Counting the Dead in India in the 21ST Century
Item
- Title
-
Counting the Dead in India in the 21ST
Century
- extracted text
-
A iz/>
V:
.
' - '' •' '.'<v ..
•. ’
I
4tl
J
a
■■
j
I
Proceedings of the International Workshop
on Certification of Causes of Death
2z~24 February, 1999
Tata Institute of Fundamental Research
Mumbia, India
Co-sponsored by the
Birhanmumbai Municioal Comoration,
World Health Organization, Wodd Bank, a id
Centers for Disease Control *.nd Prevention
Jiy, (X<-
'
MC
CKMTERS FOM
COMTMOV
AMD PHEVftfno*
Counting the Dead in India in the 21 st
Century
Proceedings of the International Workshop
on Certification of Causes of Death
22-24 February, 1999
Tata Institute of Fundamental Research
Mumbai, India
Co-sponsored
by the
Birhanmumbai Municipal Corporation
World Health Organization
World Bank
Centers for Disease Control and Prevention
Summary
By making a few simple and practicable modifications to current
practice, India can substantially improve the reliability of its birth, death,
and health statistics. Reliable statistics help in monitoring the changing
importance of particular diseases as causes of premature mortality and in
monitoring the current and future importance of external factors such as
HIV, tobacco, and alcohol as causes of death.
The International Workshop on Certification of Causes of Death,
convened on February 22-24, 1999, in Mumbai, India, strongly endorsed
the leadership of the Registrar General of India in upgrading India’s
vital statistics registration and cause-of-death systems for the next
century.
The following are the key recommendations of the Workshop:
► Add brief, simple questions on the use of smoking and chewing
tobacco to routinely collected death certificates.
► Strengthen the reporting and completeness of birth and death
registration, especially in rural areas.
► By 2005, improve cause-of-death certification by using newer
methods of collecting data and focusing on increasing the number of
completed medical certifications in urban areas.
► Support further operational research and evaluation of vital
registration and cause-of-death reporting systems.
page 2
Samira Asma and Prabhat Jha wrote this report, with input from
Workshop participants. It reflects the views of the Workshop
participants and not necessarily the sponsoring institutions.
For more information, contact Samira Asma at sea5@cdc.gov or Prabhat
Jha at pjha@worldbank.org
page 3
Introduction
Timely and accurate health data are essential to public health
surveillance efforts that monitor trends in vital events, diseases, injuries,
and disabilities. More timely and accurate release of vital statistics has
been identified as a priority by national, state, and local health agencies;
by academia; and by private organizations.
The creation of systems for collecting vital statistics in developed
countries such as the United Kingdom and the United States has had an
enormous effect on the organization and funding of disease control and
public health programs. Reliable health data and vital statistics
encourage researchers to develop appropriate interventions and
discourage them from using substantial resources against insubstantial
problems. Health data also help to direct research, inform public policy,
and change personal behavior. They also help to inform the public about
a few disturbing trends in disease, while reassuring them about other
trends (the large majority of which are favorable).
The United Kingdom’s nationwide death registration started in about
1837 and was functional within a few decades. India and other
developing countries can benefit from such experience to implement
complete systems in one to two decades. Unlike its present and past
systems, India’s new vital statistics registration, causes-of-death reports,
and analytical epidemiology must deal with India’s unique problems
related to childhood and maternal mortality, communicable diseases
such as tuberculosis and HIV/AIDS, and the emergence of noncommunicable diseases. These new emerging diseases pose major
organizational and technical challenges. It is promising to note,
however, that the Registrar General of India has taken a leadership role
in formulating new strategies for registering vital statistics (including
causes of death) during the next century.
page 4
Objectives of the Workshop
► To discuss methods that would improve the quality and coverage of
the certification of causes of death in India and to make
recommendations for improvement.
► To review current analytical epidemiology on major avoidable
causes of non-communicable diseases (such those related to tobacco
use) and to assess its impact on public health.
► To explore feasible approaches for eliminating the magnitude of
avoidable deaths due to various causes such as tobacco use.
To accomplish these objectives, the Workshop participants -
► Collaborated with the Registrar General of India and with members
of academia to develop options for improving cause-of-death data
and analytical epidemiology in India.
► Discussed key gaps and methodological issues in data collection.
► Critically reviewed the status of the epidemiology of tobacco use in
India.
This report presents selected findings and recommendations from
Workshop discussions. Appendix A is copy of the death report proposed
by the Registrar General, Appendix B consists of Workshop abstracts
and presentations, and Appendix C is a list of Workshop participants.
page 5
Revisions to the present death-certificate:
Options for the Registrar General to consider
The Workshop participants acknowledged the strong leadership of the
Registrar General in considering the suggestion of placing new, key
questions on death-report forms beginning in January 2000. Listed
below is the Registrar General’s current proposal (see appendix A for a
complete Death Report) and two suggestions (Option 1 and Option 2)
proposed at the Workshop with the advantages and disadvantages of
each.
Registrar General's Current Proposal
Questions 14-19 are listedfrom the Death Report Form (see appendix A
for complete report).
14. If used to smoke regularly (enter the following information)
a) For how many years? i) Till death; or
b) Till when?
ii) Till
years before death
15. If used to inhale snuff regularly (enter the following information)
a) For how many years? i) Till death; or
b) Till when?
ii) Till
years before death
16. If used to chew tobacco regularly (enter the following information)
a) For how many years? i) Till death; or
b) Till when?
ii) Till
years before death
17. If the used to chew arecanut in any form (including pan masala) enter
the following information)
a) For how many years? i) Till death; or
b) Till when?
ii) Till
years before death
18. If used to drink alcohol regularly (enter the following information)
a) For how many years? i) Till death; or
b) Till when?
ii) Till
years before death
19. If used to take narcotic drugs regularly (enter the following
information)
a) For how many years? i) Till death; or
b) Till when?
ii) Till
years before death
page 6
Advantage
► Questions are comprehensive and specific to Indian habits.
Disadvantages
► Too many of the questions could produce ambiguous responses,
which would decrease completion rates.
► Term regularly is unclear. In the United Kingdom’s International
Study of Infarct Survival (ISIS) case-control study, questions about
regular drinking created confusion, resulting in inability to analyze
the answers.
► Duration of smoking is hard to measure. For example, a study by
Gupta et al. in Mumbai showed that the correlation between answers
given 2 years apart about ever smoking was 95%, but the correlation
for duration ofsmoking was only 60%. Even more confusion would
result when living relatives are asked about the smoking habits of a
deceased person. Moreover, quitting is uncommon in India and other
developing countries, so data on quitting may not be reliably
obtained through questions about duration.
► Some questions are unnecessary: use of inhaled snuff is rare and so
is use of arecanut without tobacco. Data from the Mumbai Cohort
Study suggest only 0.5% of the population use snuff (Gupta et al. see appendix B). In the Mumbai Cohort Study, the use of arecanut
without tobacco did not indicate excess mortality (rate ratio adjusted
for age was 1.01 formales).
► The question about alcohol is unclear, and responders might easily
confuse 1) binge and light drinking and 2) drinking home-made
liquor rather than liquor bought from a store. Moreover, the question
does not capture data on deaths due to alcohol poisoning.
► The question about narcotics is general rather than specific to the
various types of narcotics (e.g., bhang, hashish).
► Questions about alcohol and narcotic drugs, even when they are
about the habits of dead people, could deter people from responding
to any questions, given their fears of legal implications.
page 7
Option 1 of Workshop Participants
Add two simple questions to the presentform for reporting the deaths of
adults, and omit the proposed questions 14-19.
About 5 years* ago,
Definitely
Definitely
No reliable
yes
no
answer available
□
1) Was the dead person a smoker?1
2) Was the dead person a quid chewer?* □
□
□
□
□
Mark No reliable answer available if the information is not obtained from
someone who lived with the dead person.
* Asking about a person’s habits 5 years previous to death is intended as a way of
learning about habits before they were affected by illness.
■[Smoker means somebody who smoked on most days.
XChewer means somebody who chewed a quid of any type on most days.
Advantages
► Questions are simple, easy to complete, and well validated in studies
in Mumbai, India; China; and South Africa.
► Format is consistent for all questions, with simple yes or no answers.
► Use of No reliable answer available column helps to ensure that
comparisons are among certain categories, which helps to minimize
bias in analyses of proportional mortality even if the overall number
of available answers is small.
► These simple questions would probably generate powerful
information and help to track the tobacco-use epidemic over time,
across states, and at different ages.
► The question avoids the problem of confusing current smokers’ risk
for disease with past smokers’ risk for disease. Data from Mumbai,
India, and South Africa suggest that the former smokers’ risk for
lung cancer is nearly as high as that of current smokers. In India,
quitting rates are low for smokers (<5%), and quitting is mainly
because of disease onset. Given that the death report is unable to
reliably capture long-term quitting due to insufficient knowledge
and poor recall problems, it is appropriate to have only two
categories: user and non-user.
page 8
► Asking about smoking that occurred 5 years previous to death (or 10
years previous) reduces or eliminates the chance that data will be
skewed by changes in tobacco-use habits that arise from onset of
disease (e.g., heart patient who quits life-long smoking on medical
advice and then dies).
Disadvantages
► Information on the use of chewing tobacco taken from hospital
records was not useful in a study in Mumbai (Gupta et al.), but
questions directly to surviving relatives produced useful data in
Chennai (Gajalaxmi et al. - see appendix B)).
► No question about alcohol use. This is partly because of the
difficulty of doing so in a simple and reliable way, and partly
because questions about alcohol can cause offence (compromising
the tobacco component and leading to unreliable evidence on
alcohol itself), and partly because alcohol causes fewer deaths than
tobacco does.
page 9
Option 2 of Workshop Participants
Same as Option 1 (two simple questions on smoking and chewing) plus a
question on alcohol use.
About 5 years* ago,
Definitely
Definitely
No reliable
yes
no
answer available
1) Was the dead person a smoker?t
□
□
□
2) Was the dead person a quid chewer?*
□
□
□
3) Did the dead person drink alcohol1
on most days?
□
□
□
Mark No reliable answer available if the information is not
obtained from someone who lived with the dead person.
*Asking about a person’s habits 5 years previous to death is intended as a way of
learning about habits before they were affected by illness.
'[Smoker means somebody who smoked on most days.
XChewer means somebody who chewed a quid of any type on most days.
U Alcohol includes both home-made and store-bought alcohol
Advantages
► Same as for Option 1.
► In addition. Option 2 captures data on alcohol use, which can be
analyzed by rural versus urban residents and by age differences of
deceased.
► Option 2 allows researchers to compare mortality due to tobacco use
with mortality due to alcohol use.
Disadvantages
► Question on alcohol may not be able to reliably capture different
patterns of drinking, and the question is not well validated by
studies.
► Question on alcohol would probably miss intermittent drinking.
► Questioning people about alcohol use could decrease response rates
for all questions.
► Question on alcohol can compromise the tobacco component and
lead to unreliable evidence of alcohol use itself.
page 10
Issues Related to Vital Registration
In India, registering births and deaths is a major challenge. There are
about 24 million births and 9 million deaths per year. Of all unrecorded
births in the world, 1 in 3 occur in India. There is large variation in the
extent of birth registration across states and between urban and rural
areas. This situation demands that vital registration be made a priority
(see Vijayanunni, - see appendix B).
Most deaths in rural areas occur in the person’s home, and these are the
deaths most under-reported. Overall, 7 in 10 deaths occur at home with
the remainder occurring in institutions (e.g., hospitals or nursing homes).
Deaths of elderly and young children are particularly prone to omission.
In Tamil Nadu, for example, infant deaths are under-reported by twothirds. With the sample registration system (SRS), overall death rates are
between 50% and 100% higher than they are with the civil registration
system (CRS), although the gap is narrowing in some states. In contrast,
the gap between SRS and CRS rates of infant mortality is not narrowing:
SRS rates are two- to three-fold higher than CRS rates.
Recently, the ’’capture-recapture" method was used near Calcutta to
register deaths and births. After an incomplete CRS, households were
canvassed for information on deaths. The capture-recapture method
estimated about twice as many deaths as did either the CRS or the SRS
method. (Sekar and Deming, 1993). Workshop participants did not
address in detail issues related to registering maternal deaths and the
deaths of children younger than 5 years old, since these issues were
outside its scope of work. However, they did emphasize that these issues
deserve considerable attention.
There is variation in rural-urban migration (e.g., Pondicherry has 130%)
of expected deaths due to in migration. The level of registration exceeds
100% due to people from neighboring areas outside some states come to
avail better medical facilities and due to de facto method of registration
get registered in some states. Incentives to report deaths are poor, and
the implementation of the Registrar General’s guidelines varies from
area to area. Timeliness of reporting and publishing the results of the
reports are slow (e.g., 1994 data are presented in 1999).
A National Identification Scheme is planned, but little progress has so
far been made. Evidence from the South Africa and United Kingdom
methods of collecting vital statistics suggests that assignment of a
unique number to each person is of great importance. Giving each
person an Election Commission voter card with a unique number has
been proposed in India, but the progress toward implementing the
proposal is extremely slow.
page 11
Workshop Recommendations
for Improving Registration of Births and Deaths
► Register all urban and rural burial and cremation grounds, and do
not permit any unregistered burial or cremation ground to be used.
► Do not permit any unregistered corpse to be buried, cremated, or
disposed off in any way.
► Add Deceased’s duration of stay in the city where death occurred to
the form for registering vital statistics in all municipalities. This
would help to separate data on deaths of residents from data on
deaths of migrants. We need this data in order to have an appropriate
denominator for calculating death rates.
► Waive fee for late registration of infant deaths so as not to penalize
people who are slow to report such deaths.
► Use capture-recapture methods, including better comparisons of
SRS and CRS, for key areas such as infant and maternal mortality
and for adult mortality and fertility.
► Review incentives, training and staffing for functionaries who
register deaths, including those who coordinate records with those of
other ministries (e.g., the Ministry of Health and Family Welfare).
► Encourage training for registration staff at all levels, including
computer training - perhaps, at first, focusing on the staff in a few
pilot states.
page 12
Issues Related to Reporting Causes of Death
The accuracy of cause-of-death data varies by cause-of-death and by
location where death occurred (rural vs. urban). For example in Tamil
Nadu, the percentage of medically certified deaths is about 14% of the
total registered deaths and about 40% of the registered urban deaths
(Ramodass. - see appendix B). Use of the International Classification of
Diseases-Modification 10 (ICD-10) has been proposed but is not yet
implemented in most of the country.
There is variation in the level of enforcement of burial regulations. For
example, people in Calcutta can bury their dead with only a note from a
doctor, but people in Chennai and Mumbai must have Form 8 (Medical
Certificate of Cause of Death).
Comparisons between states presented at the Workshop showed causeof-death data to be of variable quality. For example, the percentage of
"ill-defined” death codes ranges from about 10% in Goa to 30% in
Mumbai. Comparison between cause-of death reports and verbal
autopsies in Chennai (Gajalaxmi et al. - see appendix B) showed that
verbal autopsy reduced "ill-defined” codes from 50% to 25% of all
deaths, and increased specificity of reported causes - The same verbal
autopsy study shows that in comparison with the cause-of-death data
gathered on deaths that occur in hospitals, sensitivity for verbal autopsy
ranged from 100% for cancer deaths, 86% for respiratory deaths, and
74% for vascular deaths.
The rural cause-of-death survey has been discontinued. Instead the a
new SRS is being developed to gather fertility and mortality data on the
people who live in 6,000 blocks with about 1000 people in each block.
As a result, about 1.1 million households will be interviewed. On
average, each block will have about 10 deaths and 30 births per year.
At the Workshop, there was substantial discussion about sentinel sites.
China has about 140 nationally representative disease surveillance points
(DSPs). These cover a population of about 10 million people for whom
almost all the deaths are medically certified and centrally collected by
the Chinese Academic of Preventive Medicine. The DSPs were
established in the early 1980s under a World Bank loan to China at a
relatively low cost. The system has proved invaluable for generating
reliable mortality statistics for China and the world.
Workshop Recommendations
for Improving Cause-of-Death Reports
► Have widespread medical certification of all deaths in urban areas
by 2005.
►
Standardize questions on cause of death.
page 13
► Take stock of what is happening with regard to death certification in
different states and municipalities, focusing on quality. Compare
SRS results with CRS results.
► Expand training (including computerized ICD-10 training and use),
especially in medical schools, at the Ministry of Health and Family
Welfare, and at other pertinent ministries.
► Develop protocols for piloting SRS to address key risk factors in
limited areas using verbal autopsy.
► Consider making cancer a reportable disease in areas with
population-based cancer registries.
► Discuss methodologies for conducting analytic work. Most analytic
studies will need follow-up of patients for the foreseeable future. In
some areas with a high level of registration of vital statistics, studies
may be able to determine total mortality through linkages with
causes of deaths.
page 14
General Workshop Recommendations
Workshop participants made two key recommendations:
► During the next decade, support efforts by the Registrar General and
others to strengthen registration of vital statistics, cause-of-death
reporting, and analytic epidemiology in order to equip India to meet
the health challenges of the 21st century.
► Support efforts by the Registrar General to add simple, brief
questions on tobacco and alcohol use to death-report forms.
Workshop participants also recommended the following:
► At both the national and state level, the Ministries of Planning,
Health and Family Welfare, and Biotechnology should help to
finance more operational research and evaluation of vital
registration and cause-of-death reporting systems. Seek support
from international organizations such as the World Bank, the United
Nations and its agencies such as the World Health Organization, and
from the Centers for Disease Control and Prevention in the United
States.
► During the next 4 to 6 months, pilot test the feasibility of suggested
options for modifying cause-of-death reports. Aim to cover several
hundred deaths in the pilot test.
► Form a consortium of people to work on improving the vitalstatistics registration, cause-of-death reports, and non-communicable
disease control and monitoring.
► Set priority areas for research. Establish and evaluate sentinel
registration sites, and validate and expand verbal autopsy
techniques. Further discussion about making SRS work reliably
could be important, including a discussion on establishing
sustainable sentinel sites.
► Meet again in October 1999 to review the status of registration of
vital statistics in India, to assess progress toward goals, to conduct
relevant training, and to help inform further action.
page 15
Appendix A
Death Certificate
i
DEATH REPORT
DEATH REPORT
FORM NO.2
Statistical information
To be filled by the informant
Legal information
To be filled by the Informant
1.
Date of Death : (Enter the exact day,
month and year the death took place e.g.
1-1-2000)
2.
Name of the Deceased :
(Full name as usually written)
3.
4.
5.
Sex of the deceased :
(Enter “male" or “female"; do not use abbreviation)
Age of the deceased : (if the deceased was over
1 year of age, give age in completed years. If the
deceased was below 1 year of age, give age in
months, and if below 1 month give age in
completed number of days, and if below one day, in
hours)
8.
Place of death : (Tick the appropriate entry 1, 2 or 3 below and
give the name of the Hospital/ Institution or the address of the
house where the death took place. If other place, give location)
1.Hospital/
Name:
Institution
2.House
Address:
Informant’s name:
District:
c)
State:
Religion: (Tick the appropriate entry below)
1 .Hindu
2.Muslim
3. Christian
4. Any other religion: (write the name of the religion)
10.
Type of medical attention received before death: (Tick
the appropriate entry below)
1. Institutional
2. Medical attention other than institution
3. No medical attention
I
If used to smoke regularly (enter the following
14.
Information)
a) For how many years?
I) Till death; or
b) Till when? uj tju
years before death
15.
If used to inhale snuff regularly (enter the
following information)
a) For how many years?
u o 0 TUI death; or
b) Till when? jq jm
years before death
16.
If used to chew tobacco regularly (enter the
following Information)
If used to chew arecanut In any form (Including
pan masala) (enter the following Information)
17.
a) For how many years?
-u O 0 Till death; or
I '
b) Till when? yj Tj||;
years before death
s
Was the cause of death medically certified?: (Tick the
appropriate entry below)
I.Yes
2. No
Sto 12.
Cause of Death : (For all deaths Irrespective of whether
medically certified or not)
8 11.
j
a) For how many years?
I) Till death; or.
b) Till when? j[j -pm
years before death
Occupation of the deceased :
(If no occupation write ‘Nil’)
■o
If used to drink alcohol regularly (enter the
following information)
18.
a) For how many years?
'rm 4, o 0 ™death; or
’
b) Till when?
Tj||
years before death
.
<3
Address :
19.
€
so
(After
filling
all
columns 1 to 19,
informant will put
date and signature
here:)
Date:
b)
; ANNEX 4 ■
| s :
FORM NO.2
I
To be filled by the informant
Residence of the deceased: (Residence of the deceased
is the place where the deceased actually lived and can be
different from the place where the death occurred. Only
name of the Town/Village, District and State is required to
be entered here and not the house address)
a) Town/Village:
9.
2
3.0ther Place
6.
7.
___
i
TJ
5
o
13.
In case this is a female death, did the death occur while
pregnant or within 6 weeks after the end of pregnancy:
(Tick the appropriate entry below)
I.Yes
2. No
(Columns to be filled are over. Now put signature at left)
Signature or left thumb mark of the informant
To be filled by the Registrar
Name
Registration No.:
District:
Registration Date:
Tahsil:
Registration Unit:
Town/Village:
Town/Village :
If used to take narcotic drugs regularly (enter the
following Information)
a) For how many years?
n I) Till death; or
b) Till when? j-q Tm
years before death
To be filled by the Registrar
Registration No. :
Code No.
Date of Death:
Registration Date :
Sex: I.Male ZFemale
i
.
Age:
Years/months/days/hours
Place of Death : 1 .Hospital/lnstitution 2.House 3: Other Place
Registration Unit:
District:
Name and Signature of the Registrar
Name and Signature of the Registrar
Appendix B
Presentations and Abstracts
RECENT DEVELOPMENTS IN THE CIVIL
REGISTRATION SYSTEM IN INDIA
Dr. M. Vijayanunni
Registrar General St Consensus Commissioner, India
Civil registration was started in some areas of India in the last century. It
started with the registration of deaths with a view to collecting
information for control of pestilence and disease. Subsequently,
registration of births was introduced. In 1886, a Central Act-the Births,
Deaths and Marriage Registration Act was placed in the Statute Book.
However, registration was not made compulsory but was kept voluntary.
Different Acts were enforced in different parts of the country at different
points of time and even in a single state there were many Acts in force in
different areas. A few states had their own Acts that were adopted by a
few other states. Some Municipalities had separate Acts for registration
and some had enabling provisions in their Act providing for registration
of births and deaths.
The enactment of the Registration of Births and Deaths Act, 1969
provided the first uniform law for the whole country which made
reporting and registering of births and deaths compulsory and also
replaced all the diverse laws that existed on the subject and unified the
system. Foetal deaths of gestation period of less than 28 weeks of
duration are not required to be registered. It provides for a statutory
authority at the centre and in each state. It aims at enabling the Central
Government to regulate the registration and compilation of vital
statistics in the country so as to ensure uniformity and comparability,
leaving enough scope to develop an efficient system of registration on
the lines suited to the particular characteristics of the respective
administration. The responsibility of the implementation of the Act in
the states is that of the statutory authority in each state and union
B-2
the registration centres. This is true also for states that have already
achieved cent percent registration viz. Kerala and Goa (last report 1994).
The delay in reporting of the statistics from the local registrars,
eventually delay the compilation of vital statistics and its publication at
the state and national level.
The Registrar General's Office was seized of the problem and after an
in-depth study of the registration and statistical functions discovered that
a lot of paper work at the level of registrars is largely responsible for
such malaise. Therefore, it was felt necessary to review and rationalise
the actual process of registration and statistical work done by the local
registrar and find ways of reducing his workload, thereby ensuring
quicker transmission of data/records from the registration centres. One
of the. ways of reducing the workload was to keep the transcription work
of the registrar at the barest minimum. Keeping this objective in view it
has been now decided to replace the multi-event register by the loose
leaf single event form that is itself the legal part of the reporting form.
The registrar now has to register the event in the legal part of the
reporting form itself, thereby totally eliminating the need to have
separate register. The reporting forms that have been modified to
accommodate these new registration procedure now have two parts-legal
part, that contains only those items that, required for legal purposes and
statistical part, where only items that are statistical in nature are
included. At the end of every month the registrar only has to detach the
statistical parts of the reporting forms and despatch them for statistical
tabulations. The modification of the forms gave us an opportunity to
give a fresh look at the items of information currently being collected
with a view to excluding items that are not necessary and include those
which are important. Instructions for filling up the forms appear along
with each item for easy reference and making them, user-friendly.
Wherever possible, the items have been pre-coded to facilitate data
entry. Different colours have been suggested for the three different
forms (Reporting forms for birth, death and still birth) for easy handling
B-4
marriage, which in turn can be cross tabulated with the order of
birth to get estimates of fertility levels. In itself the age at
marriage data is a very useful data item.
c)
Method of delivery - This information is used to relate the
method of delivery with certain characteristic of mother like age
group, educational level, type of attention, at delivery, etc.
Information from this item can be used to monitor delivery
trends across the country and over the years.
d) Birth Weight - This is the single most important characteristic
associated with infant mortality. It is also related to the age of
mother, her literary status and other factors surrounding the
birth. Consequently, it is used with other information to plan for
and evaluate the effectiveness of health care. This item can be
easily collected from the medical institutions and through the
health workers/ANMs, wherever they have been notified to
collect the information on births and deaths from the informant.
e) Duration ofpregnancy - Duration of pregnancy when cross
tabulated by Weight and age group of mother gives a very good
insight into the reproductive health status.
The items "Nationality" and Permanent Address" have been excluded
from all the reporting forms.
The following items of information have been included in the new death
report form (Form No.2)
a) Residence of the deceased - The actual purpose it collect usual
residence of the deceased. This will help in tabulating the events
by place of usual residence.
b) Pregnancy-related death - This item is introduced for collecting
B-6
information on female deaths that occur during pregnancy or
within 6 weeks of the end of the pregnancy. It may be noted that
the end of pregnancy may. be due to abortion and may result in
still birth or live birth. This Will provide an upper limit of the
maternal deaths and the corresponding rates over space and
time. This item has been recommended for collection by the
World Health Organisation.
c) Tobacco, alcohol and narcotic habits - It has been decided to
collect details of addictive habits of the persons of the deceased.
The addictions included are smoking, inhaling of snuff, chewing
tobacco, chewing arecanut, drinking alcohol, and taking
narcotics. The duration of the habit and the time since person
had given up toe habit will also be collected. This information
will be related to the cause of death provides a deep insight into
the addiction-induced death. This information can be tabulated
by age of deceased for providing useful clue to the addiction
pattern prevailing in the country.
In the Still Birth Report form only those statistical part that will be used
for tabulation have been retained. These are "Age of mother at time of
birth", "Mother's level of education" and "Type of attention at delivery".
The new item that has been included in this form is "Duration of
pregnancy" as still birth, when cross-tabulated by age of mother and
duration of pregnancy, provides important health information.
The proposed reduction in the workload of the Registrar and the changes
recommended in the forms are likely to correct the situation to the extent
that the basic returns reach the designated compilation offices in time.
However, the problem would still persist if the data so received in the
compilation offices were not processed in time. Therefore, we are
preparing a time-bound action plan on processing of the vital statistics
data to ensure that the reports are made available to the data-users in
B-7
time. Common program software to be used by all the states will be
developed for use of all the states to integrate and unify the processing
and tabulation of CRS based vital statistics. We expect that with the
successful implementation of the new system, the civil registration
system would be in a position to provide the much needed vital statistics
at levels even below the state on a continues and permanent basis, at
least in the states where the registration levels have reached 90%.
B-8
Annex - 1
AGENCIES FOR REGISTRATION OF BIRTHS AND DEATHS IN STATES/UTS
CHIEF REGISTRARS
Health
Economics & Statistics
Others
Andhra Pradesh, Assam,
Gujarat, Haryana,
Himachal Pradesh, Jammu
& Kashmir, Maharashtra,
Manipur, Meghalaya,
Orissa, Punjab, Sikkim,
Tamil Nadu, Tripura, Uttar
Pradesh, West Bengal,
A&N Islands Chandigarh,
Lakshadweep,
Arunachal Pradesh, Bihar,
Goa, Karnataka, Madhya
Pradesh, Nagaland,
Rajasthan, Delhi
Kerala (Panchayats),
Mizoram (Chief Secretary),
Dadra & Nagar Haveli (
Secy.Admn.), Daman &
Diu (Finance Secretary),
Pondicherry (Local Admn.
Deptt.),
(16 States + 3 UTs)
(8 States+lUT)
(2 States + 3 UTs)
DISTRICT REGISTRARS
District
Health/Medical
Officer
District Statistical
Officer
' / "•
Deputy
Commissioner
Collector
Others
Andhra Pradesh,
Assam, Gujarat,
Haryana, Himachal
Pradesh, Jammu &
Kashmir,
Maharashtra,
Manipur,
Meghalaya, Orissa,
Punjab, Sikkim,
Uttar Pradesh,
Andaman &
Nicobar
Islands,Chandigarh,
Lakshadweep.
Madhya Pradesh,
Nagaland,
Rajasthan, Delhi.
Arunachal Pradesh,
Goa, Karnataka,
Mizoram, Tamil
Nadu, Tripura,
West Bengal, Dadra
& Nagar Haveli,.
Daman & Diu.
Bihar, (Dy.
Development
Commissioner),
Kerala (Distt.
Panchayat),
Pondicherry
(Municipal Admn.)
(13 States + 3 UTs)
(3 States + 1 UT)
(7 States + 2 UTs)
(2 State + 1 UT)
RURAL REGISTRARS
Andhra Pradesh, Bihar, Goa, Gujarat, Himachal
Pradesh,
Kerala, Maharashtra,
Manipur (Sub
Registrars), Rajasthan, Tripura (non - TTAADC
areas), Uttar Pradesh, West Bengal (Sub-Registrars)
Daman & Diu, Delhi, Pondicherry.
Panchayat Secretary
,
Police Officer
Haryana, Jammu & Kashmir, Madhya Pradesh, Punjab,
Chandigarh.
Viilage/Reveniie Official
Karnataka, Tamil Nadu, Dadra & Nagar Haveli.
Block Development Officer
Manipur (Non-hilly areas).
Tehsildar Sub-Divisional
Officer/Extra Asstt.
Commissioner
Arunachal Pradesh (Extra Assistant Commissioner
/Circle Officer), Tripura (Tehsildar in TTAADC
areas).
Primary Health Centre
Incharge/Health Inspector/
Block Sanitary Inspector
Assam, Manipur (in hilly areas), Meghalaya,
Orissa, Sikkim, West Bengal, A&N Islands,
Lakshadweep.
Teacher of Govt. School
Arunachal Pradesh (Sub-registrars), Mizoram,
Nagaland, Rajasthan (Sub-registrar in Some areas)
URBAN REGISTRARS
In urban areas registration is done by the Municipal authorities.
-
-----
*
ANNEX 2
Progress of Registration of Births and Deaths - India, States and Union Territories
Level of Registration and Receipt of Returns for the years 1992,1993 and 1994
India/
State/
Union territory
1
1992
2
Level of Registration1
Birth
Death
1992 1993
1993 1994
1994
1992
3
4
5
6
India
50.8
51.4
51.6
Percentage of
Returns Received
1992
1993
1994
8
9
10
1994
7
46.3
80.0
83.6
79.7
30.0
34.4
34.4 29.9
15.2
15.5
19.5
NA.
NA
NA NA.
29.6
29.8
24.1
102.7* 110.0* 112.3*
64.8
62.4
62.4 70.5
67.6
64.7
75.5
38.0
37.4
42.9
NA
NA
NA
75.5
66.5
55.3
89.3
87.9
89.9
53.2
48.1
52.6
69.4
68.9
71.7
16.5
27.1
15.1
NA.
NA
NA.
NA
NA
N.A.
49.9
52.1
60.2
58.4
NA
54.7
81.3
82.8
77.7
29.6
24.7
26.7
9.2
7.9
9.0
82.0
76.6
80.8
44.1
37.2
41.8
28.3
25.9
25.9
24.9
24.7
23.2
36.8
100.0
NA
98.8
100.0
94.1
100.0
71.0
100.0
N.A.
100.0
99.9
91.6
46.7
NA.
81.5
82.8
100.0
100.0
66.4
100.0
92.7
100.0
81.9
76.5
41.6
100.0
N.A.
96.8
100.0
96.5
95.9
73.0
100.0
84.2
100.0
99.6
95.5
57.8
N.A.
88.1
82.3
100.0
100.0
70.5
100.0
92.7
100.0
86.6
70.3
39.8
100.0
N.A.
78.6
100.0
91.2
99.6
75.7
100.0
85.3
100.0
99.6
91.2
55.0
NA
81.5
73.1
NA
100.0
73.2
100.0
93.9
100.0
80.6
78.2
44.1
46.0
States
Andhra Pradesh
Arunachal Pradesh
Assam
Bihar
Goa
Gujarat
Haryana
Himachal Pradesh
Jammu & Kashmir2
Karnataka
Kerala
Madhya Pradesh
Maharashtra
Manipur
Meghalaya
Mizoram3
Nagaland
Orissa
Punjab
Rajasthan
Sikkim
Tamil Nadu
Tripura
Uttar Pradesh
West Bengal
33.2
33.1
35.0
67.1
77.3 66.3
N.A.
NA N.A.
17.4
21.2 21.6
119.9* 121.5* 117.4*
95.9
92.7 92.1
68.5
63.8 61.4
58.4
52.8 51.5
N.A.
N.A.
N.A.
75.6
68.8 72.4
99.2 100.7* 99.2
49.7
49.9 51.0
76.2 '
77.6 77.2
13.7
18.8
18.1
NA.
N.A. NA
N.A.
NA
N.A.
94.6
87.9 88.3
67.1
N.A.
70.1
89.6
86.9 89.0
21.8
20.5 22.0
22.3
22.9 23.6
91.4
88.6 88.9
68.3 74.6 103.3*
32.6 35.1
37.0
53.4
47.0 44.7
Union Territories
100.0
100.0
NA
78.5
71.4
N.A.
Andaman & Nicobar Islands NA 120.0* 118.7*
100.0
100.0
100.0
259.5* 277.1* 220.6*
163.3* 147.9* 128.2*
Chandigarh
100.0
100.0
100.0
52.7
61.4
57.6
80.6
78.6 79.2
Dadra & Nagar Haveli
N.A.
N.A.
NA
67.4
NA
N.A.
135.1* NA NA
Daman & Diu
100.0
100.0
100.0
113.6* 109.6* 116.5*
115.9* 111.8* 110.1*
Delhi
N.A.
NA
N.A.
NA
N.A.
NA
N.A.
NA NA
Lakshadweep
100.0
100.0
216.4* 205.7* 200.6*
150.6*
146.8*131.3*100.0
Pondicherry
Notes:
1 The level of registration is the percentage of registered events to the SRS estimated events.
2 S.R.S. data for Jammu & Kashmir are not available.
3 S.R.S. was implemented in Mizoram only from 1996 onwards and hence level of registration could
not be worked out
* The level of registration exceeds 100 percent in these States/UTs because the people from the
neighbouring areas outside these States/UTs
States/UT^ ^ome
co
here to avail of better medical facilities and
due to the de facto method of registration all siith (events get registered in these States/UTs.
NA Not available since reports are yet to be received from these States/UTs.
|o <
■*
n vm w
0j713
AHO
MORTALITY DATABASE IN CHENNAI
Gajalakshmi CK,1 Peto R,2 Shanta V1
’Cancer Institute, Chennai, India
2University of Oxford, UK
Introduction
The population of Chennai is 4.2 million as on 1st July 1997 with a male
female ratio of 1:0.93. It has 33 registered burial grounds. The health
authority requires a death certificate at the time of disposal of the body.
A copy of the death certificate is sent to the Vital Statistics Division
(VSD) in Chennai for deaths that occur at the hospital. A medical
certificate is obtained from the private medical practitioner for deaths
that occur at home. If medical certificate is not available, then
symptoms, complaints, duration of illness etc are collected from the
relatives at the burial ground. Based on this, the probable cause of death
is determined by the Medical officer in the VSD. Autopsy of the dead
body is not routine. Deaths due to unnatural, accidental, or violent
causes are usually subjected to postmortem examination.
Accurate information on cause of death could help health authorities to
prioritize problems and interventions. Despite inaccuracies death
certificates remain as an important source of mortality data. Our earlier
study has shown that the sensitivity of death certificates to detect cancer
deaths is 57% (Gajalakshmi et al., 1998). So in order to compute cause
specific mortality for the residents of Chennai we need a tool other than
the death certificate. Verbal autopsy (VA) has been used extensively to
assess cause of death among children and adult females. VA is an
attractive alternative tool to assess the cause of death in countries where
a large proportion of people die at home with out medical care at the
time of terminal event and the registration of death is almost complete.
In Chennai the death registration is almost complete and about 70% of
deaths occur at home. Hence we undertook a pilot study on VA among
B-12
randomly chosen male deaths that occurred in Chennai during 1995-97
to assess whether it can be used as an indirect method to compute cause
specific mortality in a developing environment like India.
Methods
The Population-based Cancer Registry at the Cancer Institute (WIA)
commenced data collection in Jan. 1982 from the VSD. The procedure
followed to collect the morbidity data is described in detail elsewhere
(Shanta et al., 1994). The data required for the registry are not
computerized by the VSD but maintained in the registers by trained
staff. The Registry staff abstract data from death registers in the VSD to
a standardized format which includes the following: deceased’s age,
gender, marital status, occupation, place of death, permanent address,
date and cause of death (immediate/ underlying/ contributory), and
informant’s name. These were computerized in the registry and linked
with the incident cancer cases registered since 1982 through linkage
programs using computer. From 1982 to 1991, mortality data only on
cancer deaths were collected from the VSD. Since 1992, data on all
deaths irrespective of stated cause of death on the death certificate have
been collected from the VSD.
The Division of Epidemiology and Cancer Registry at the Cancer
Institute (WIA) has given special training to male university graduates
to interview the spouse and/or close relative of the deceased to collect
data on symptoms, signs and circumstances leading to death when they
visited the family of the deceased person. The field visit report was
entered in the blank page. The questionnaire has filter questions only for
heart attack and respiratory diseases. To avoid distress over the terminal
event the field visit was carried out 6 months after death. We assume
even the longer recall period will not affect reporting the circumstances
that lead to the terminal event.
B-13
cancer, 7-8% due to respiratory diseases including pulmonary
tuberculosis, 29-34% due to vascular disease, 40-44% due to ill-defined
causes and for about 4-9% of deaths, the cause of death was not
available in the VSD.
The pilot study on VA was carried out with 2500 randomly chosen
deaths that occurred in Chennai during 1995-97. About 311 houses
could not be traced due to the following reasons: shifted the house after
death, house demolished or inadequate address to locate the house.
About 71% of deaths in the random sample occurred at home (Table 6).
Table 7 shows that more deaths were seen over 55 years of age and most
of the responders were in the age group of 35-64. About 27% of the
deceased and 37% of responders were illiterates (Table 8). Table 9
compares the cause of death stated on the death certificate with the cause
of death derived from VA. As a result of verbal autopsy the deaths due
to ill-defined causes were decreased from 41% to 25% and the
percentage of deaths due to cancer, respiratory diseases and vascular
disease were increased compared to the data from the VSD. The cause of
death was derived based on verbal autopsy report for deaths which were
grouped under ‘unknown cause’ at the time of death registration, at the
VSD.
The diagnosis derived from verbal autopsy was compared with hospital
diagnosis of cause of death stated on the death certificate (Table 10).
The sensitivity of verbal autopsy to detect cancer deaths is 100% ,
vascular deaths 73.6% and respiratory deaths 86%. In Table 11 the
cause of death stated on the death certificate of deaths that occurred at
home was compared with verbal autopsy report. The sensitivity of death
certificate issued for deaths that occurred at home to detect cancer
deaths is 66.3% , vascular deaths 53.4% and deaths due to respiratory
diseases 38.6%.
B-15
Discussion
Accuracy of cause of death stated on the death certificate is essential to
compute cause specific mortality, estimate survival rates and for health
care planning. The death registration in Chennai is almost complete and
certification of deaths, which occur at home, is of poor quality. The
‘mortality to incidence ratio’ or ‘death in period’ which is one of the
indices of quality of registry data was 22% when the Registry collected
data on only cancer deaths from the VSD. This is because of the low
(57%) sensitivity of death certificate for cancer as underlying cause of
death [(1955/3420) *100]. i.e. 43% of cancer patients die due to causes
other than cancer. Thus collecting data only on cancer deaths will result
in registering only 57% of total deaths of cancer patients available in the
VSD, and 43% of deaths due to non-cancer causes will not be registered.
In India 65% to 70% of cases are seen in late stage of the disease at the
time of initial diagnosis. So majority of cancer patients might have died
due to cancer even though a non-cancer cause of death was mentioned
on the death certificate. Hence Registry started collecting data on all
deaths since 1992; this procedure has improved the mortality registration
in the population-based cancer registry in Chennai.
We need an alternative method to collect data on the cause of death to
minimize the misclassification of cause of death stated on the death
certificate. The VA appeared to be an alternative method, which is
widely used to collect data on cause specific mortality. The pilot study
on verbal autopsy carried out in Chennai among randomly chosen male
deaths that occurred during 1995-97 showed the possibility of
ascertaining the leading causes of death, reducing the misclassification
of cause of death on the death certificate and deriving cause of death
when it is not available in the VSD focusing on the importance of
collecting complete address with pincode would help in tracing the
houses for field visit and improve quality of death data being collected
from the VSD. They also should be given orientation courses on
abstraction of cause of death from death certificate. Organizing
B-I6
Table 1: Mortality data in Chennai PBCR: 1982-95
Years
1982-84(3yrs)
1985-91(7yrs)
1992-93(2yrs)
1994-95(2yrs)
VSD
(no)
Active F-up
(no)
Mortality
Total deaths
(no)
1601
3893
3420
3335
373
6116
548
424
1974
10009
3968
3759
__________
Morbidity
Incident cases
(no)
M:I ratio
7418
21555
6642
6887
26.6
46.4
59.7
54.6
%
Table 2: Extent of disease at the initial diagnosis-PBCR: 1992 & 1993
Extent ofdisease
Localised
Regional extension
Advanced/metastas.
Lymph & Leukae
Unknown
Total
No
237
4147
1020
470
476
6350
%
3.7
65.3
16.1
7.4
7.5
100.0
DCO excluded (n = 292)
Table 3: Extent of disease at diagnosis and cause of death by place of death from VSD
Data - PBCR: 1992 & 93
!
Extent ■of disease
Govt hospital
Ca
Nbn-ca
10
12
138
76
128
9
56
5
20
13
Localised
Regional extension
Advanced/metastas.
Lymph & Leukaem
Unknown
Total
No.
%
352/467
75.4
115/467
24.6
DCO excluded (n = 292)
B-18
Place of death
Private Hospital
Ca
Non-ca
11
15
148
85
147
20
57
5
37
18
400/543
73.7
143/543
26.3
Home______
Ca _____ Non-ca
53
100
577
801
146
141
61
45
74
120
911/2118
43.0
1207/2118
57.0
Table 6: Place of death of the deceased in Pilot-Study
On verbal autopsy
Place of death
No
%
Hospital
Home
631
1558
28.8
71.2
Total
2189
Table 7: Age group-wise distribution of the deceased and responders in Pilot Study
on verbal autopsy
Age group
<25
25-34
35-44
45-54
55-64
65-74
75+
Total
Deceased
No _______
%
100
176
334
456
512
611
4.6
8.0
15.3
20.8
23.4
27.9
Responders
No________ %
1.7
38
10.1
220
19.6
429
25.2
552
23.3
511
15.4
337
4.7
102
2189
100.0
2189
100.0
Table 8: Educational level of the deceased and responders in Pilot Study on verbal
autopsy
Responders
Deceased
Educational level
No
%
Illiterate
<6 yrs
6-8 yrs
9-12 yrs
>12 yrs
584
321
429
566
289
Total
2189
------B-2CT------
26.7
14.7
19.6
25.9
13.2
No
817
380
454
408
130
No(%)
37.3
17.4
20.7
18.6
5.9
100.0
2189
100.0
Table 9: Causes of death based on the data from the vital statistics division (VSD) and
verbal autopsy
Causes of Death
among males______
Cancer
Respiratory diseases
including PT
Vascular disease
Infection except
Tuberculosis
Illdefmed causes
Other medical illness
Unknown
External causes
Total
Based on Death
Certificate from VSD
No
%
Based on Field
visit report
No
%
118
5.4
165
7.5
155
653
7.1
29.8
304
760
13.9
34.7
26
896
193
148
1.2
40.9
8.8
6.8
56
549
300
2.6
25.1
13.7
55
2.5
2189
2189
Table 10: Causes of death of those who died at the hospital were compared with
Verbal autopsy
t
Verbal autopsy
Causes
Cancer
Vascular
Respiratory
Other causes
Total
Hospital deaths from VSD
Cancer
51
51
Vascular
5
170
17
39
231
Respiratory
1
1
50
6
58
Other causes
7
100
22
162
291
Total
64
271
89
207
631
Verbal autopsy compared against hospital deaths:
Sensitivity of verbal autopsy to detect cancer deaths is 100% [(51/51)* 100]; vascular
deaths 73.6%[(170/231)*100]; respiratory deaths 86% [(50/58)* 100].
B-21
*
Table 11. Causes of death of those died at home compared with verbal autopsy
VSD
VSD
Cancer
Cancer
67
Vascular
13
Respiratory
1
Other causes 20
Total
101
Verbal Autopsy
Vascular
Respiratory
Other causes
261
4
224
489
45
83
87
215
103
9
641
753
Total
67
422
97
972
1558
Home deaths co: apared against verbal autopsy:
Sensitivity to decect cancer deaths is 66.3% [(67/101)* 100], vascular deaths
53.4%[(261/489)*100] and respiratory deaths 38.6% [(83/215)*100],
j
------ B-22---
•»
DEATH REGISTRATION IN RURAL TAMIL NADU
R. Ramodass
Tamil Nadu
Introduction
The registration of vital events keeps a continuous check on
demographic changes. Once the registration of births and deaths is
complete and accurate, it can serve as the back bone for the Health
Information System.
Tamil Nadu has a long tradition of registration of births and deaths.
Prior to the introduction of ’’Registration of Births and Deaths Act
1969’’(Central act 18 of 1969) by the Government of India, registration
of births and deaths in Tamil Nadu was carried out under the provisions
of Madras Panchayat Act HI of 1899 in the Rural areas, the Madras
Districts Municipalities act 1920 in the Municipalities and select Town
Panchayats and the Madras city Municipal act 1919 in Madras
Corporation. Even though these acts provided the procedures for
registration of births and deaths, they were not covering the entire state
and not uniform.
The Registration of births and deaths was made compulsory at the place
of occurrence with the introduction and implementation of ’’Tamil
Nadu Births and Deaths Registration Rules 1977’’. with effect from
15th March 1977 in accordance with the provisions in section 30 of the
RBD act 1969.
B-23
Registration Hierarchy
The Registration hierarchy in Tamil Nadu is as follows:
Director of Public Health and Preventive
Chief Registrar
Medicine
Joint Director (State Bureau of Health
Deputy Chief Registrar
Intelligence)
District Revenue Officer/Additional
District Registrar
Collector
Deputy Director of Health Services
Additional District Registrar
Village Administrative Officer
Registrar of Births and Deaths in the
Village Panchayat
Executive Officers/Sanitary Inspectors of
Registrar of Births and Deaths in the
the Town Panchayat
Town Panchayat
Sanitary Inspector of the Ward
Registrar of Births and Deaths in the
Municipalities/Corporations
The powers or legal actions against the defaulters are vested with the
Additional District Registrars in the Non Municipal Areas and
Municipal Health Officers/Commissioners in the Municipal Areas.
A time limit of 14 days for births and still births and 7 days for deaths is
provided in the rules for registration of births and deaths. The rules also
provide for late registration of births and deaths beyond the time limit
specified above with late fee and other administrative procedures.
There are more than 13,000 registration units in Tamil Nadu as detailed
below:
Village Panchayats
12.456
Town Panchayats
643
Municipalities
104
Corporations
6
Cantonments
2
13,202
Total
B-24
The birth and death returns from the village panchayats are sent to the
Deputy Director of Health Services concerned through the Tahsildars
and in the case of town panchayats the returns are sent direct to the
Deputy Director of Health Services concerned for compilation. The
returns from the Municipalities/Corporations/Cantonments are sent
directly in the consolidated form to the Director of Public Health and
Preventive Medicine.
Efficiency of the Registration System
The efficiency of the registration system in the State is monitored
through indicators like registration efficiency and reporting efficiency.
The reporting efficiency under the Civil Registration System (CRS) has
improved from a level of about 80% to above 90% over the last 10
years. The registration efficiency of births has increased from a level of
about 70% to 90% over the last 10 years. The registration efficiency of
deaths has increased from a level of about 60% to 75% over the same
period.
Status of Death Registration
The responsibility to report the death for registration is fixed on:
1. The eldest male member of the household or the nearest adult relative
of the deceased in the case of domiciliary deaths.
2. The medical officer incharge of the institution in the case of
institutional deaths.
3. The person incharge of the vehicle in case of deaths occurring in
moving vehicles.
4. The person incharge of lodging houses/choultries in the case of
deaths occurring in lodging houses and choultries.
B-25
5. Police officials in case of death occurring in the platforms or due to
road accidents. In the case of deaths involving inquest the responsibility
is fixed on the enquiring officer.
About 3.6 lakhs of deaths get registered every year in the State as
against an expected 4.8 lakhs of deaths. Of these 3.6 lakhs of deaths,
about two thirds get registered in the rural areas and about one third get
registered in the urban areas. About 19% of deaths are institutional
deaths in the State as a whole. The percentage of institutional deaths in
the urban areas is about 46% and only 5% in the rural areas. The
percentage of medically certified deaths is about 14% of the total
registered deaths and about 40% of the registered urban deaths. The
medical certification of cause of deaths is obtained in form 8 for
institutional events and in form 8A for non-institutional deaths. But the
certification in the prescribed forms is missing in quite a few cases, for
which action is being taken to rectify the deficiency.
In corporations and municipal areas, cause of death is insisted upon at
the time of registration of deaths. In Chennai Corporation, doctor’s
certificate on cause of death is insisted upon even at the time of disposal
of the dead body. In the rural areas this is not insisted upon due to lack
of facilities.
Impediments in Death Registration
The following are the obstacles in ensuring 100% death registration and
in improving upon the cause of death statistics.
1. Infant deaths especially those happening within the first seven days
of life seldom get reported and registered, Infant deaths from around
12% of total deaths as per the recent survey conducted in the state while
only 4% of infant deaths get registered under civil registration. The
omissions in registration of infant deaths is probably due to some infant
deaths getting misclassified as still births, and also due to the prevalence
B-26
of female infanticide in some pockets and the consequent fear of legal
action.
2. The deaths of persons getting discharged against medical advice do
not get registered especially when the body is taken away to native
villages far away from the urban institutions.
3. Deaths involving police inquests are registered belatedly or registered
without correct cause of death particulars for want of post-mortem
reports or inquest reports.
4. There are no registered burial/burning grounds in the rural areas and
also in most of the municipal towns, due to which systematic
notification of all deaths cannot be ensured.
5. Section 10 (3) of the RBD act provides for the issue of a certificate
regarding the cause of death from the medical practitioner who attended
during the last illness of the deceased. The proof of attendance lies with
the medical officer and hence enforcement of this portion of the act is
rendered difficult. Most of the poor people are made to obtain a
certificate at a cost.
6. About 40% of the deaths occur to old age people and it is possible that
such cases are not attended to by any medical practitioner and hence
getting cause of death certificate is a problem in such cases. If cause of
deaths certificates are insisted for such cases they may turn out to be a
routine certificate with respiratory arrest as cause of death.
Measures Taken To Improve Death Registration
Tamil Nadu Govt, have introduced social security measures such as the
following:
1. Grant of financial assistance from the Chief Ministers Relief Fund to
B-27
the family of the persons died in natural calamity, fall from trees and
snake bites.
2. Grant of financial assistance for the funeral rites of scheduled
caste/tribes.
Both of the above two measures help in improving death registration, as
the cash is disbursed on the production of death certificate.
Suggestions For Improvement of Death Registration
1. All burial/burning grounds must be registered and no corpse should
be allowed to be disposed of in unregistered burial/burning grounds or
river sides.
2. The local bodies may arrange to open a death registration counter at
the major institutions to enable on the spot registration and issue of
extracts.
3. Late fee should be waived I the case of infant deaths reported or
detected after the time limit, by suitable amendments to RBD Act.
4. For deaths occurring in moving vehicles the place of registration may
be changed as "permanent place of residence of the deceased’ instead of
the "place of first halt".
5. To improve the cause of death statistics and medical certification of
cause of deaths all the doctors both in government and private sector
must be given a sensitization on the needs to certify the cause of deaths
in the ICD format. Trainers training should be arranged at the
National/Regional levels.
B-28
Rural Survey of Cause of Death Scheme in Tamil Nadu
In the rural areas of Tamil Nadu, 95% are domiciliary deaths and it is
here that the relevant statistics on cause of death is missing.
In order to have some reliable data on cause of deaths the scheme of
"survey of cause of death" was implemented in 150 Health Sub Centers
of Tamil Nadu covering a population of 7.3 lakhs for a period of 3 years
from 1996 to 1998. In the earlier years, lesser number of units were
covered under the banner of "Model Registration Scheme".
In these 150 Health Sub Centers, the Female Health Worker at the health
Sub Center level acts as the field agent who is responsible to record
births and deaths that occur to the usual resident population. The
probable cause of death is arrived at through an after death enquiry of
the facts and circumstances of deaths as recalled from the memory of the
household. The medical officer of the Primary Health Care Center test
checks the cause of deaths. Have yearly surveys are conducted once in
six months by the Health Inspector of the Health Sub Centers during
January and July every year and all the births and deaths that take place
in the area to the usual resident population are netted. Now the Rural
Survey Cause of Death scheme stands withdrawn by GOI with effect
from 1-1-99. Since it will take a long time before civil registration
system attains the desired level of efficiency, the scheme of survey of
cause of death needs to be continued. The statistics on cause of deaths is
also useful for planning purposes especially those related to Infant
Deaths and Maternal Deaths. If this scheme is continued with an
increase in the sample size it will definitely yield cause of death
statistics at state level and at district level also. It is suggested that 40
Health Sub Centers per district covering 2 lakhs population can be taken
up.
B-29
Cause of Deaths as per Civil Registration System - 1996
Tamil Nadu
S.No
01
02
03
04
05
06
07
08
09
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
%
CAUSE OF DEATH
CHOLERA
LEPROSY
WHOOPING COUGH
POLIOMYELITIS
MALARIA
TYPHOID & PARA - TYPHOID
OTHER FEVERS
DIARRHOEA-DYSENTERY - A.G.E
ANAEMIA
TUBERCULOSIS
BRONCHITIS & ASTHMA
JAUNDICE
CANCER
HEART ATTACK
DIABETES
PARALYSIS (CEREBRO - VASCULAR)
TETANUS
DIPHTHERIA
SUICIDE
INFLUENZA
FOOD POISONING
CHRONIC LIVER DISEASE
ULCER OF STOMACH
RABIES
BITES OR STINGS OF ANIMALS, INSECTS ETC
SYPHILIS & OTHER GEN I TO-URINARY DISEASES
MATERNAL DEATHS - OTHERS
MATERNAL DEATHS - ABORTIONS
PREMATURITYz BIRTH INJURY
MENINGITIS
SENILITY
MEASLES
BURNS - ACCIDENTS
APPENDICITIS
FALLS r DROWNING
ACCIDENTAL POISONING (EXCEPT FOOD POISON)
TRAFFIC ACCIDENTS (ALL TYPE OF TRANSPORT)
OTHER ACCIDENTS PNEUMONIA
HOMICIDE
ALL OTHER CAUSES
B-30
-
0.02
0.04
0.02
0.02
0.00
0.16
8.94
1.15
0.91
3.96
5.26
1.27
2.51
22.28
1.57
3.37
0.23
0.003
1.19
0.002
0.024
0.34
1.67
0.04
0.33
0.34
0.20
0.03
2.84
0.13
18.69
0.03
1.09
0.03
0.14
0.05
1.13
0.32
0.31
0.11
19.25
SURVEY OF CAUSE OF DEATH ( RURAL )
S.No
1997
Major Cause Groups
No.of
Deaths
i^
2^
3.
4^
5.
6.
7.
io.
] Accidents and Injuries
801
] Child Births and Pregnancy ___ 28
100
I Fevers
310
Digestive Disorders
636
Disorders of Respiratory
System
467
| Disorders of Central Nervous
| System_____ ____________
1220
| Disorders of Circulatory
| System__________________
| 780
| Other Clear Symptoms
]| 594
1 Causes Peculiar to Infancy
| 802
| Senility
]| 5738
1 Total
B-3I
0 3713
%
|r 13.96
]| 0.49
| 1.74
5.40
11.08
8.15
21.26
13.59
10.35
I 13,98
IW-OO
'Vi
Chapter wise Percentage of Deaths
As per ICD
S.No
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14
15.
16.
17
%
Cause of Deaths
Infectious and Parasitic Diseases__________ 10.2
2.7
Neoplasm___________________________
3.2
Nutritional and Metabolic Diseases and
Immunity Disorders___________________
1.3
Diseases of Blood and Blood Forming
Organs_____________________________
0.5
Mental Disorders_____________________
2.8
Disease of Nurvous System and Sense
Organs________________ _____________
24.1
Disease of Circulatory System___________
3.3
Disease of Respiratory System___________
3.4
Disease of Digestive System_____________
2.3
Disease of Genito Urinary System________
1.1
Complication of Pregency, Child Birth and
the Puerperium______________________ _
0.2
Disease of Skin and Subcutaneous Tissue
0.1
Disease of Musculoskeletal System and
Connective Tissue____________________
0.8
Congenital Anomalies_________________
13.4
Certain condition Originating in the Perinatal
Period_____________________________
11.7
Symptoms Signs and Ill defined Conditions
18.9
Injury and Poisoning
100.0
Total
B-32
BAR CHART SHOWING THE
REGISTRATION EFFICIENCY OF
BIRTHS AND DEATHS
INTAMILNADU
1001
90
80
70
□ 1986
H I
I
III
eoJI
rv" ‘i
i
8:-
i
•/
1
i
ft111
n: Ml®
I igfl
.It
I HIH
- ■
si<i
I
-
40J II t 'I
I
30
201004
-
-
is
Iflll
I
t
i
cl
J!
- -1 ■y' ’ i
-.|
1
Mil
•B <
<? > 3
a
|
II IIII ■I ■ ''VIII H
<
k
I
i
i,
fl s
CEMHS
BRMS
B-33
i
Ir
□ 1988
S
□ 1990
H
11992
H
S1994
□
11996
CHART SHOWING PERCENTAGE OF
INSTITUTIONAL DEATHS AND DOMICILLARY DEATHS
IN RURAL AREAS
3.1
I
lag
II
Sig
5?
It
96.9
SINS ilDOM
B-34
i
PERCENTAGE OF MEDICALLY CERTIFIED DEATHS
IN TAMILNADU
-- \a
f ■
45
--
Xrt
35
. ?. • SJ«
p >/? 3^7
4u
Affifi
II
'^/-v
i
<__
30 r
25
tii
p__
f«W
zu-f^
1®
B iigiSS^
w
■fc
i
IL
s| IpfsflfH
p
s
gio!s
Silgi I
illlllilB
p- ■•“•
Sl!l
5iili!
ol
G>
0>
I
J
f If;
II
_
I
s
- -■•
■I
i SI
I
CM
O>
G>
—
co
CD
CD
i
r"
■
■
;I SH■ 1
.__1JI
_______
'
S’
g
s
si
j|
o>
a>
LO
G>
O>
BSTATE DURBAN
B-35
iI
II i3
■f
-Si.s
p .
1
_____
Ort
15
10
■I ?iiii
- . _ '
iiiii ; flsai' ISBIiff
■
3
.-
''''$
1
1
»
R
■f
I
i
CD
O>
G>
TAMIL NADU - 1997 (SCD(R))
13.96
13.98
rr
l-J
10.35J
?>••■.<•/■.• :-XW
fil
J
I ■
’T’/l
.4
11.08
w
r
13.50
1^8
L
b‘1
J;
■ ■]
21.26
A
S Accident
□ Fever
OCough
B Disorder of CS
9 Peculiar to infancy
B-36
H Pregnancy compli
□ Digest Disorders
S Disorder of CNS
H Other clear symptoms
S Senility
CANCER AS A NOTIFIABLE DISEASE - IMPROVEMENTS IN
SYSTEM OF REGISTRATION AND CERTIFICATION OF DEATH
Nandakumar
Bangalore
B-37
Cancer As A Notifiable Disease
Improvements in System of registration and Certification of Death
B-38
What is Notification?
Compulsory for a diagnosing or treating physician/laboratory/medical
institution or any other allied departments to report to a specified
authority about a particular disease soon after its confirmed diagnosis
B-39
Need for such a rule in India?
• Quality of Average Medical Record
• Numerous and varied sources of cancer registration
• High proportion of cases from private sources
• Inadequacies in system of registration of death
• Incomplete certification of cause of death
B-40
Broad Aim/Purpose
• Enhance Cooperation to registry of all medical and allied personnel
• Contribute to Active Registration
• Wherever feasible initiate Passive Registration
• Improve system of registration and certification of causes of death
B-41
Specific Areas where such a rule will help
• Minimize cases where Residential Status is unknown
• Minimize cases where Duration of Stay is unknown
• Increase accuracy of identifying and eliminating duplicates
• Improve mortality data
B-42
Legal Basis of Cancer Registration
• Voluntary
• Compulsory ■ Notifiable
- Legislation
- Administrative Order
Latter is more common.
Could also depend on source of notification.
B-43
Reporting of Cancer is Compulsory
o%
• Africa
• Central and South America
55.6%
• North America
82.6%
• Asia
38.1%
• Europe
47.2%
• Oceania
100%
B-44
<
)
)
)
)
)
)
)
)
)
)
)
)
)
I
Types of Confidential Information
• Core Information ■ Diagnosis of Cancer
• Epidemiologic Data ■ Personal: Sexual/Drinking
• Details of Treatment - Treating physician's name
• Information on Follow-up - House visits for survival information:
When Patient is alive:
Knowledge of cancer
Knowledge of treating physician
B-45
)
1
Medical Records
• Treated as Legal Documents and therefore are Confidential Documents
• No written code or legislation or Government Regulation to treat as
Confidential
• Only Court of Law has the right to Requisition
• Information on Diagnosis and Treatment also furnished to Insurance
Companies
B-47
Medial Law and Ethics in India
H.S. Mehta
V.]. Taroporvala
• CONSENT - Defined as:
Voluntary Agreement
Compliance
Permission
• May be Express or Implied
Express may be Verbal or Written
• Therefore in a Cancer Registry -
Answering or Completing Question
Indicate Implied Consent
B-48
Patient/Relative
• Need and Importance of Information
• Reassured about holding information on Identity and Disease as Strictly
Confidential
• Information used for Scientific purpose only
B-49
Right to refuse Information
• Has the Individual (Patient or Treating Clinician) Really the Right to
Refuse
8-50
I
Hospital/Medical Institution
• Has the Right to utilize Data of Patients Without Identifying
the Individuals(s)
B-51
Follow-up in Mumbai Cohort Study - Preliminary Results
Prakash C Gupta, H.C Mehta
Tata Institute of Fundamental Research, Mumbai, India
The major objective of the Mumbai Cohort Study (MCS) is the
estimation of tobacco attributable mortality in India. A cohort of about
160,000 individuals, recruited during 1991-97, is being followed up
through house-to-house visits. Until Jan. 1999, some 52568 individuals
age 35 years and over were followed up. Among them 0.1% were
unidentified, 14.1% had migrated and 8.4% were reported dead. A total
of 293,368 person years were observed, giving crude annual mortality
rate (per 1000) among men as 20.1, among women. Among women who
had reported using smokeless tobacco, the age adjusted relative risk
(AARR) was 1.35 and among men smokeless tobacco users, it was 1.22.
Among men smokers the age adjusted relative risk was 1.73 and
although the number of women smokers was very small, the AARR
among them was 1.28. These results are similar to those reported in
earlier cohort studies from rural areas. The relative risks were high for
bidi compared to cigarette smoking. A dose response relationship was
also apparent.
B-53
Tobacco Attributable Mortality in South Africa
Freddy Sitas,' Rosana Pacella-Norman,1 Debbie Bradshaw,2 Danuta
Kielkowski,3 Yussuf Saloojee,4 Sulaiman Bah,5 Alan Lopez,6
Derek Yach6 and Richard Peto7.
'National Cancer Registry, South African Institute for Medical
Research, Johannesburg. 2Medical Research Council of South Africa,
National Center for Occupational Health, Department of Health
4National Council Against Smoking Statistics SA 6WHO, Geneva
Epidemiological Studies Unit, University of Oxford, UK
In order to estimate tobacco attributed mortality in South Africa, the
Minister of Health was approached in 1995 and approved the addition of
two questions on the death registration form:
■ Was the deceased a smoker five years ago?
■ Was the informant/next to kin a smoker five years ago?
A smoker was defined as someone who smoked on most days. These
two questions allow for the construction of a case control study and
follow up study of the next of kin.
National and provincial facilitation/implementation teams comprising
key role players were formed in all 9 provinces and the new death
B-54
registration forms were introduced in July 1998. An instruction manual
for filling in the new death registration forms also now available for the
first time.
‘Statistics South Africa’ is responsible for coding of the death
registration details and the production of mortality reports. However,
there is a 2-3 year backlog, so all of the 200-odd regional Home Affairs
offices around the country were asked to send copies of the first 200
death forms (about 40,000 deaths) from 1 October - 1 December 1998,
i.e., 2 months after implementation for fast track coding and evaluation
of the new forms. These forms have now been gathered and are about to
be processed.
Depending upon the level of completeness of the information on tobacco
an initial case control analysis will be carried out to measure tobacco
attributable mortality for selected causes of death.
B-55
Tobacco Prevalence and Death Registration System in Bhopal
Rajesh Dixit
Gandhi Medical College, Bhopal, India
The knowledge of tobacco usage as well as mortality data is particularly
important in Bhopal population as this population underwent exposure to
methyl isocyanate due to chemical as accident in 1984 and hence is
different from other parts of the world.
Tobacco survey was conducted in Bhopal population to estimate the
prevalence of various kinds of tobacco use. The area wise voter list
were obtained, and by random sampling, individuals were selected for
interview using random numbers. Interview were conducted by the
cancer registry staff on pre coded performa by house to house visit.
Total number of 2560 males and 2576 females of the ages 18 years and
above were interviewed. 46.6% of male and 43.1% of females were
found to be users of tobacco, while 12.1% had habit of both smoking
and as chewing. Among females only 0.16% were smokers while 42.5%
were tobacco chewers. Only 0.47% of females had double habit of
smoking and chewing. Among smokers bidi smoking were more
common.
B-56
The mortality statistics has not been used so far in Bhopal to observe the
tobacco related deaths. Total number of 5213 deaths were recorded in
Bhopal urban area during 1997, however, the death registration system
in Bhopal is poor. The details of methodology of death registration
system, its problem and scope for improvement will be discussed. The
possibility of using existing death registration system to carry out
epidemiological studies will also be presented.
B-57
Death Registration and Cause Specific Mortality in a Semi-
urban Adult Population in South Kerala
Babu Mathew1, R. Sankaranarayan2, Binu Jose Jacob1, Gigi
Thomas1, K.T. Shenoy3, M. Krishnan Nair1
’Regional Cancer Center, Trivandrum, India international Agency
for Research on Cancer, Lyon, France
3Medical College, Trivandrum, India
We aim to specify cause-specific mortality in a cohort of 114 601
subjects (48 809 males and 65 792 females) aged 35 years or more,
recruited for a population-based oral cancer screening intervention trial
in the northern suburban area of Trivandrum City, Kerala, India. This
cohort, which consists of residence of 13 panchayaths (a local
administrative structure) in Kazhakuttam and Chirayinkil development
blocks, has been recruited over a period of three years during 1996-
1998. Half of the males are tobacco recruited over a period of three
years during 1996-1998. Half of the males are tobacco smokers, while
less than 1% of females smoke. Tobacco chewing is practiced in the
form of betel quid by a fourth of subjects; a fourth of males report
alcohol drinking habit.
Deaths occurring in the target population are registered in the concerned
panchayats registration office, if it occurred in home or in a hospital
B-58
within the geographical limits of the panchayaths. Usually a relative of
the deceased notifies death to this office. On the other hand, if a resident
of the panchayats expires in a hospital in Trivandrum City or other
nearby towns, such as deaths are notified to the concerned urban death
register. Physicians certify mostly hospital deaths. The urban death
registers
receive a copy of the hospital death certificates.
Though the death register ultimately registers most deaths, the cause
specific mortality data are either not available or unreliable or non
specific. Some of the commonly stated ‘cause of death’ is ‘cardio
respiratory arrest’ or ‘old age’.
For research purposes, we have established an active method to collect
information on antecedent medical history from hospitals/households. A
health worker visits death registers on a regular schedule and relevant
details are abstracted. These are traced back to hospitals/households and
relevant records are reviewed and information is abstracted. A panel of
physicians review all the information thus collected to assign a cause of
death aposteriori.
We present preliminary findings based on 1652 deaths occurring in 1996
in subjects ages 35 years (942 males and 706 females) and above. The
cause of death was either not available or non-specific for 71.3% of the
cases. A panel of physicians reviewed the actively collected information
B-59
and assigned the cause of death aposteriori. This exercise reduced the
proportion of deaths with no-specified causes to 6.3%, coronary heart
disease 922.1%), cancer (15.4%), obstructive pulmonary disease
(14.6%), cerebrovascular disease (13.7%), and diabetes mellitus (6.8%)
were the broad categories of cause of death assigned by the physicians
after review.
Improvement in the completion and quality of death registration in India
would require sincere inputs in legislation, physician education, public
awareness and further investments in death registry organization and
training of staff in coding and analysis.
B-60
Cancer Mortality in Greater Bombay
B.B. Yeole
Bombay Cancer Registry, Indian Cancer Society
Mumbai, India
Mortality statistics have an impressive history as a useful tool for
undertaking epidemiological studies of cancer. The mortality analysis of
various occupational groups has provided the evidence, which led to the
discovery of several chemical carcinogens. Examination of time trends
of the death rates, has in turn lead to the development of new etiologic
hypotheses. Furthermore, international comparison of mortality data has
been productive in outlining new directions for undertaking
epidemiological field studies.
The significant role played by mortality data in epidemiologic studies, in
the past, was largely due to the unavailability of morbidity data, which is
considered more valuable for undertaking epidemiologic investigations.
Gradually, the role of mortality studies has diminished with the
establishment of population based cancer registries in various countries
throughout the world and the availability of adequate morbidity data.
The value of mortality data has also decreased with the increasing use of
epidemiological field studies undertaken to test specific etiologic
hypotheses, developed as a result of analysis of mortality statistics.
B-61
At our registry, mortality data has been obtained from the death records
maintained by the vital statistical division of the Bombay Municipal
Corporation. Copies are made by us of all death certificates which
mention the word ‘cancer’ or ‘tumor’ as being the cause of death.
During the year 1995, a total of 5422 cancer deaths were recorded by the
Bombay Municipal Corporation. Out of these, 860 were known non
residents of greater Bombay. 4112 were residents for over one year and
the remaining 450 did have adequate addresses in Bombay at the time of
death but their duration of stay in the city was not mentioned on the
death certificate. The percentage of total resident cancer deaths tot he
total new cancer registration, was 51.7% during 1995.
The age-adjusted death rates for 1995 at all sites were found to be 67.8
for males and 63.8 for females per 100,000 population. Deaths from
cancers of the lung, top the list in males, followed by deaths from
cancers involving oesophagus, larynx, and stomach. Cancers of the
female genital organs predominate in this sex, the breast being the
leading site followed by the cervix and oesophagus. The age-specific
rates were found to follow the general pattern of increase with age. An
increase for the first time, was seen to occur in the fourth decade, in both
sexes. The age-specific rates for females were found to be generally
lower than males, except in the age-group 20 to 59. This can be
explained by the high mortality rate of female genital cancers that occurs
in the reproductive age groups.
B-62
Problems in Mumbai Feasibility Study
R.Y. Agarkar
Medical Section, Tata Institute of Fundamental Research
Mumabai, India
Tobacco has been identified as one of the most important modifiable
risk factor in leading causes of death. The mortality effects of tobacco is
very well documented in developed countries. Even though there is
enough evidence to demonstrate the higher overall mortality due to
tobacco in India, there remains considerable uncertainty about total
number of deaths tobacco will eventually cause. To monitor the growth
of the epidemic of tobacco deaths in India, a series of long term
epidemiological studies with large population base is required.
WHO recommended that tobacco status on death registrations, along
with cause of death reporting, provide an inexpensive and long-term
method to monitor the tobacco epidemic. Such simple methodology
requires validation in India. A study to evaluate the validity and
practicability of questions on smoking and chewing tobacco on death
certificate in Mumbai, was planned at TIER. The study group
encountered many problems due to lack of cooperation from the
professionals, bureaucratic hurdles, and indifferent attitude due to lack
of awareness of concerned medical and non-medical hospital staff.
B-63
A Feasibility Study in Mumbai - Recording Tobacco Use on
Death Certificates
H.C. Mehta, P.C. Gupta
Tata institute of Fundamental Research, Mumbai, India
It has been suggested that recording of tobacco use status prior to death
on death certificate will provide a reasonably robust and cost-effective
method for monitoring the tobacco epidemic. A feasibility study was
initiated in Mumbai to investigate such an approach. In the first phase,
the feasibility study was limited to hospitals where a large proportion of
deaths (about 70%) occur. It was reported that cultural practices do not
favor interviewing relatives after death of patient and the common
practice is to fill out details in the death certificate from existing medical
records. This study therefore investigated feasibility of abstracting
information about tobacco use from existing medical records of patients
who died in three large government hospitals in South Mumbai. An
investigator sta in the medical record room with a small laptop computer
and looked at the death certificates starting from most recent ones. For
individuals over 35 years, information on age, gender, occupation and
causes of death were abstracted. Corresponding case paper was
identified and scanned carefully to abstract the tobacco use status and if
available details of tobacco use for estimating reliability of the
information. The cause of death was coded as per the ICD10.
B-64
Among total of 1658 death records abstracted from three different
hospitals, 25.9% had some information on tobacco use, 1.3% did not
mention any tobacco information at all and 72.8% were mentioned ad no
tobacco users. Major causes of death were circulatory system disease
(29.3%), respiratory diseases (18.6%), tuberculosis (18.2%), digestive
system disease (8.8%) and neoplasm (5.4%). Classifying tuberculosis.
respiratory and neoplasm by tobacco use and comparing them with rest
of the causes combined, the odds ratio was 1.97. The majority of
tobacco habit recorded was smoking habit (20.5%). Since the
prevalence of smokeless tobacco use is much in the population it
appears that smokeless tobacco use is not adequately recorded in
medical case paper.
B-65
The Death Registration System in Calcutta - A review of Death Cases
with reference to Calcutta PBCR- 1997
Urmi Sen, O.P. Ghosh, S.S. Mondal
Chittaranjan National Cancer Institute, Calcutta, India
The divisional offices of the Vital Statistics department of Calcutta
Municipal Corporation are the major source of mortality for PBCRCalcutta. All information is available in the death registers maintained
for the different crematoriums and burial grounds at these units of
Calcutta Corporation. However, mortality data from some of the
suburban areas of Calcutta covered by the registry area was not available
due to lack of proper documentation. Analysis has been made on the
basis of the available data.
The total number of cases registered for the year 1997 with PBCRCalcutta was 4081. This being the first year of operation the possibility
of some under registration of cancer cases should be taken into account.
The total number of cancer deaths as per the registry record was 448.
The total number of Death Certification Notification Cases (DCN) for
1997 obtained from the VSD units was 2121. Out of these only 277
deaths cases could be matched with the existing registry data. Attempts
have been made to trace the remaining 1844 unmatched DCN cases by
B-66
pre-paid postal inquiries. The response rate observed is poor. An
analysis of the postal replies obtained so far has been given. An overall
analysis of total death cases due to cancer has been made. Attempt has
also been made to analyze the confirmed death cases of 1997 with the
existing registry records for some major cancer sites in both males and
females.
B-67
Death Registration in a Rural Area in India ■ An Evaluation
B.M. Nene
Tate Memorial Center Rural Cancer Project
Barshi, Maharashtra, India
Panchayat office in each village maintains the death records. The
information is registered with the Block development Officer at tehsil
Head quarters and then sent onwards to the Joint Director of Health
Services of the region. New Government Resolution was passed in June
1997. However, its implementation is poor. The drawbacks of the
present system are highlighted and some suggestions are made, which if
implemented will improve the death registration system.
B-68
Appendix C
List of Participants
Dr. R Y Agarkar
Tata Institute of Fundamental Research
National Center of the
Government of India for Nuclear
Science and Mathematics
Homi Bhabha Road, Mumbai- 400 005
India
Tel:
91 22 215 2329/2971/2979
Fax:
91 222152110
Email: agarkar@tiff.res.in
Dr.‘ Samira Asma
Office on Smoking and Health
Centers for Disease Control and Prevention
4770 Buford Highway, MS K-50
Atlanta, GA- 30341
USA
Tel:
770 488 5719
Fax:
770 488 5848
Email: sea5@cdc.gov
Mr. A M Budukh
Ashwini Rural Cancer Research
Relief Society Cancer Hospital
Agalgaon Road, Barshi- 413 401
Sholapur District
Maharashtra
India
Dr. Rajesh Dixit
Population Based Cancer Registry
Department of Pathology
Gandhi Medical College
Bhopal, MP- 462 001
India
Tel:
755 540 768
Fax:
755 541 775
Dr. C K Gajalaxmi
Division of Epidemiology & Cancer Registry
Cancer Institute (WIA) Annex
18 Sardar Patel Road
Chennai- 600 036
India
Tel:
91 44 235 0131/0241
Fax:
91 44 491 2085
Email: gaja_l @hotmail.com
/
Dr. Prakash Gupta
Tata Institute of Fundamental Research
National Center of the
Government of India for Nuclear
Science and Mathematics
Homi Bhabha Road,
Mumbai- 400 005
India
Tel:
91 22 215 2329/2971/2979
Fax:
91 222152110
Email: pcgupta@tiff.res.in
Ms. Binu Jose Jacob
Trivandrum Oral Cancer Screening Study
Sindhu Nivas, Opp Post Office
Mangalapuram, Thonakkal P O
Trivandrum- 695 313
Kerala
Dr. Prabhat Jha
Health, Nutrition, and Population
The World Bank
Room G3-085
1818 H Street, NW
Washington, DC- 20433
Tel:
202-458-7384
Fax:
202-522-3489
Email: pjha@worldbank.org
Dr. Alka S. Karande
Executive Health Officer
Birhanmumbai Municipal Corporation
Municipal Head Office
Annexe Building, 2nd Floor, RN 217
Mahapalika Marg, Fort,
Mumbai- 400 001
India
Tel:
262 0588/0251 ext: 2218
Fax:
262 6437
Dr. B Kuruvilla
Rural Cancer Center
Thiruvananthapuram
Kerala, India
Dr. Alan Lopez
World Health Organization
21 Avenue Appia
Geneva 1211
Switzerland
Tel:
41 22 791 2374
Fax:
41 22 791 0746
Email: lopeza@who.ch
Ms Hemali Mehta
Tata Institute of Fundamental Research
National Center of the
Government of India for Nuclear
Science and Mathematics
Homi Bhabha Road
Mumbai- 400 005
India
Tel:
91 22 215 2329/2971/2979
Fax:
91 222152110
Email: hemali@tifr.res.in
Mr. R G Mitra
Office of the Registrar General
2A, Man Singh Road
New Delhi- 110011
India
Tel:
91 11 338 3761
Fax:
91 11 338 3145
Dr. A Nandakumar
National Cancer Registry
Indian Council of Medical Research
Kidwai Memorial Institute of Oncology
POB 2930, Hosur Road
Bangalore- 560 029
India
91 80 642 649/632 302 •
Tel:
Fax:
91 80 664 801
Email: ank@blr.vsnl.net.in
Dr. B M Nene
Ashwini Rural Cancer Research &
Relief Society Cancer Hospital
Agalgaon Road, Barshi-413 401
Sholapur District, Maharashtra
India
Professor Richard Peto
Medical Statistics and Epidemiology
Harkness Building
University of Oxford
Radcliffe Infirmary
Oxford 0X2 6HE
United Kingdom
Tel:
44 0 1 865 557 241
Fax:
44 0 1 865 558 817
Email: gale.mead@ctsu.ox.ac.uk
Dr. Thiru R Ramodass
Office of the Director of Public Health & Preventive Medicine
259 Anna Salai
Chennai- 600 006
India
Fax:
91 44 434 4158
Dr. Shekar Salkar
Goa Cancer Society
Vaidya Hospital
9 Pestana Road
Panaji, Goa- 403 001
India
Fax:
0832 230 833
Email: sgvaidya@bom2.vsnl.net.in
Dr. R Sankaranarayanan
Unit of Descriptive Epidemiology
International Agency for Research on Cancer (IARC)
150 cours Albert Thomas
F-69372 Lyon Cedex 08
France
Tel:
334 72 73 8599
. Fax:
334 72 73 8575
Email: sankar@iarc.fr
Dr. Urmi Sen
Dept of Epidemiology & Biostatistics
Chittaranjan National Cancer Institute
37, S P Mukerjee Road
Calcutta- 700 026
India
Tel:
91 33 476 5101/5102/5104 Ext-318
Fax:
91 33 475 7606
Email: cncinst@giasclO 1 .vsnl.net.in
Dr. Surendra Shastri
Department of Preventive Oncology
Tata Memorial Hospital
Dr Ernest Borges Road
Parel, Mumbai 400 012
India
Tel:
91 22416 1413
Fax:
91 22 416 4440
Email: medimail@tmc.emet.in
Dr. K T Shenoy
Director
Epidemiology Program
Medical College
Thiruvanathapuram, Kerala
India
Email: ktshenoy@md3.vsnl.net.in
Dr. Freddy Sitas
National Cancer Registry
South African Institute for Medical Research
Johannesburg
South Africa
Tel:
011 489 9171
Fax:
011 489 9152
Email: freddys@mail.saimr.wits.ac.za
Dr. Sharad Vaidya
Goa Cancer Society
Vaidya Hospital,
Gov. Pestana Road, Panaji
Goa- 403 001
India
Tel:
91 83 222 3526/225648
Fax:
91 83 223 0833/224127
Email: sgvaidya@bom2.vsnl.net.in
Dr. M Vijayanunni
Registrar General & Census Commissioner, India
2A, Man Singh RoadNew Delhi-110011
India
Tel:
91 11 338 3761
Fax:
91 11 338 3145
Email: rgindia@hub.nic. in
Dr. B B Yeole
Deputy Director
Bombay Cancer Registry
Indian Cancer Society
74 Jerabai Wadia Road
Parel, Mumbai- 400 012
India
Tel:
91 22 412 2351
Fax:
91 22 416 1447
/
- Media
5713.pdf
Position: 1773 (5 views)