5530 Wisconsin Ave. • Chevy Chase, MD 20815 USA • [301)654-2550 • Cable: URCINTER • Telex: 64693
PR1COR is operated bv the Center for Human Services for the .Agency for International Development under Cooperative Agreement No. AID/DSPE-5920-.A-00-1048-00.
119
and the effects of community support at each health care facility visited.
The results of the case studies generally reinforced what had been learned
from the household survey.
In all 10 case studies, patients showed a
willingness to pay for or otherwise support health care services, provided
they thought the services were of high quality.
Alternative financing mechanisms were developed using the
information
collected through the observation of families, the survey, and the case
studies, and from interviews with key personnel in the MOH and other
health-related agencies.
Two analytical techniques, decision flowcharts and
P/N/I force field analysis, were used by a multidisciplinary group to evaluate
the financing alternatives. The four alternatives considered most appropriate
were:
(1) standard fees-for-services (cuotas), (2) payment for medications,
(3) community contributions of labor for construction and maintenance of
health centers, and (4) rotating drug funds managed by local health
committees. These alternative solutions were recommended to the MOH.
This study resulted in the adoption of MOH community financing policies. The
data on common illnesses, illness behavior, and expenditures on drugs,
gathered for the PRICOR study, convinced decisionmakers in the MOH of the
importance of making low-cost, high-quality, essential medications available
to all Hondurans.
Several mechanisms (some of them proposed by the PRICOR
study) are now under study or are ready for field testing.
These include a
system of "popular" pharmacies; community rotating funds for medications;
commercialization of oral rehydration salts (ORS); and expansion of Lab-PANI,
a semi-autonomous, government-sponsored pharmaceutical production facility.
There has been, however, a considerable gap between the adoption and
implementation of these policies to extend the availability of basic drugs.
As an indirect result of this study of PHC financing alternatives, the MOH has
instituted a policy that encourages hospitals to recover a greater portion of
their operating costs.
The MOH has directed all hospitals to increase
recovery to a minimum of 30 percent of their annual operating costs through
fees-for-services from patients.
In 1984 hospitals recovered an average of
4.2 percent of their annual operating budgets, a 30 percent increase over 1983.
The operations research technique itself had significant impact on the way the
Honduran MOH makes decisions.
Because of the satisfaction of key MOH
personnel with OR as a decisionmaking tool, a Science and Technology Unit has
been created within the MOH to coordinate OR studies.
Ongoing OR studies
a
include: a study on the effectiveness of trained vs. untrained TBAs;
feasibility study of the commercialization of ORS; and a comparison of
alternatives for expanding hospital services without reducing PHC services.
A
*
*
This study was conducted from August 1982 through December 1984 by the
Ministry of Health of Honduras and Management Sciences for Health (MSH), a
nonprofit organization based in Boston, Massachusetts, Further information is
available from the principal investigator. Dr. Frederick Hartman, MSH, 165
Allandale Road, Boston, Massachusetts
02130,- or from Dr. Jack Reynolds,
PRICOR study monitor (Chevy Chase).
120
PRI€®R
CF/CO
India
Primary Health Care Operations Research
Study Abstract
AN OPERATIONS RESEARCH STUDY OF FINANCING, ORGANIZATIONAL,
AND MANAGERIAL PROBLEMS OF COMMUNITY HEALTH PROJECTS IN INDIA
To meet the complex demands of grossly underserved rural populations of India,
numerous voluntary agencies have started community health projects in the last two
decades.
In their efforts to work out effective methods and processes of
delivering health care, these projects have had to develop solutions to a number of
financial, managerial, and organizational problems.
This study, conducted by
researchers from the Institute of Rural Management, Anand (IRMA) in India, and
U.S.-based Management Sciences for Health (MSH), examined eight such voluntary
health projects and identified the most important operational problems they have
faced, along with solutions and results achieved.
The study focused on financing, community organization, and personnel management
problems..To gather information on the eight health projects, the researchers
interviewed the directors and randomly—selected staff from the projects as well as
community leaders, practitioners, and a sample of community members in villages
served by the projects.
Focus group discussions were also held with staff and
community members.
Case studies were conducted by soliciting information from
management and by using currently available service statistics of the projects and
literature published about or by the projects.
After data analysis, the
researchers organized a three—day workshop in Anand for representatives of the
eight study projects, other voluntary health projects, the Voluntary Health
Association of India, and international donor agencies. The purpose of the
workshop was to discuss the priority problems and alternative solutions identified
through the study, point to unresolved issues, and make recommendations.
Financing.
The PRICOR investigators found that three major approaches had been
used to finance the health projects.
First, all projects raised some funds and
in-kind assistance from outside the community for starting up the project. Half of
the project directors felt that external resources were absolutely essential at
this point. Second, the projects obtained resources for recurring expenses from a
variety of local and external sources. Local sources included donations of land,
labor, and buildings; fees for drugs and services; and membership and insurance
Third, the
The government was an important source of external funding.
fees.
To
increase
efficiency,
projects pursued a policy of relentless cost containment,
the eight health projects used various strategies to economize on buildings.
Strategies included using existing
transport, and the purchase and use of drugs,
buildings; allowing project personnel to own <and maintain two-wheeled vehicles; and
purchasing drugs in bulk at competitive prices.
The single, most difficult financial problem cited by the project directors was how
to achieve greater community self-reliance.
In particular, how can projects
involve community members in contributing toward the cost of preventive health
services which are usually disassociated from current health problems and in
developing resources for treating incurable diseases?
Local organizations proved
to be important resource bases.
5530 Wisconsin Ave. • Chevy Chase, MD 20815 USA • (301]654-2550 • Cable: URCINTER • Telex: 64693
PR1COR is operated by the Center for Human Services for the .Agency for International Development under Cooperative Agreement No. AID/DSPE-^920-A-00-1048-00.
121
Community organization. Based on the experience of the eight health projects, the
researchers concluded that starting with previously existing or related community
organizations instead of new organizations helped achieve broad coverage more
quickly and more easily — at least over the short run.
However, the fast and
efficient start achieved through existing organizations did not necessarily result
■‘
in greater breadth of participation in the
long run.. There is some evidence that a
represents
only part of the community. may
pre-established organization, if it i .
perpetuate restricted participation.
Overall, multi-purpose organizations demonstrated an advantage in breadth of
participation, i.e.. high percentages of community members who participated.
However, in terms of depth of participation (community involvement in project
planning, management, and service delivery), only newly organized multi-purpose
organizations achieved greater participation. In previously existing multi-purpose
organizations, members tended to assume fewer active roles.
In support of multi-purpose organizations, some respondents explained that their
projects had achieved wide participation by starting with a strongly felt need in
On the
the community, even if that need was only indirectly related to health,
potential
other hand,
adherents of single-purpose organizations warned of
bankruptcy of multi-purpose organizations, where failure of a single activity could
drag down the entire organization.
Community poverty was a real barrier to the projects in achieving broad community
participation, but so was wealth.
Not only did poor people refrain from joining
health schemes, but affluent community members gravitated to private physicians.
Serving the entire spectrum of economic strata therefore required, on the one hand,
services such as hospital referral to attract the wealthy, and on the other hand,
services such as food supplementation and income generation to attract the poor.
Personnel Management.
The researchers concluded that turnover and salaries are
likely to be lower and dedication higher among locally trained staff as opposed to
professionally trained personnel. The experience from the projects suggested that
it would be wise to train local personnel to undertake many tasks which would
otherwise be performed by professionals. They also felt that those professional
staff who are recruited for health projects should learn the skills of the support
staff to lessen social distance and reduce dependence.
For example, health workers
could be allowed to drive and maintain their own two-wheeled vehicles instead of
riding as passengers with drivers of more expensive four-wheeled vehicles.
During the workshop, channels for further dissemination of information were
identified and plans made for networking among nongovernmental community health
projects.
The sponsoring institution, IRMA, will distribute the PRICOR study
findings widely so that other voluntary health projects facing similar problems
might benefit from solutions that have been effective elsewhere.
IRMA is also
considering proposals to include training in the management of rural community
health programs in its curriculum and to provide opportunities for student
internships in the management of volunary health projects.
it
*
■k
This study was conducted from January 1985 through March 1986 by Management
Sciences for Health (MSH) in collaboration with the Institute of Rural Management,
Anand (IRMA), Gujarat, India.
Further information may be obtained in India from
Professor D. Nagabrahmam (IRMA), and in the U.S. from the principal investigator.
Dr. Henry Elkins, Jr., MSH, 165 Allendale Road, Boston, Massachusetts
02130, or
from Dr. Jeanne S. Newman, PRICOR study monitor (Chevy Chase).
122
^R|g®R
CF/CHW
Jamaica
Primary Health Care Operations Research
"Study Abstract
THE STUDY OF PRIMARY HEALTH CARE
TEAMS IN JAMAICA
Study researchers investigated productivity problems of primary health care
(PHC) teams in Jamaica and developed a resource allocation planning model,
Trial computations of the model, run for two districts in the Cornwall Region,
have provided projections on demand for service, personnel needs, cost of
personnel, and clinic hours and configurations, These computations have shown
not only that services and population coverage can be significantly increased
but that personnel costs can at the same time
tl— be
L- decreased.
PRICOR undertook this study at the request of the Health Management
Improvement. Project (HMIP) whose objective! is to improve the delivery of PHC
In Jamaica,
by improving management within the Ministry of Health (MOH).
and
curative
services
are
provided
by
three
types
of health
preventive
These
centers
range
from
small,
simple
centers
offering
few
centers.
to
large,
sophisticated
centers
providing
full
medical
and
dental
services,
The centers are linked by a patient referral system and by staff
services.
visits from the larger to the smaller centers.
The Department of Social and Preventive Medicine of the University of the West
Indies (UWI) and Price Waterhouse Associates, Management Consultants, Jamaica
were identified as the research teams. They, together with PRICOR and HMIP,
developed a study protocol and four study objectives:
(1) to develop a
methodology for measuring the productivity and cost effectiveness of PHC
teams; (2) to describe how the various categories of personnel in the health
center distribute their working time among certain predefined activities;
(3) to determine how productive and cost effective PHC teams are now and how
these factors relate to coverage of the population with essential services;
and (4) to develop strategies to improve productivity and to work with the MOH
to implement selected strategies on a trial basis.
Researchers observed 496 randomly selected workers representative of the
various types of staff working in health centers -(HCs) throughout the
country. These observations, part of a work sampling survey, showed that the
Based on
nonproductive time of health workers ranged from 26 to 66 percent,
Price
Waterhouse
(PW)
and
the MOH
similar studies done in the private sector, r.l—
—I—~
expect that nonproductive time should be kept below 25 percent. The PW team
developed a productivity index which was determined by comparing the actual
output of a clinic to the expected output given the critical work station time
available. Analysis showed wide variations in productivity indices, ranging
A cost index was also determined by comparing
from 30 to 150 percent.
personnel cost units used in serving patients to the personnel cost units
5530 Wisconsin Ave. • Chevy Chase, MD 20815 USA • (301)654-2550 • Cable: URCINTER • Telex: 64693
PR1COR is operated bv the Center for Human Services for the .Agency for International Development under Cooperative Agreement No. AID/DSPE-?920-A-00-1048-00.
123
allocated to scheduled clinics or HCs. The analysis showed an average cost
index of 46 for scheduled clinic hours and of 20 percent for HCs. The low
cost indices suggested that there is great opportunity to improve the
efficiency of the HCs.
The most
Productivity broadly defined is the effective use of resources.
Manpower
allocated
expensive resource employed in the PHC sector is manpower,
to an HC is tantamount to creating service capacity. Therefore, such capacity
should be based on expected demand for services and on time required to
deliver a unit of service. A model was developed that improved productivity
through manpower allocation and clinic rescheduling.
This model helped to
optimize productivity by determining the demand for the major services offered
at the health center, based either on projections of ideal coverage or past
trends; the mix of services the various clinics at a HC should offer; the
number of clinic hours that should be programmed in a year for the various
clinics to meet demand; the scheduling of the clinics; the number of each type
of personnel that should be allocated to a HC; and the assignment of critical
personnel (expensive and in limited supply, such as doctors) to geographic
clusters of HCs to maximize their use. Thus the model can help PHC managers
adjust their system to make optimum use of available resources.
The model
also allows a sensitivity testing for each of the variables.
The model is now being tested in two pilot districts in the Cornwall region,
where the test will be used to determine the best ways to restructure clinics
in those districts and to reallocate personnel. Changes in productivity will
be closely monitored over a 6-month period. If the test is successful, the
plan is to implement the model nationwide as a dynamic planning tool for
resource allocation.
*
*
*
This study was conducted from June 1983 through March 1986 by the Department
of Social and Preventive Medicine of the University of the West Indies, and
Price Waterhouse Associates, Management Consultants, Jamaica, with the
cooperation of the USAID-funded Health Management Improvement Project in
Jamaica and the Jamaican Ministry of Health. Further information is available
from Ms. Patricia Desai, Department of Social and Preventive Medicine,
University of the West Indies, Mona, Kingston 7, Jamaica; from Mr. Bobby
Zachariah, Price Waterhouse Associates, Box 372, Kingston, Jamaica; or from
Dr. David Nicholas, PRICOR study monitor (Chevy Chase).
124
PRI€®R
CO/CHW
Korea
Primary Health Care Operations Research
Study Abstract
PRIMARY HEALTH CARE PROGRAMS IN KOREAN RURAL COMMUNITIES WITH
THE SUPPORT OF EXISTING VILLAGE ORGANIZATIONS
Korean researchers used an operations research approach to address the
problems caused by insufficient attention to preventive and promotive health
activities in rural areas of the Keyonggi Province of Korea. The objective of
their study was to contrast the! feasibility and effectiveness of Primary
Health Care (PHC) programs that use non-health community organizations as
support structures.
The researchers initially went into the study areas (three counties in
Keyonggi province) and developed good working relationships with community
leaders, community health practitioners (CHPs), and officials of local,
provincial, and central government. This was accomplished through multiple
Channels of communication
visits, meetings, telephone calls, and letters.
established
between
CHPs,
the
community
people,
and referral facilities.
were
This set the stage for working closely with the community.
Baseline data were collected through a household survey, analysis of CHP
activity records, and surveys that focused on the characteristics of CHPs,
community leaders, and the study villages. Several problems related to the
health care system. health care utilization, and leadership capacities of the
communities were discovered: (1) The emigration of the younger generation to
urban areas caused many hardships for those people left in the rural areas;
(2) A high birth rate indicated the need for maternal and child health (MCH)
and family planning services;: (3) The morbidity rate was high and many people
on self-diagnosis and purchase of drugs at pharmacies; (4) CHP
were relying
i
activities were mainly concentrated on clinic-based medical care services for
minor illnesses; (5) Community leaders had limited knowledge about health care
and did not serve as good role models for the community in health matters.
Based on these results and those of earlier studies, the researchers wanted to
determine if community organization support of community health practitioners
would improve the quality of PHC delivery.
They decided to test the
effectiveness of informal community organizations vs. formal community
organizations in this role. In the study areas where CHPs were to work with
community organizations, the following interventions were implemented:
1) further training of CHPs with a focus on outreach activities and
preventive/promotive* care; 2) establishment of better communication between
the CHP and the community through existing community organizations;
3) education of community health leaders (CHLs) about the importance of
community participation in PHC; and 4) development and training of community
leaders to serve as communicators, health educators, motivators, and health
care providers.
5530 Wisconsin Ave. • Chevy Chase, MD 20815 USA • (301)654-2550 • Cable: URCINTER • Telex: 64693
PR1COR is operated bv the Center for Human Services for the Agency for International Development under Cooperative Agreement No. AID/DSPE-5920-A-00-1048-00.
125
A quasi-experimental research design (two experimental counties and one
control county) was used to test the effects of these strategies.
Experimental group I (Yang Pyong county) incorporated Bansang—Hoe (formal
village organizations) as substructures for PHC services.
Experimental
group II (Icheon county) incorporated various informal community groups, such
as mothers' clubs, 4-H clubs, church groups, agricultural cooperatives, clan
meetings, and development committees, as substructures for PHC services.
In
the control group (An Sung county), CHPs continued to provide PHC services
without the assistance of any community organization.
The first project input in the experimental counties was the training of CHPs
in field-based activities through 3-day workshops. After their own training,
CHPs trained leaders of the community organizations to serve as CHLs. CHLs
and CHPs were later trained to fill out monthly checklists of their PHC
activities. The research team also developed educational leaflets with health
messages for use and distribution by the CHPs and CHLs.
After the CHLs had been active in the experimental counties for several
months,, the researchers carried out a process evaluation.
One-third of all
CHLs in Yang Pyong and Icheon were interviewed, as well as a sample of
community residents in both counties.
In response to the survey, the CHLs
said that they had tried to inform the community about the activities of the
CHP and tried to inform the CHP about health problems in the community. They
also indicated that they ha4 discussed MCH, family planning, chronic health
problems, and health hazards with community people when the topics came up in
conversation.
The CHLs complained, however, that their new role was not
recognized by the community.
This discouraged the CHLs from performing the
health education and information tasks assigned to them. In response to this
problem, the CHPs conducted educational sessions for villagers on the CHLs'
role and the study team held a two-day workshop for CHPs to bolster their
enthusiasm.
After a period of 15 months, the researchers carried out a final evaluation in
order to compare the effectiveness and efficiency of the PHC services
delivered by CHPs in the three study counties. Each CHP program was evaluated
according to:
(a) the effectiveness of the program in terms of selected
health and health service indices;
indices; (b) productivity of CHPs and CHLs in terms
of quantity of services; and (c) efficiency in terms of cost incurred per
population served.
The comparisons were made using three different methods:
simple statistical analysis, a computer simulation technique, and a cost
effectiveness analysis.
The major conclusion of the study was that PHC services that incorporate
formal community organizations are more effective and efficient than both ■
those
that
incorporate
informal organizations and those that do not
incorporate any community organizations.
All three methods of data analysis
confirmed this general result.
*
★
*
This study was conducted from January 1983 to January 1986 by the Department
Further information is available
of Nursing of Seoul National University.
University,
from the principal investigator. Dr. Yeo Shin Hong, Department of Nursing,
College of Medicine, Seoul National University, 2 8, Yundeun-Dong, Chongno-Ku,
Seoul, Korea, or from Dr. Stewart Blumenfeld, PRICOR study monitor (Chevy
Chase).
126
PRI®g)R
CF
Liberia
Primary Health Care Operations Research
Study Abstract
PLANNING AND EVALUATING COMMUNITY FINANCING
IN LIBERIA
Although considerable resources, both public and private, have been devoted to
services
often
improving primary health care (PHC) services in Liberia, these
1--- ------ ---In response to this
cannot be sustained when external funding is withdrawn,
problem the Liberian Ministry of Health (MOH), in cooperation with the
Christian Health Association of Liberia (CHAL), conducted an operations
research study from 1984 - 1985 to develop effective ways for communities to
generate funds to finance some or all of their PHC services.
The financing schemes had to satisfy certain conditions: (1) they must cost
less than the income generated;
(2) a high proportion of the target
population (children under 5 and pregnant and lactating women) must use the
services provided; (3) a large fraction of households must contribute; and 4)
the villagers themselves must be able to sustain the financing scheme(s)
developed.
Problem analysis began in January 1984 when three villages were selected to
participate in the study. A Health Service Utilization Survey was conducted
to generate demographic data on these villages.
The survey found that
households were headed by older men (age 55+), most of whom were farmers with
no formal education, and the average household income was considerably less
than the national average of $280 per year.
In each village the town council and other village leaders met to consider
four issues: what health care services would be provided, who the health care
provider would be, who would participate, and how much the services would
cost. Each village established a village health committee (VHC) to manage the
project and chose a member of that committee to be trained as the community
health worker (CHW).
A major health care concern of the villagers during
these discussions was the availability of drugs. A large revolving drug fund
at the Kolahun District Health Center was seen as a resource and possible
model for revolving drug funds at the village level.
Community leaders in each of the study villages discussed eight alternative
financing schemes for generating PHC funds within the community. Each of the
three study villages constructed a preference matrix and, on each, the same
four schemes ranked highest, although in different order: (1> drug sales, (2)
production-based prepayment,
(3) community and individual labor, and (4)
donations and ad hoc assessment. During a 12-month field test, the villages
were successful in partially financing their PHC services through their chosen
5530 Wisconsin Ave. • Chevy Chase, MD 20815 USA • (301)654-2550 • Cable: URCINTER • Telex: 64693
PR1COR is operated bv the Center for Human Services for the Agency for International Development under Cooperative Agreement No. AID/DSPE-5920-A-00-1048-00.
127
Revolving drug funds were established and managed by the VHC, and
schemes.
monthly supervisory visits to each village were made by the principal
investigator or the research assistant, Training and logistical support were
provided by CHAL and MOH.
Ad hoc assessment and drug sales were highly successful in raising sufficient
funds to establish and sustain village-level revolving drug funds.
Ad hoc
assessment provided seed money of $59.00 to $209.60 in each village, with 75
to 90 percent of the households participating in these assessments.
The
revolving drug funds have been sustained by means of a 25 percent markup at
the village level on drugs obtained from the Kolahun District Revolving Drug
scheme. Stocks of drugs have been increased and diversified while drug costs
have remained low.
Sales revenues per case treated by CHWs range from $0.27
to $0.95.
Production-based prepayment and labor in communal rice fields have not yet
provided any direct resources for PHC funds but might be called upon if
needed. Some community labor has been provided in the fields of individual
CHWs, although the CHWs have indicated they prefer to be paid a salary.
The
assumption by the investigators and village leaders that CHWs would be
supported by the traditional application of community labor, as is the case in
the compensation of traditional healers, has been recognized as erroneous.
The project's success and replicability have been amply demonstrated by.the 10
Six of these have revolving
villages that now participate in the project.
drug funds, and in another four, CHWs have been selected and are now being
trained; plans for revolving drug funds in these four are underway.
Additional funds and support raised through community labor, donation, and
farm contract work have been applied to the direct support of PHC.
The
results for those communities participating in the full scheme include locally
available health care providers, drugs, and PHC management. The establishment
of the village health committees has also provided the framework for continued
efforts to improve PHC. Both CHAL and the MOH have agreed to continue their
support for these efforts, and the district medical officer will continue to
provide supervision.
Adherence to a systematic operations research approach has enabled this
project to move successfully from problem identification,
through the
solutions, to field
development and systematic assessment of alternative solutions,
implementation.
The selection of a solution combining multiple financing
schemes has allowed participants to contribute to PHC according to their means
and has provided a broader base for project success.
it
it
it
This study was conducted from January 1984 through October 1985 by the
(CHAD.
Christian Health Association of Liberia (CHAL).
Further information is
available from the principal investigator. Dr. Andrew Cole, Christian Health
Association of Liberia, P.O. Box 1046, Monrovia, Liberia, or from Dr. Jeanne
Newman, PRICOR study monitor (Chevy Chase).
128
PRI®g>R
CF
Mali
Primary Health Care Operations Research
Study Abstract
COMMUNITY FINANCING OF PRIMARY HEALTH CARE
IN PERIPHERAL AREAS
The Republic of Mali’s formal health infrastructure of hospitals and health
centers serves only a small part of the country’s inhabitants. Eighty percent
of the people live in rural areas and receive only basic primary health care
(PHC) services provided by small dispensaries, village health workers, and
village pharmacies.
The problem of financing PHC delivery in these
underserved areas is acute and stems largely from the inability of communities
to sustain PHC services once outside funds have been depleted. The Director
General for Planning and for Health and Social Services Education therefore
undertook an operations research study to find ways for communities to help
meet the recurrent costs of maintaining PHC, particularly at the village level.
With the participation of members of the Comite d'Orientation et de
Coordination des Etudes et Programmes Socio-sanitaires (COCEPS) of the
Ministry of Health (MOH), the research team first identified the PHC services
to be provided at the village and ’’secteur de base” (a group of 5-10 villages)
levels. These were, in order of priority:
Essential drugs;
Maternal and child care;
Treatment of acute illnesses and wounds;
Health education;
*
Immunization against the six EPI diseases; and
Drinking water and basic sanitation.
This group then determined the best strategies for providing the services.
These strategies included establishing an initial stock of drugs; training
various types of health workers for the villages and "secteurs de base ” and
establishing a system for their supervision; holding quarterly health
education sessions in the villages; establishing a logistics and supply system
to provide regular immunizations in the villages; and refurbishing existing
water sources and creating and ensuring maintenance of new ones.
The next step was to determine the types of recurrent costs involved in
providing the targeted services. These costs included items such as drugs,
vaccines, immunization equipment and supplies, salaries, transportation, food,
and lodging for health workers and supervisors.
5530 Wisconsin Ave. • Chevy Chase, MD 20815 USA • (301)654-2550 • Cable: URCINTER • Telex: 64693
PR1COR is operated bv the Center for Human Services for the .Agency for International Development under Cooperative Agreement No. AID/DSPE-5920-A-00-1048-00.
129
To collect data for developing a solution to the financing problem, the study
team used two questionnaires.
One was used to interview local leaders of
60 villages in the areas of Koro and Kita. The other questionnaire was used
to interview 1,800 family and household heads in the same villages. Answers
to the questionnaires provided information on topics such as health worker
remuneration, drug supplies at the village level, family expenditures for
health services, and villagers’ attitudes toward those services.
In light of the replies to the questionnaires, the study team proposed that
PHC services in rural areas be financed as follows:
Restocking of village pharmacies: payment of a five percent markup
over the purchase price or a fixed fee for services.
Payment of village health workers: in kind at the discretion of the
recipients.
to be
Living expenses of village health workers during retraining:
paid by the village council.
Remuneration of "secteur de base" health workers: fixed salary to be
paid from a budget supported by fee-for-service payments or through
prepayments.
"secteur de base" health workers during
Living expenses of
retraining: to be paid by the "arrondissement" (level above "secteur
de base") council.
Lodging for head medical center nurse on supervisory visits: to be
paid by the "arrondissement" council.
Vaccines, lodg-ing, and transportation of vaccinator: to be paid from
a budgetary program * financed by the vaccination recipients, the
recipient villages, and from "arrondissement" resources.
Maintenance of water source: to be paid by the village council.
The first specific evidence of implementation of these results has been the
agreement of the Development Council of Koro to finance the 1986 budget
presented by the Chief Medical Officer of the Koro Health Center, This budget
In
includes many of the financing strategies developed in this study.
general, it is hoped that as a result of the study’s recommendations, villages
will finance certain costs associated with the delivery of curative services,
with "arrondissements" and "cercles" assuming financial responsibility for
preventive and supervisory activities.
*
*
*
This study was conducted from June 1984 to March 1986 by the Direction
Nationale de la Planification et de la Formation Sanitaire et Sociale of the
Ministry of Health of Mali.
Further information is available from the
principal investigator. Dr. Mamadou Traore, Directeur, Direction Nationale de
la Planification. Ministere de la Sante' Publigue et des Affaires Sociales,
Bamako, Mali, or from Dr. David Nicholas, PRICOR study monitor (Chevy Chase).
130
PRI®g)R
ORT
Mexico
Primary Health Care Operations Research
Study Abstract
A MARKET RESEARCH STUDY OF THE QUANTITATIVE AND QUALITATIVE
ASPECTS OF THE MARKETING AND DISTRIBUTION OF
ORAL REHYDRATION SALTS IN MEXICO
A market research study of oral rehydration salts (ORS)
(UKS) was carried out in
Mexico by Promotora de Planificacion Familiar (PROFAM), a Mexican nonprofit
organization.
Mexican health agencies have widely
i ’ ' \ encouraged the use of
packets of ORS granules to combat high infant morbidity and mortality from
diarrheal diseases.
PROFAM was interested in producing, marketing, and
distributing an inexpensive, easy-to-use ORS tablet and wanted to know if the
public would find this product acceptable.
Since drugstores are a popular
source of ORS products, it was decided that a survey of pharmacists would
provide a relatively good measure of supply and demand for ORS.
The survey included 116 private drugstores:
55 in Mexico City, 30 in the hot
region, and 31 in the temperate region. The researchers asked pharmacists and
other drugstore personnel to answer questions regarding the distribution and
sales of ORS, their knowledge of the purpose and correct use of ORS, the
demand for ORS, and their opinions on the presentation of ORS products.
Distribution and Sales.
Analysis of the survey results showed that the
majority of the drugstores surveyed (99 percent) sell ORS products, more than
too high for
half without a prescription. Although the prices are probably
j.
many consumers who need ORS products (US$ .73 - .94 per 500 ml bottle), the
pre-mixed liquid sells well, The pharmacists report that most sales are made
to women.
Knowledge of Purpose and Use. Most of the pharmacists stated correctly that
ORS products are used to treat dehydration caused by diarrheal diseases, Many
of the pharmacists, however, were not sure of
cl the
J— directions
2--- ---- for use. Only
one of
<_ the ORS supplier companies occasionally provided the pharmacists with
The
The
information on the purpose and the correct use of their ORS product.
pharmacists reported that- pre—mixed liquid ORS products had the highest
demand.
.
The demand for ORS products was seasonal, with the
Demand and Presentation.
largest quantities being bought in the spring andL summer.
A few pharmacists
felt that demand could be
be-• increased by adding flavoring to existing ORS
products. Opinions were mixed about the acceptability of an ORS tablet.
On
the one hand, many of the pharmacists thought that tablets
would be easy for
t---the consumer to use and more economical than other ORS products. On the other
hand, many pharmacists feared that use of contaminated water to dissolve the
5530 Wisconsin Ave. • Chevy Chase. MD 20815 USA • (301)654-2550 • Cable: URCINTER • Telex: 64693
PR1COR is opened by the Center for Human Services for the .Agency tor international Development under Cooperative Agreement No. AID/DSPE-5920-A-00-1048-00.
131
tablets would be a problem and that the tablets might be difficult to
dissolve. When asked to indicate whether ORS tablets or granule packets would
be more suitable, 33 percent of the pharmacists chose tablets, 48 percent
chose granule packets, and 19 percent had no opinion.
The investigators
emphasized that these results are the opinions of pharmacy personnel and do
not necessarily represent those of consumers.
Based on the study results and subsequent discussions, PROFAM decided not to
proceed with the production of ORS tablets. The two major reasons for this
decision were suitability and cost. The survey showed no particular consumer
advantage of the ORS tablets over granule packets, and the tablets also cost
slightly more than granule packets to produce. Instead, PROFAM will produce
packets of granules that can be dissolved in an 8-ounce glass of water. The
8-ounce glass is a more convenient container for consumers than the 1-liter
vessel required to dissolve existing granule packets.
PROFAM was to begin
manufacturing the packets as soon as they received Government approval.
it
*
It
This study was conducted during July 1983 by Promotora de Planificacion
Familiar (PROFAM), a private, nonprofit Mexican association.
Further
information is available from Ing. Luis de la Macorra, President, PROFAM,
Apartado Postal 34, El Pueblito, 76900 Villa Corregidora, Queretaro, Mexico,
or from Dr. David Nicholas, PRICOR study monitor (Chevy Chase).
132
CF
Philippines
^RI€®R
Primary Health Care Operations Research
RStudy Abstract
ALTERNATIVE STRATEGIES FOR FINANCING PRIMARY HEALTH CARE
IN THE PHILIPPINES
With PRICOR assistance, a team from the University of the Philippines in the
Visayas (UP-V) conducted a 2-year operations research study in Iloilo Province
aimed at developing feasible approaches to mobilizing community resources to
help pay for primary health care (PHC) services.
The study was part of a
larger project (Panay Unified Services for Health-PUSH) carried out by the
National Economic Development Authority in the region to improve PHC services
on the island of Panay.
The UP-V/PRICOR study had three major objectives:
(1) to help the people in
each of the study villages (barangays) determine what services they would
support; (2) to help them find appropriate means for raising funds and then
help them develop and implement effective financial management schemes; and
(3) to encourage them to use part of their resources to pay for preventive and
promotive services.
Six barangays, selected to represent different economic sectors of the island,
participated in the study.
In each barangay a baseline survey was carried out
to determine what people believed to be the most important health problems m
their community. The survey also identified people's perceived needs in terms
of health services; their attitudes toward, and utilization of, available
local health services, particularly the Barangay Health Workers; their current
expenditures for health services in the public and private sectors; and their
stated willingness to pay for additional services not available locally
through some sort of community financing mechanism. The major health problems
Annual
were perceived to be respiratory and gastro-intestinal illnesses,
($29)
to
health expenditures per household were estimated to range from P200
>
were
The unavailability of drugs and poor water supply facilities
P300 ($43).
as
major health-related problems.
The majority of households
1---perceived
willingness
to
participate
in
community
financing
of
health
expressed
In
the
six
barangays,
the
purchase
of
drugs
and
the
operation
of
activities.
drug
depot
(botika)
were
projects
that
community
members
were
most
willing
a
to finance.
Solution development consisted of several steps that involved the barangay
First, the study team charted the
residents in community financing projects,
pictorially
and
presented them to the barangay
results of the baseline survey
.
After
learning
the results of the baseline
residents at community assemblies,
communities
selected
health activities they
survey and discussing them, the (.
financing project and the type of financing
would fund through a community
c
mechanism they would use. Five barangays chose to finance and run community
5530 Wisconsin Ave. • Chevy Chase, MD 20815 USA • (301)654-2550 • Cable: URCINTER • Telex: 64693
PR1COR is operated bv the Center for Human Services for the Agency for International Development under Cooperative Agreement No. AID/DSPE-5920-A-00-1048-00.
133
drugstores, botikas sa barangay, and one barangay decided on an emergency
hospitalization loan fund.
These would be revolving funds in which user
payments for drugs (or repayment of loans) are used to replenish stocks (or
replenish the loan fund).
Most barangays selected a flat rate contribution
from households for the
initial capitalization of the
project,
but
supplementary fundraising activities, such as taxes on sales of produce and
livestock, raffles, and parties, were also included in some of the plans. One
month after the community financing schemes were initiated, the researchers
held a workshop in each barangay for members of a core group of villagers who
had agreed to take responsibility for the project. These workshops were used
to plan the community financing project in more detail, to strengthen the
management capabilities of the core group, and to teach concepts of primary
health care. During this workshop, community women were selected for the jobs
of "lead mother."
These unpaid volunteers were to assist the barangay
residents in implementing preventive and promotive health activities.
Most of the barangays were able to collect an average of 46.1 percent of the
targets they had set for fundraising through flat rate contributions.
The
botikas were managed and run by volunteers from the communities.
One person
was generally in charge of dispensing drugs, collecting money, and maintaining
stocks. Each botika made its own arrangements to buy its initial stock and
replenishments.
Pricing policy was set by each barangay.
Stock turned over
at annualized rates ranging from 132 to 913 percent and profit on sales ranged
from 16 to 22 percent, despite claims of very small markups.
Mine months
after the research study had ended, the botikas and the emergency fund were
found to be still functioning.
Between 83 and 92
percent of barangay
residents had contributed to the capitalization of the botikas (99 percent to
Utilization of the boticas ranged between 54 and 77 percent
the loan fund).
The results of the lead mothers program were mixed.
Those
of households,
barangays where lead mothers were active and effective did make clear,
however, that potential beneficiaries of health services (lead mothers) can
take on roles as preventive and promotive health care motivators if they are
given adequate support.
From the success of the botikas sa barangay, it can be concluded that these
Filipino villagers valued having a local source of desired drugs sufficiently
to pay for this service.
They were not, however, willing to pay for
preventive and promotive health services, as had been initially hoped by the
PRICOR researchers.
With considerable outside assistance, the barangay
residents did learn how to capitalize, organize, and manage revolving funds.
Extensive community participation in designing, financing, and managing the
botikas seemed to be a key factor in their performance and sustainability.
*
*
*
This study was conducted from January 1983 to May 1985 by the University of
the Philippines in the Visayas Foundation.
Further information is available
from the co-principal investigators. Dr. Trinidad S. Osteria, Institute of
Southeast Asian Studies, Heng Mui Keng Terrace, Pasir Panjang, Singapore, and
Professor Ida M. Siason, University of the Philippines in the Visayas, Iloilo
City, Philippines, or from Dr. Stewart Blumenfeld, PRICOR study monitor (Chevy
Chase).
134
PR|(S®R
CF
Senegal
Primary Health Care Operations Research
Study Abstract
COMMUNITY FINANCING OF PRIMARY HEALTH CARE IN
RURAL AREAS OF SENEGAL'S SINE SALOUM REGION
One of the objectives of the second phase of the Sine Saloum Rural Health
Project (SS-RHP) in Senegal is to decrease the dependence of the project on
USAID funding. Given rigid budget constraints, the Senegal Ministry of Health
(MOH) needs to find alternative means of financing the basic health services
provided by the project.
An operations research study carried out by the
Harvard Institute of International Development in 1983 addressed this
problem.
The objective of the study was to determine whether community
financing could adequately support rural primary health care (PHC) in Sine
Saloum.
The Sine Saloum Rural Health Project has been in operation since 1977.
The
project provides PHC services to 880,000 people in five of six departments of
the region through a network of self-sustaining "health huts" (village-based
health care facilities) in about 600 villages.
Villagers are expected to
construct the huts, encourage their fellow community members to use the new
services, and compensate the village health workers (VHWs) trained by the
Funds to cover costs are to come from the fees collected for
project,
Ideally, government supervisors visit each
medicine and other services.
village once a month to supervise VHWs and to encourage village health
Phase II of this project
committees in their support of the VHW and of PHC.
was being designed at the time of the PRICOR study.
In order to examine the potential for community financing of PHC, the
principal investigator proposed to analyze the economic feasibility of 13
government-initiated PHC activities.
These activities were broken down into
those that were involved in establishing a local PHC program and those that
were involved in operating and maintaining such a program.
Establishment
phase
activities
included:
sensitization
of
villagers,
agreement of village leaders to participate in the PHC program, nomination of
community members to serve as volunteer health workers (VHWs), preservice
training of VHWs, construction of a health hut, and provision of an initial
stock of drugs and other medical supplies to the VHWs.
The operation and
maintenance phase included: patronage of VHWs by the villagers, availability
of VHWs to serve the villagers, resupply of drugs and related products,
physical maintenance of the health hut, management of project income by
community
representatives,
supervision by government health officials,
recording and reporting of service data, and inservice training of VHWs.
5530 Wisconsin Ave. • Chevy Chase, MD 20815 USA • (301)654-2550 • Cable: URCINTER • Telex: 64693
PR1COR is operated bv the Center for Human Services for the .Agency for international Development under Cooperative Agreement No. AID/DSPE-5920-A-00-1048-00.
135
Through a review of existing documentation on PHC interventions in Senegal and
interviews of health care providers and decisionmakers at all levels of the
health system, the principal investigator described activities relating to
He also attempted to
supervision, VHW1 renumeration, and drug distribution.
determine whether each activity is essential to the maintenance of rural PHC
and, if essential, how it can be financed.
He then made the following
conclusions regarding the community financing potential of each of these
recurrent cost components.
The major supervision costs to the Sine Saloum Project were those associated
with supervisors’ transportation to the villages (e.g. motorcycle fuel,
maintenance, and repair).
None of the possible financing sources examined
seemed likely to pick up a substantial share of these costs in the near
future. The investigator therefore proposed that a less intensive supervision
system be established.
In the area of VHW remuneration, an in-depth sociological observation of the
existing situation was proposed. The investigator thought that this would be
necessary to assess the present level of VHW compensation in Sine Saloum, as
well as its variability, its stability, and its adequacy to sustain VHW-based
PHC.
The investigator also concluded that the existing network of private commerce
in Sine Saloum could handle the distribution of pharmaceuticals at a fraction
of the cost incurred by public sector agencies, and with greater reliability
and frequency of resupply.
Using these conclusions, the PRICOR investigator developed recommendations for
the design of the second phase of the Sine Saloum Rural Health Project.
Concerning the extension of the project to the remaining departments of the
region, he recommended that a preliminary sociological and anthropological
investigation be a prerequisite to the implementation of the program
(construction of the health hut) in any village.
The purpose of this
investigation would be to determine the villagers' perceptions of their health
problems; the procedures the villagers are currently following to obtain
health care; how much time, effort, and money it cost villagers to obtain
health goods and services; and what improvements in health care they expect
from the new PHC program.
The investigator also recommended developing
studies to assess the effects of phasing out onsite supervision and to
determine
how modified arrangements
for drug procurement affect the
reliability of supply in those areas where USAID support for the PHC program
is gradually being withdrawn.
it
it
it
This study was conducted from July 1983 through December 1985 by the Harvard
Further
information is
Institute of International Development (HIID).
1737
available from the principal investigator.
investigator, Dr. Clive Gray, HIID,
Cambridge Street, Cambridge, Massachusetts 02138, or from Dr. David Nicholas,
PRICOR study monitor (Chevy Chase).
136
W®R
CCD
Somalia
Primary Health Gare Operations Research
■Study Abstract
OPERATIONAL PROCEDURES TO
IMPROVE AVAILABILITY OF PHC DRUGS
An operations research study focusing on the use of modern drugs in rural
areas of Somalia was carried out by researchers from the Somali National
Academy of Arts and Sciences and Medical Service Consultants/ Inc. (USA). The
study, supported by PRICOR and UNICEF/ with the cooperation of the Somali
Ministry of Health (MOH) and USAID/Somalia, was conducted in villages in six
geographically diverse regions of Somalia.
The research team first discussed the problems of drug delivery in the rural
areas of Somalia with officials from the MOH and several international health
These decisionmakers believed that the shortage of drugs in rural
agencies.
areas was a major constraint to implementing the Primary Health Care (PHC)
Factors that might contribute to breakdowns in drug
Program in Somalia.
delivery were identified.
Three country-specific survey documents were designed and pretested to gather
facts about (1) patient use of modern drugs at the village level; (2)
prescriber practices in dispensing these drugs; and (3) drug stocks actually
present at the prescribing facilities.
Thirty rural villages were selected for a household survey that used a
In each village/ 24 households were
multistage cluster sample design.
The total number of households interviewed
randomly chosen for the survey,
Thirty-two percent of households had purchased drugs in the past
was 716.
6 months, The 716 households interviewed reported average drug expenditures
A significant
of 68.64 Somali Shillings (US$ .83) in the past 6 months.
portion of Somali villagers seem to be willing and able to make out-of-pocket
payments for drugs, based on the fact that they are already doing so.
Some drug users (approximately 7.0 percent of the total sample or 16.7 percent
of apparent drug users) reported purchasing drugs from a PHC or MOH clinic
(where drugs are supposed to be free). Apparently/ some goverment health cafre
facilities were charging for drugs in order to help meet costs.
Forty-nine
percent of respondents indicated that PHC or other MOH facilities were their
first choice when seeking modern medicines.
The demand for modern drugs in
the rural areas is, however/ much greater than the government PHC facilitie’s
have been able to meet.
Of 31 different diseases named by dispensers surveyed at clinics, health
posts, and drug stores used by villagers, the three most often cited were
seems
to correspond
This
malaria, diarrhea/dysentery, and bronchitis.
5530 Wisconsin Ave. • Chevy Chase. MD 20815 USA • (301]654-2550 • Cable: URCINTER • Telex: 64693
PR1COR is operated bv the Center for Human Services for the .Agency for international Development under Cooperative Agreement No. AID/DSPE-5920-A-00-1048-00.
137
approximately with the most common illnesses reported by the households:
cough, diarrhea, fever, headache, influenza, malaria, and stomach distress.
The most frequently mentioned drugs for the "most common diseases" were
penicillin, chloroquine, and aspirin. Drugs reported by households to have
been received from PHC or other MOH facilities included chloroquine, aspirin,
ORS, ferrous sulfate, cough medicine, ampicillin, and cilfotrim. Of the 47
items recommended by the dispensers as treatments for the "most common
diseases," only 12 items would be sufficient to treat the spectrum of
illnesses.
The survey data were then organized to focus on the following areas of
interest: (1) household profiles; (2) household reports of drug use, cost, and
source; (3) practitioner or dispenser reports of most common diseases and the
drugs recommended for them; and (4) selected PHC drug items existing at
facilities furnishing pharmaceutical services to the target villages.
A summary of the findings from the three surveys was circulated to seven
decisionmakers in the MOH, together with a list of strategies to improve the
delivery of drugs to the rural areas of Somalia. The MOH decisionmakers were
asked to comment on the findings of the study and to rank the alternative
strategies in terms of effectiveness and feasibility. Six of the participants
cited the following three options as the most feasible:
improve the drug
distribution system, improve inventory control at the central drug warehouse,
and improve dispenser awareness of appropriate drug use. The same six options
were ranked from most to least important for improving the availability of
drugs to
the
rural
population.
Seeking
the
local
production of
pharmaceuticaIs was ranked as most important. Limiting the number of drugs to
be handled and improving the distribution system tied for second place.
After examining the findings of the PRICOR study and the strategies suggested
by the Somali MOH decisionmakers, the study team developed recommendations for
improving drug delivery to the rural areas. The researchers felt that both
the public and the private sectors should be used for PHC drug delivery. The
study team also recommended the preparation of a written "Guide to PHC Drug
Use" that could be used to teach the Somali villagers. A functioning Somali
women's organization was identified as a group to work with in developing the
education program. It was also recommended that the MOH review the PHC drug
list with the goal of limiting the number of items included.
*
*
*
This study was conducted from April 1985 to December 1985 by the Somali
National Academy of Arts and Sciences, and Medical Services Consultants (MSC),
Inc. Further information is available from MSC, 1716 Wilson Blvd., Arlington,
Virginia 22209, or from Dr. Stewart Blumenfeld, PRICOR study monitor (Chevy
Chase).
138
PRI€®R
CF/CHW
Swaziland
Primary Health Gare Operations Research
Study Abstract
DEVELOPMENT OF APPROPRIATE METHODS
FOR SUSTAINING RURAL HEALTH MOTIVATORS
The Rural Health Motivator (RHM) is a key primary health care (PHC) element m
rural Swaziland. The program, however, suffers from high turnover rates.
In
conjunction with the Primary Health Unit (PHU) of the Ministry of Health (MOH)
from the
and with the support of the PRICOR Project, a team
i--- ---- Social Science
Research Unit (SSRU) of the University of Swaziland undertook an operations
research (OR) study to find ways to stabilize the RHMs.
The solution to the problem was identified as having three main components:
community participation, stronger supervision, and dependable compensation.
RHMs receive a monthly stipend from the Ministry,, but it is small and delays
In setting priorities to resolve the problem,
in payment are not uncommon,
the
basis of previous work with groups of RHMs,
the MOH/SSRU team decided, on
that small, irregular compensation was probably the major cause of RHM
turnover.
Supplementation by the community was thought to be an appropriate solution
because it offered the benefit not only of increasing the RHM’s payment for
services, but also promoting increased community involvement in assuring
availability of PHC.
However, the researchers determined that before
communities would be willing to pay for RHM services, the skills of the RHM
would have to be strengthened.
The objectives of the research, therefore,
became to define a set of skills within the capabilities of the RHM that were
acceptable to the MOH and that the community would consider worthy of
compensation, and to develop an appropriate compensation scheme.
Several OR techniques were used in the problem analysis phase of the study to
examine the organization of the country's PHC system and to identify
constraints, inputs, processes, outputs, and outcomes.
A nationwide survey,
which elicited information crucial to solution development, focused on four
areas: utilization patterns of health care, health expenditure, willingness
of communities to support the RHMs, and attitudes of the RHMs themselves
toward their work.
An interesting finding of the survey was that while 28
percent of the respondents said they would be willing to support the RHM
financially, not one of the RHMs interviewed believed the community would be
willing to do so, mainly because the RHMs were, in principle, already
receiving a. salary from the government.
Based on comments from the RHMs,
experience gained from other projects, discussions with health professionals,
and the results of the community survey, the research team developed matrices
The
to determine what RHM activities should be strengthened or introduced.
matrix results, in combination with discussions with MOH officials, determined
that immunization. oral rehydration therapy (ORT), and growth monitoring
5530 Wisconsin Ave. • Chevy Chase. MD 20815 USA • (301)654-2550 • Cable: URCINTER • Telex: 64693
PR1COR is operated bv the Center for Human Services for the .Agency for International Development under Cooperative Agreement No. AID/DSPE-5920-A-00-1048-00.
139
should be emphasized, Another matrix helped identify prepayment for services
as the most appropriate form of compensation.
Solution testing took the form of a demonstration field test in a single
chieftaincy over a period of 6.5 months, The research team initiated the test
by conducting community meetings and a 2-day training course for the eight
RHMs.
The test was carefully monitored through meetings with RHMs and
communi ty 1eade rs. A community survey and interviews with RHMs and community
leaders were used to evaluate the field test.
The field test evaluation revealed that the PHC skills training for the RHM
had served to broaden the community’s perception of the role of the RHM in
relation to children’s health. For example, respondents perceived the RHM to
be the primary source of information on ORT, and mothers complied well when
the RHM referred their children for immunization. The newly-introduced growth
monitoring skills were well received by both the community and the RHMs, and
RHMs proved to be proficient and active in performing these skills.
The RHM support scheme the community chose involved the donation of communal
land and agricultural labor toward the production of a crop to be given to the
RHM, who could then sell the crop for cash.
This plan did not quite reach
fruition because the area chief, who alone has the power to direct the people
to perform civic duties, was absent at the crucial field preparation period
due to the coronation of a new king.
The field set aside for the RHM was
The
prepared by the community, but the RHM deemed it too late for planting,
chief and the community agreed that the RHM should remain active and that the
An
community would prepare the field
again for the next planting season,
1
adjacent chieftaincy, without any promotion from the research team, has
indicated that it would also try the same scheme.
The following three recommendations were made to the MOH as a result of the
study.
First, RHMs should be trained to perform growth monitoring at the
community level.
This training should include local-language curriculum
materials, careful referral guidelines, proper tools such as scales and growth
cards, and adequate supervision and inservice sessions. Second, RHM inservice
training should be revised to emphasize immunizations and ORT.
Third, the
Public Health Unit (PHU) should try to improve community support for RHMs.
These efforts should include encouraging communities to devise their own,
locally-appropriate forms of RHM compensation, considering ways to make RHM
spouses more aware and supportive of RHM activities, and increasing efforts to
keep the traditional leaders informed of and involved in PHU and RHM
activities in their communities.
The study team emphasizes what is probably a key finding of the study, i.e.
that only after community support for RHMs is improved should further efforts
to establish community-based in-kind or in-cash contribution schemes be
pursued.
Hr
*
*
This study was conducted by researchers from the Social Science Research Unit
of the University of Swaziland from April 1984 through March 1986.
Further
information is available from the principal investigators, Ms. Laurie H. Dunn,
P.O. Box 4, Malkerns, Swaziland, or Ms. B. Dlamini Vilakati, Ministry of
Health, Mbabane, Swaziland, or from Dr. Stewart Blumenfeld, PRICOR study
monitor (Chevy Chase).
140
PRie®R
CF
Thailand
Primary Health Care Operations Research
Study Abstract
COMMUNITY FINANCING OF PHC ACTIVITIES
IN NUTRITION, WATER, AND SANITATION
Researchers from the National Economic and Social Development Board (NESDB)
conducted a study to identify and test cost-effective models of community financing
for primary health care (PHC) activities in nutrition, water, and sanitation.
The Thai Government recognized that nutrition, water, and sanitation activities
were crucial to an effective PHC program, but the Ministry of Public Health (MOPH)
did not have the funds to pay for these health interventions, nor would the
interventions be sucessful without community participation and support. Thus, the
purpose of the study was to identify a model or models of community financing of
PHC activities in nutrition, water, and sanitation that would best mobilize
community resources in support of these activities.
The solution development phase of the study employed three data collection
First, the MOPH sent a letter to all 5,000 tambon health officers
activities.
(tambons are subdistricts consisting of 6 to 10 villages) asking them to identify
PHC funds in their jurisdictions. Over 70 percent of the health officers responded
and identified over 12,000 funds.
Next, more detailed questionnaires were mailed
to the health officers asking for specific information about the 12,000 funds:
their age and origin, management characteristics, procedures, diversification, and
problems; as well as services they provide and households they serve.
Completed
questionnaires were received providing information on 4,631 funds.
Finally, 63
in-depth case studies were carried out in 22 provinces around the country.
The data were analyzed to: (1) distinguish and describe existing viable models of
community financing; (2) <explain variations in viability and performance; and (3)
The analysis showed
propose alternative modelsi for testing and implementation.
that there are five types of funds:
1.
2.
3.
4.
5.
Single-purpose funds (drugs, nutrition, water, and sanitation)
Single-purpose subsidized funds
Comprehensive PHC funds (support all PHC services)
Multipurpose funds (not limited to PHC)
Health card funds.
The data also showed that funds vary systematically with the primary health care
activity they are set up to finance and the population they serve.
For example,
drug funds are the oldest, most numerous, most consistently profitable PHC funds in
Thailand. They serve more households, have more diversified income sources, show
more potential for diversification of services and less regional variation than
nutrition or sanitation funds.
5530 Wisconsin Ave. • Chevy Chase, MD 20815 USA • (301)654-2550 • Cable: URCINTER • Telex: 64693
PR1COR is operated by the Center for Human Services for the .Agency for International Development under Cooperative Agreement Na AID/DSPE-5920-A-00-1048-00.
141
Nutrition funds are the next most widespread, located primarily in the northeast
and north.
They tend to be found in smaller, poorer villages, have little
financial base for profitability or capital appreciation, be heavily dependent on
labor contributions of women, and/or are decapitalizing.
Ironically, the most
profitable of these are often the least active in reducing malnutrition and the
least successful by other PHC criteria.
Finally, sanitation funds are the least
numerous, with half located in the northeast.
There are regional variations in
working capital, profitability, interest rates, share purchases, and services.
They tend to be located in larger and more prosperous villages.
The most
successful concentrate on providing loans at market interest rates for construction
of water-sealed privies.
The research team examined various financing models and compared them against a set
of standard criteria, including viability, profitability, services, coverage, and
ability to support basic PHC services.
They concluded that the best solution to
the operational problem is multipurpose funds, for several reasons:
(1)
income
sources are multiple and diverse, risk is spread, and income is likely to be more
dependable, profits higher, and capital growth more rapid; (2) purchase of shares
by households is encouraged by the real prospect of profit; (3) multipurpose funds
conserve on scarce management time and skills—one multipurpose fund requires fewer
people and less time to manage than 5-10 single-purpose funds; and (4)
higher
profits and rapid capital growth enable a multipurpose fund to support nutrition
activity, even if it does not make money, and to make loans for sanitation
improvements (particularly to low income people).
These results and recommendations were presented to and accepted by the MOPH. The
Secretary General of the NESDB presented a summary of the findings and
recommendatons to the Thai Cabinet at the request of the Prime Minister. The MOPH
has accepted the recommendation that existing PHC funds, particularly drug funds,
should be encouraged to diversify and that new multipurpose funds should be
established when conditions permit. Starting in 1985, the MOPH began carrying out
these recommendations. In the solution validation phase of the study the research
team conducted longitudinal testing of the various models of establishing
multipurpose funds.
An interesting finding of the study was that the success of community financing was
not the ability to put up a certain form or models of financing schemes, but the
ability to transmit the concept of community financing to the rural community, who
will in turn formulate their own finance scheme that is responsive to their
particular needs and settings,
The essential elements of such financing schemes
are:
(1) the pooling of capital and non-capital resources within the community
that are rotatable and self-generating; (2) the ability to pool resources from
within and without the community; (3) the community has the ultimate decision
making power over the administration of the pooled resources; (4) the community has
the ultimate decision-making power over the utilization and determination of
activities of/for the pooled resources; and (5) the community has a central body or
network of central organization to overload the pooled resources.
*
*
*
This study was conducted from March 1983 through January 1986 by the National
Economic and Social Development Board (NESDB) of Thailand. Further information is
available from the principal investigator, Ms. Orathip Tanskul, NESDB, Krung Kasem
Rd., Bangkok 10100, Thailand, or from Dr. Jack Reynolds, PRICOR study monitor
(Chevy Chase).
142
PRI<$®R
CF
Zaire
Primary Health Gare Operations Research
Study Abstract
TESTING ALTERNATIVE PAYMENT SCHEMES IN HEALTH
CENTERS IN ZAIRE
In 1982, the Zaire Government decentralized public health activities and created
health zones which were to be largely self-financed.
The Basic Rural Health
Services Project (SANRU), a collaborative effort of the Government of Zaire, USAID
and the Church of Christ in Zaire, was created in 1982 to coordinate this
reorganization. One of the tasks of the SANRU project was to identify sustainable
PHC community financing schemes.
The study team identified four PHC payment schemes being used in rural areas,
namely, (1J fixed fee for episode, regardless of illness; (2) fixed fee for visit,
with varying medication fees related to daily dosage and drug cost; (3) fee for
episode, varying with severity of illness and cost of drugs; and (4) fixed fee for
visit and drugs, but reduced fees for necessary repeat visits.
The first two
schemes were selected for study because the third and fourth were viewed as
variations of the second.
First, a baseline household
Two principal data collection techniques were used,
survey was carried out at each study site before and after a new scheme had been
introduced.
Second, a cost analysis of health services was carried out in five
health zones supported by the SANRU Project.
In each of five rural health zones, two health centers were chosen to participate
in the study.
In four of the zones, one center was asked to change to a scheme
based on fee per episode of illness. The other center continued to use a payment
scheme based on a fixed consultation fee and variable drug fee. In the fifth zone,
a health center with a fee per episode scheme changed to - fee per consultation while
the other continued with the fee per episode. These schemes were observed for one
year.
Curative visits per capita varied enormously from one health center to another,
from a low of 3 visits per year per 100 population in Katanda and Tshileo to a high
of 245 per 100 persons in Lukunga. One reason for this was that the proportion of
persons seeking health care who went to the SANRU health center ranged from 1.4
percent in Katanda (where the nurse set up a competing private practice) to 83.2
percent in Lukunga (a small, well-defined, and isolated service area). A second
explanation was that people in some areas reported very infrequent use of any
health care provider (only 10 visits per 100 persons per year in Tshileo, for
Visits
example), while in other areas visits were up to 35 times more frequent.
per episode of illness varied from scarcely more than 1.0 in Lukunga, Muadi Kayembe
and Kaniama to over 6.0 in Kangoy.
Curative care costs per capita, per visit, and per episode also varied enormously.
(Costs are reported in US dollars, and are based on an exchange rate of US$ 1 =
Z 43.8 from October 1984 - March 1985 and US$ 1 = Z 52 from April 1985-September
1985.).
In Tshileo (the area with low utilization), direct operating costs for
twelve months were only 2 cents per capita, while in Lukunga they were 90 cents.
Costs per visit ranged from 4 cents in Kangoy to $3.46 in Katanda.
The latter
5530 Wisconsin Ave. • Chevy Chase, MD 20815 USA • (301)654-2550 • Cable: URCINTER • Telex: 64693
PR1COR is operated by the Center for Human Services for the Agency for International Development under Cooperative Agreement No. AID/DSPE-5920-A-00-1048-00.
143
*
center experienced extremely low utilization so that fixed costs were spread over a
relatively small number of visits, while the former had extremely high
utilization.
Health center operating costs per episode ranged from 27 cents in Kangoy to $7.46
in Katanda, with a median of 95 cents. At the average center, a patient payment of
95 cents per episode would have covered personnel salaries, zonal supervisor and
mobile team salaries for the time they spent on field activities, all inservice
training, administrative and maintenance supplies, drugs, vaccines, kerosene, minor
building repairs, and transport for personnel and materials including supervisors
and the mobile team.
Medical supply costs (mainly for drugs) accounted for nearly half of all curative
care costs, ranging from 8.8 percent in Katanda (where fixed personnel costs were
high for an underutilized clinic) to a high of 70.1 percent in Kaniama.
Most
health centers spent 22 to 34 percent of their funds on preventive care.
The operating costs of four of the five zones were $3,500 to $5,500 per year
although the number of health centers that they supervised differed significantly
from 6 to 47.
The fees that clinics set varied from one health center to another and were set in
diverse ways. No technical formula for setting prices was agreed upon, yet the 6
centers with usable revenue data reported that they covered a median of 107.3
percent of their curative costs, including costs of zonal supervision and mobile
teams.
Preventive revenues covered between 0.4 percent and 10.6 percent of
preventive costs, but three health centers collected revenue from curative care in
excess of costs permitting cross-subsidization of preventive care.
Overall, the
care.
six clinics reported median cost recovery of 66.5 percent.
Investigators learned a great deal about the process of introducing or changing
community financing schemes and about the effect of this process on clinic
operations and community utilization.
Fee levels set through the '’unscientific”
process of community decisionmaking and crude cost calculation proved to be
adequate for significant cost recovery, especially for curative care.
Clinic
management, staff quality and morale, drug supply, and relations with the community
as a whole were probably more important influences on utilization in these ten
centers than was payment scheme, yet these factors were themselves affected by
staff and community reaction to change. Clinic staff in several locations disliked
either the payment scheme they were asked to implement or the fee levels that had
been set. Utilization declined severely in one center because the nurse quit and
established a competing private practice; his successor was subsequently dismissed
for malfeasance. The community which was asked to switch from fee per episode to
fee per visit objected to what they considered a less desirable system and began
making greater use of other providers.
During this time period in this country,
the way in which a payment scheme was established and managed appeared to affect
utilization, cost, and cost recovery, more than did the precise nature of the
scheme.
*
*
*
This study was conducted for SANRU through the African Intermennonite Mission from
September 1983 through March 1986.
Information is available from the principal
investigator. Dr. Lusamba Dikassa or from Dr. Frank Baer, P.O. Box 3555, Kinshasa,
Zaire or from Ms. Marty Pipp, PRICOR study monitor (Chevy Chase).
144
APPENDIX 2. CASE STUDIES OF COSTSHARING SCHEMES
THAILAND
In Thailand, an insurance scheme or "health card fund”
operated as a conceptually simple and inventive way of sharing
costs for MCH, EPI and other preventive services and of
rationalizing referral patterns for treatment of illness. In
18 surveyed villages, health cards were sold at a modest
price, affordable by most households. The card entitled
households to treatment of up to a fixed number of illness
episodes per year - usually eight - and to free MCH and EPI
services. There were notable variations in price, coverage,
benefits and the use of capital among the programs surveyed.
Money collected from the sale of cards was used to make loans
to card holders and, at the end of the year, to reimburse
service providers.
An analysis of household survey data confirmed that the
price of health cards - usually 100-200 Baht (or US$4-8) - was
affordable to most households, even those below the poverty
line or with heads in low-income occupations. Average monthly
expenditures for health care, when either the wife or husband
was ill, exceeded the annual price of the card. An estimated
one percent of the population, largely rural households in the
South, could not afford health cards. Free cards have been
made available for such households.
The health card was purchased by most rural households,
where available, but sales were sensitive to limitations on
coverage, the number of people eligible to use the card, and
competition from private sector providers. There were large
variations in ceilings and limitations on coverage. One fund
imposed no limitations; others imposed ceilings on hospital
coverage or the annual patient charges per household. One
fund excluded certain diseases from coverage. Some funds
covered everyone in the household; while others were limited
to specified family members, usually a husband, wife and two
children.
The point of first contact for treatment of illness was
the health volunteer or the subdistrict health center.
Referral to a district or provincial hospital required a
letter or slip from providers at a lower level, except in an
emergency. With the slip, the card holder was entitled to
quick attention in the hospitals honoring the card.
145
Residents purchased health cards because of:
their modest price, which reduced the cost of
medical care;
the availability of loans;
and, reduced waiting time for services.
Funds varied in their use of capital. MOPH guidelines
for the first year suggested that funds set aside 40 percent
for loans or other community activities that would increase
fund capital and 60 percent for provider reimbursement and
fund management. The suggested proportions were 20/80,
respectively, in the second year and 10/90 in the third. As
most funds did not reimburse providers until the end of the
year, a fund could loan most or all of its capital on a short
term basis in the interim. Each fund had the discretion to
set loan terms and later allocations among providers in
consultation with them.
Health card funds were designed to encourage the use of
preventive services, increase the use of Tambon health
centers, and raise capital to finance health services,
However, a significant threat to the viability of health card
funds came from inadequate reimbursement of providers and the
inability of supply to keep pace with increased demand. The
200 Baht card covered only a fraction of the cost of service
delivery. Analysis suggested that the price of the card was
too low in 1984. Demand was likely to put heavy pressure on
health centers and MOPH hospitals reducing cost recovery in
district and provincial hospitals. Models which included
higher priced cards and lower or graduated benefits were
recommended as viable alternatives. The MOPH subsequently
accepted these recommendations altering both price and
benefits.
BRAZIL (LASSNER)
Cost sharing selected by community organizations
included:
a.
contributions of labor and supplies for
cleaning and maintenance of health units;
b.
contributions of cash to pay health unit
utility bills;
c.
contributions of expendable office and clinic
supplies;
146
d.
recruitment and employment of community members
as PHC promoters;
e.
payment of CHWs.
Although community leaders were involved in planning,
these schemes were often not successfully implemented;
community contributions generated the third lowest total
revenue of the nine strategies tested. There.were limited
successes, however. In one community, the neighborhood
association was only able to pay the electric bill for the
first month of their agreement. Therefore, they successfully
introduced a collection box to finance payment of utility
bills and cleaning of the health unit. The contribution box,
suggested by community merchants and other residents, resulted
in a broader-based cost sharing scheme.
Researchers attributed the limited success of these
schemes to weak planning and limited community support. Many
community groups were poorly organized and had financial
difficulties and political problems. Representatives of
neighborhood associations often changed, and agreements made
by previous representatives were often not upheld. Some
leaders did not adequately represent their communities, In
addition, details of how organizations were to solicit
monetary and labor contributions were never worked out.
SWAZILAND, BOLIVIA AND LIBERIA
In Swaziland, Bolivia and Liberia (Cole I), communities
contributed labor to help CHWs in agricultural work. In
Swaziland, a plan was developed for villagers to donate labor
crop
to produce a crop on communal land. rThe
"’
. was to be given
to the CHW for her own use or sale as compensation, The CHW
•'
was to provide seed2 for the
crop, The land was prepared as
r
r
the
village
chief was away when it was time
planned; however, 1
to plant, and villagers would not plant the field without
direction from the chief. This situation vividly illustrates
how a single, key individual can be vitally important.
Although the crop was not produced the first year of the
project, the community agreed to prepare and plant the field
the following year.
Investigators in Cochabamba, Bolivia developed a similar
scheme to reduce CHW attrition. Through interviews,
investigators determined that villagers were willing to pay
CHWs in-kind provided that they were reliable and worked
exclusively in their own communities. However, the level.of
in-kind payment which the villagers were willing to contribute
(equivalent to US$ 7 a year) would cover less than half the
CHW’s current salary. Therefore, a lower-level health
promoter was trained and deployed at a reduced salary.
147
Villagers donated an annual quota of wheat or potatoes as
an in-kind payment. Contributions were collected, stored and
distributed by the sindicato (a traditional and highly visible
organization in rural communities) and by the health
committee.
In Liberia, villagers decided to contribute part of their
personal crops of coffee, cocoa and rice for PHC and to use
part of the proceeds of a communal rice farm. In addition,
villagers agreed to work on the CHW’s field. Many of these
plans could not be implemented due to insufficient community
support, even though healers were traditionally compensated in
this manner.
148
APPENDIX 3
CASE STUDIES OF PREVENTIVE/PROMOTIVE ACTIVITIES
THAILAND
Design. Researchers in Thailand conducted surveys of revolving funds and
detailed case studies of 63. They found that revolving fund loans,
ocassionally from revolving drug funds, financed water and sanitation
improvements. Single-purpose sanitation funds accounted for 12 percent (538)
of 4,631 revolving funds represented in the detailed survey. Loans for
water-sealed privies were the least expensive and most frequent activity.
Working capital, profitability, interest rates, share purchase prices and
services varied, but funds had many common characteristics. Most were less
than one year old, and half were in the Northwest.
Sanitation funds tended to be located in more prosperous villages than
other revolving funds as sanitation improvements are relatively expensive.
All but two of the case study funds were established by the MOPH or outside
donor agencies with initial contributions of construction molds, materials and
cash. In addition, the MOPH trained village craftsmen in construction of
privies, well heads, water jars and rain-water cisterns. Initial capital was
supplemented by sale of fund shares to local community members. The
percentage of households purchasing shares ranged from 2 to 40 percent.
Households too poor to borrow from the fund had little incentive to purchase a
share. Many community members who could not afford the shares or did not wish
to borrow from it fulfilled a social obligation to support the fund through
labor to construct sanitation improvements. Workers received only meals
during the construction, and the benefiting household was obligated to
contribute labor in future projects.
Half of the 17 sanitation funds which PRICOR studied in depth did not
charge interest or have profits. Interest rates for those funds which did
charge ranged from 8 to 18 percent compared to prevailing money lender rates
of 36 to 60 percent. Only those funds which charged interest close to market
rates were highly profitable and showed capital appreciation. In the national
survey, the percentage of funds charging interest or with other income sources
was nearly identical to the proportion showing a profit.
Management. Sanitation funds were managed by a local committee whose
members were 85 percent male and an average of 44 years old with a modest
income. Interest rates and share prices were set by the committee, and loans
were allocated either by lottery or by committee selection. CHWs were members
of 36 to 46 percent of the sanitation fund committees, varying by region.
Managers were elected in 75 percent of the case study sanitation funds
with the remainder appointed by MOPH officers. All managers were shareholders
and served unlimited terms with almost no turnover. They were responsible for
supervision of construction, and for record keeping, inventory control and
loan collection. Only 6 to 13 percent of the fund managers were compensated
in cash or in-kind, but many were given preference for low or no cost loans
for themselves, family or friends.
149
Issues. Equity must be considered in management of sanitation funds.
Community members did not have equal access to loan funds. First, only those
who could afford to purchase shares could borrow from the fund. Second, in
cases where committees decided who got the loans, preference was given to fund
managers, committee members and their family and friends. Those who could
afford the fund shares and were selected to borrow from the fund were
subsidized by a low or no cost loan.
As half of the funds studied did not experience capital growth, the number
of persons who could be served and the type of sanitation projects which could
be funded was limited; demand for loans often exceeded supply. Investigators
found no or low interest sanitation loans to be inconsistent with rural credit
markets in Thailand, and with fund viability and growth.
Based upon an analysis of 14 multipurpose funds and of drug funds which
had diversified, researchers recommended that near market rates of interest be
charged and that multipurpose funds be established.
HAITI
Design. Incentives were needed in Haiti to motivate CHWs to deliver
preventive services. Researchers concluded that communities were unwilling to
pay for preventive care directly but would donate cash to obtain access to
credit. Based upon this finding, researchers devised and implemented an
innovative credit program to raise funds on an on-going basis for recurrent
expenses of CHWs and to induce women to learn child survival skills.
Loaning to a group was seen to be less risky than loaning to individuals.
Researchers found, however, that women were reluctant to borrow money for
commerce or production outside the family as a member of a group because they
did not trust each other with common funds. Therefore, the first step in
implementing a group credit scheme was to create affinity groups whose members
would be willing to borrow as a group.
Mothers with children in severe malnutrition (a major group at health
risk) were indentified from health records and invited to attend a special
three-day seminar in their village, The seminar covered health as well as
other topics of interest to women, Participants were asked to bring two
trusted friends to the seminar. The seminars were then organized on a monthly
basis, and each newly created group was given the opportunity to invite other
friends. Only those trusted by all other members could be invited to join the
group. Completed groups had three to seven members. Facilitators supported
the group formation process.
As each group became established, members were asked to collect dues to
create a group fund. Viable groups able to successfully create a fund and
engage in samll-scale, on-going projects were invited to join a credit scheme
by purchasing a health/credit card at a cost of $2 per group member. The
credit scheme was linked to preventive care by requiring members to learn a
set of child survival skills before they could purchase a card and by
directing proceeds of the card sales toward the payment of CHWs. In order to
be eligible to buy a card, each member of the group had to demonstrate
competency in four child survival interventions (ORT, immunization, growth
150
monitoring and family planning); their children had to be up-to-date in their
vaccinations and had to participate regularly in growth-monitoring sessions.
Arrangements were made with a local agricultural bank to lend each group
money at 14 percent interest (pre-paid) providing that a sponsoring PVO put
1/5 the amount of the loan in a bank savings account as a frozen guarantee
fund. (Interest rates charged by private credit sources ranged from 60% to
300% a year.) The amount of credit for which each group was eligible was
determined by the amount of money in their group fund. The loan was given to
the group as a whole. Individuals could then borrow from the group. The
repayment period was flexible, ranging from a few months to one year.
Management. Mothers in each affinity group managed their respective
funds and fixed the amount of money they were willing to contribute.
Sixty affinity groups have been formed to date and interest in the project
is keen. Credit schemes are attractive to rural women in Haiti because there
is a great demand for credit, and interest rates for loans from private
sources are very high.
Future plans in the credit scheme include establishing procedures to
"roll-over" loans to support longer-term commercial enterprises and to advise
women about financial management techniques and types of income generating
projects likely to be successful.
151
! 401
- Media
14014.pdf
Position: 2229 (4 views)