You are here

Sources of Health Information

Lecture on 'sources of health information - part 1:
Lecture on 'sources of health information - part 2:
1. Census
2. Registration of Vital Events
3. Sample Registration System (SRS)
4. Notification of Diseases
5. Hospital Records
6. Disease Registers
7. Record Linkage
8. Epidemiological Surveillance
9. Other Health Service Records
10. Environmental Health Data
11. Health Manpower Statistics
12. Population Surveys
13. Other routine statistics related to health
14. Non-quantifiable information

1. Census
Census is taken in most countries of the world at regular intervals
Definition of Census (UN): ‘The total process of collecting, compiling and publishing demographic, economic and social data pertaining at a specified time or times, to ALL persons in a country or delimited territory’.
It is massive undertaking to contact every member of the population in a given time and collect a variety of information
It needs
• Massive preparation and
• Several years to analyze the results after census is taken
• This is the main problem of census as a data source i.e. that the results are not available quickly
Census in India
The first regular census was undertaken in 1881
Census is undertaken after every 10 years
Usually, it is conducted at the end of the first quarter of the first year in each decade.
The last census was held in March 2011
The Census Act of 1948 is the legal basis of the censes
The Census Commissioner of India is the highest officer responsible for guiding and operating the census
Although the primary function of census was to collect demographic information i.e.
• Total count of population
• Break-up of the population into subgroups such as age and sex distribution
But over the years, much more information is collected along with this
• Social and economic characteristics of the households
• Their living conditions e.g. type of fuel used, type of house etc.
• Occupation
• Income Etc.
Census data provides a frame of reference and baseline information for administration, planning, action and research in medicine and social sciences
Population census provides the data used to compute health, demographic and socio economic indicators

2. Registration of Vital Events
Unlike Census, ‘’Registration of vital events’’ maintains a continuous record of demographic trends.
Definition (UN): ‘legal registration, statistical recording and reporting of the occurrence of and the collection, compilation, presentation, analysis and distribution of statistics pertaining to vital events’
Vital events include:
a. Live births
b. Deaths
c. Fetal deaths
d. Marriage
e. Divorce
f. Adoptions
g. Legitimations
h. Recognitions
i. Annulments and
j. Legal separations
In India, the civil registration system mandates registration of births, deaths and still births.
Vital events registration in India
In 1873, GOI passed the ‘’Births, Deaths and Marriages Registration Act’’. This provided only for voluntary registration.
Subsequently, individual States like TN, Karnataka and Assam passed their own Acts
The Vital event registration in India is unreliable. It lacks:
• Accuracy
• Timeliness
• Completeness and
• Coverage
Also there is lack of uniformity in the collection, compilation and transmission of data, which is different for rural and urban areas and multiple registration agencies (e.g. health agency, panchayat agency, police agency and revenue agency)
In 1960, Vital statistics was transferred to the office of the Registrar General, India from the Director General of Health services. Thus, population census and vital statistics, including civil registration, came under one office
a. The Central Births and Deaths Registration Act, 1969
This was promulgated with an aim to improve the civil registration system
The Act is in force since 1 April, 1970
The Act provides for compulsory registration of births and deaths throughout the country and
Compilation of vital statistics in the States so as to ensure uniformity and comparability of data
The implementation of the Act required adoption of rules for which also, model guidelines have been provided
The Act also fixes the responsibility for reporting births and deaths
While the public are to report events occurring in their households, the heads of hospitals, nursing homes, hotels, jails or dhamashalas are t report events occurring in such institutions to the concerning Registrar
The time limit for registering the event of births and that of deaths is 21 days uniformly all over India.
In case of default, a late fee can be imposed
The Act makes the beginning of a new era in the history of vital statistics registration in India
The Chief Registrars of Births & Deaths guide in States and Union Territories, whereas the central authority is the Registrar General of India.
b. Lay reporting
Because of slow progress in the development of a comprehensive vital registration system, some countries have attempted to employ first line health workers e.g. village health guides, to record births and deaths in the community.
Indeed, one of the important functions of a primary health worker is to collect and record data on vital events and other health information in his or her community
Definition of Lay Reporting: ‘the collection of information, its use, and its transmission to other levels of the health system by non-professional health workers’
Many countries have adopted this approach to obtain information on vital statistics
‘Demographic Survey’ is another approach but not a satisfactory one. It is best regarded as a temporary substitute for some kind of vital events registration system

3. Sample Registration System (SRS)
The civil registration system in India has a lot to be improved and would require a huge effort and time. A need for having an alternate source of such information was felt. The Government of India, in the late 1960s, initiated the Sample Registration System (SRS).
SRS aims to provide reliable estimates of birth and death rates for the States and also at All India level. Following features of SRS ensure the completeness of vital events reporting:
1. A representative sample of the population is covered and NOT the WHOLE population
2. It is based on Dual – record system:
a. First a baseline survey of sampled units is done
b. This is followed by continuous enumeration of vital events of the areas by an enumerator (who is a volunteer from the community)
c. Independent retrospective 6-monthly surveys are done for recording births and deaths which occurred in the preceding 6 months and the number is matched with the one reported by the enumerator
Not only the numbers of vital events should match, it is also verified if the births and deaths reported by the enumerator are the same ones which the survey has found out. This is called as ‘Matching of events’ by the observer
At present, the Sample Registration System (SRS) provides reliable annual data on fertility and mortality at the state and national levels for rural and urban areas separately. In this survey, the sample units, villages in rural areas and urban blocks in urban areas are replaced once in ten years

Various birth and death rates calculated for India and State levels are based on SRS data only;
Birth rate, death rate, IMR, Total Fertility Rate, sex ratio at birth and in 0-4 yr age group, Child Mortality rate, institutional deliveries etc.

4. Notification of Diseases
Disease notification is a practice of reporting the occurrence of a specific disease or health-related condition to the appropriate and designated authority.
A notifiable disease is any disease that is required by law to be reported to government authorities.
Effective notification allows the authorities to monitor the disease, and provides early warning of possible outbreaks
A notifiable disease is one for which regular and timely information regarding individual cases is considered necessary for the prevention and control of the disease.
Reasons for Declaring a Disease as Notifiable may be:
a. It is of interest to national or international regulations or control programs
b. Its National/ State/ District incidence
c. Its severity (potential for rapid mortality)
d. Its communicability/ Its potential to cause outbreaks
e. Significant risk of international spread
f. The socio-economic costs of its cases
g. Its preventability
h. Evidence that its pattern is changing
i. In other words, diseases which are considered to be serious menaces to public health are included in the list of notifiable diseases
5. Hence the list of notifiable diseases vary from
a. Country to country and
b. Within the same country
i. Between the States and
ii. Between urban and rural areas
Limitations of notification system:
• Notification data can include only a small portion of the total illness in the community
• The system suffers from a good deal of under reporting
– Atypical and subclinical cases escape notification e.g.
• Rubella
• Non paralytic polio etc.
– Accuracy of notification is dependent upon availability of facilities for diagnosis
– Often the disease is not reported because:
• Healthcare worker is not aware
• Intention to report a lower incidence to prove effective administration etc.
• In spite of these limitations,
– notification provides valuable information about fluctuation in disease frequency
– It also provides early warning about new occurrences or outbreaks of disease
– It has become a valuable source of morbidity data i.e. the incidence and distribution of certain specified diseases which are notifiable.
Under IHR following diseases were notifiable to WHO:
1. Cholera
2. Yellow fever and
3. Plague
In 2005 revision, following diseases were added:
1. smallpox,
2. poliomyelitis due to wild-type poliovirus,
3. SARS and
4. cases of human influenza caused by a new subtype
. Diseases under International surveillance are
• Louse borne typhus
• Relapsing fever
• Polio
• Influenza
• Malaria
• Rabies and
• Salmonellosis
Now some non-communicable diseases are in the ambit of notification e.g
 Cancer
 Congenital malformations
 Mental illness
 Stroke and
 Handicapped persons

6. Hospital Records
a. The cases constitute only the tip of the iceberg – i.e. they provide information on only those patients who seek medical care but not on a representative sample of the population. Mild cases may not attend hospitals; subclinical cases are always missed
b. The admission policy may vary from hospital to hospital; therefore hospital statistics tend to be highly selective
c. Population served by a hospital cannot be defined. Hence there is no the denominator. hospital statistics are a poor guide to the estimation of disease frequency in a community
 A study of hospital data provides information on the following aspects:
o Geographic sources of patients
o Age and sex distribution of different diseases and duration of hospital stay
o Distribution of diagnosis
o Association of between different diseases
o The period between disease and hospital admission
o The distribution of patients according to different social can biological characteristics and
o The cost of hospital care
 Hospital information is useful in planning of health services. Indices needed for this purpose are:
o Bed occupancy rates
o Duration of stay
o Cost effectiveness of treatment policies is useful in monitoring the use of hospital facilities
 The Medical Record Department (MRD) should be established in each hospital to enable this

7. Disease Registers
• A registry is basically a list of all the patients in the defined population who have a particular condition.
• It is different from ‘notification’ where the case is reported and is counted once.
• A register requires that a permanent record be established
– that the cases be followed up and
– that basis statistical tabulation be prepared both on frequency and on survival
– Hence mostly for chronic conditions
– In addition the patients on a register are frequently be the subjects of special studies
• It helps keep track of
– when they were seen last,
– what tests and treatments they have had and
– Survival etc.
• Morbidity registers are a valuable source of information such as the
– duration of illness, case fatality and survival
– Natural history of disease especially chronic diseases
• Limitations:
– Absence of a defined population base, hence incidence and prevalence rates can’t be calculated
• Population based registry can provide this if complete reporting can be ensured
– Morbidity registers exist only for certain diseases and conditions such as stroke, myocardial infarction, cancer, blindness, congenital defects and congenital rubella etc.

There are two types of registries:
Hospital based:
It involves recording of information on the patients seen in a particular hospital. The primary purpose of hospital based registries is to contribute to patient care by providing readily accessible information on the patients, the treatment received and its results. The data is also used for clinical research and for epidemiological purposes.
The objectives of hospital based registry:
• Assess patient care
• Participate in clinical research to evaluate therapy
• Provide an idea of the patterns of cases
• Contribute to follow up of the patients
• Epidemiological research through case control studies
• Show time trends in the stage of diagnosis
• Help study the quality of care for the patients in the hospitals
Population-Based Registry:
A population-based disease registry contains and tracks records for people diagnosed with a specific type of disease who reside within a defined geographic region (i.e., a community, city, or state-wide)
The major concern of Population Based Registries is to calculate the incidence rates.
Population Based Registry systematically collects information on all reportable cases occurring in a geographically defined population from multiple sources.
• The comparison and interpretation of population based incidence data support population-based actions aimed at reducing the burden in the community.
• The systematic ascertainment of incidence from multiple sources can provide an unbiased profile of the burden in the population and how it is changing over time.
• These registries can provide data for planning and evaluating control programmes.

Disease Registry in India:
National Centre for Disease Informatics and Research, ICMR operates following disease registries in the country:
• National Cancer Registry Programme
• Registry of people with diabetes in India with young age at the onset
• National stroke registry program

8. Record Linkage
• Record linkage means bringing together information that relates to the same individual or family, from different data sources. In this way it is possible to construct chronological sequences of health events for individuals
• The records may originate in different time or places.
• Medical record linkage implies the gathering and maintenance of one file for each individual in a population, with records relating to his/her health.
• The events commonly recorded are
o birth, marriage, death,
o hospital admission and discharge
o sickness absence from work,
o prophylactic procedures
o use of social services, etc.
• Record linkage is a particularly suitable method of studying association between diseases; these associations may have etiological significance.
• The challenges with record linkage are
o the volume of data that can accumulate
o Need for technology
o Privacy issues
o Involvement of different agencies for different records
o High chances of incomplete retrieval of information
• Therefore in practice record linkage has been applied only on a limited scale e.g. twin studies, measurement of morbidity, chronic disease epidemiology and family and genetic studies.
• Due to these constraints, record linkage is beyond the reach of many developing countries.

9. Epidemiological Surveillance
Surveillance systems are set up for select diseases under the respective control/eradication program as a procedural matter
The purpose of this surveillance is to keep an eye on the incidence, prevalence and changing pattern of the particular disease so as to adjust the control measures accordingly.
• AFP surveillance data under polio eradication
• Leprosy cases surveillance under NLEP
• No. of children give JE vaccine etc.
These programs have yielded considerable morbidity and mortality data foe the specific diseases.
10. Other Health Service Records
• Apart from hospitals, records are also maintained in:
o Primary health centers
o Polyclinics
o MCH centers
o School health records
o Anganwadis
o Special clinics like diabetic clinic, autism clinic, cardiac clinics
• Records in Anganwadis can provide information about:
o birth weights,
o weight, height, arm circumference,
o Immunization etc.
• However, this data relates only to a certain segment of the general population.

11. Environmental Health Data
• Sometimes information on environment may be needed for studying its effects on health
• Data is then required on various aspects of
o air, water and noise pollution;
o harmful food additives;
o industrial toxicants,
o inadequate waste disposal and other aspects
• Collection of environmental data remains a major problem for future research

12. Health Manpower Statistics
• For planning, administration and evaluation of health services, information on health manpower is important.
• Such information includes the number of
o physicians (by age, sex, specialty and place of work),
o dentists (classified in the same way),
o pharmacists,
o veterinarian,
o hospital nurses,
o medical technicians, etc.
• Their records are maintained by the State medical/dental/nursing councils and the Directorates of Medical Education.
• The census also provides information about occupation.
• The Institute of Applied Manpower Research attempts estimates of manpower, taking into account different sources of data, mortality and out-turn of qualified persons from the different institutions.
• The Planning Commission also gives estimates of active doctors for different States.
• Regarding medical education, statistics of number admitted, numbers qualified, are given every year in ‘health Information of India’, published by the GOI, in the MoHFW

13. Population Surveys
The health information from the sources discussed before do not provide information for all the diseases prevalent in and the health status of the community
The statistics regarding diseases prevalent in the community and the health status of the community are of importance for public health administration. Also for research in any one disease or health related state one may not find the required information in these records.
This calls for conduct of population surveys for the specific health problem/s
The term ‘health survey’ is used for surveys for studying any aspect of health – morbidity, mortality, nutritional status etc.
Survey for studying the diseases in a community is known as ‘morbidity survey’
Health surveys are of the following types:
1. Survey for studying the diseases prevalent and their distribution in the population and the health status of the individuals
Also known as ‘Community Diagnosis’
This is needed for planning and developing the health services for this population
2. Surveys for studying factors affecting health or ‘risk factors’
3. Surveys for assessing administration of health services e.g. utilization rates, out-of-pocket expenditure on health, unmet health needs etc.
Sampling techniques are available so that estimates at any level of precision desired can be achieved. Hence the precision level may be set in accordance with the constraints of available resources and the appropriate sample size may be surveyed.
Health surveys may be
• cross sectional or longitudinal
• descriptive or analytical (or both)
Health surveys on regular basis are operational in only a few countries
• Japan since 1953,
• USA since 1957 and
• UK since 1971
Data collection in surveys can be carried out in one or more of the following methods:
• Health examination
• Health (face to face) interview
• Health records
• Mailed questionnaire
Each method has its own advantages and disadvantages and may be suitable in varying situations.
1. Health examination of the members is suitable when morbidity survey is carried out
• The survey team includes doctors, technicians, interviewers etc.
• Hence it is expensive and not suitable for an extensive scale
• It carries the issue of providing treatment to those found ill.

2. The health interview (face to face) survey is an suitable for assessing subjective phenomena such as
• Perceived morbidity, disability and impairment;
• Economic loss due to illness,
• Expenditure incurred on medical care;
• Opinions, beliefs and attitudes;
• Some behavioral characteristics etc.

In India, the National Sample Survey Organization (NSSO) has been active in conducting interview surveys; these surveys have yielded nationwide data in general morbidity, family planning and vital events, but the morbidity data is not reliable because of the limitation of the interview method.
Better still is to combine interview with health examination surveys and/or laboratory measurements.
An alternative method of measuring subjective phenomena is the self-administered Questionnaire

3. Health records survey involves collection of data from health service records. This is evidently the cheapest method of collecting data. This method has several disadvantages:
• The estimates obtained from the records are not population based
• Reliability of data is open to question, and
• Lack of uniform procedures and standardization in recording of data.

4. The use of questionnaire is simpler and cheaper and they may be sent, for example, by mail to persons sampled from a given target population.
• A certain level of education and skill is expected from the respondents
• There is usually a high rate of non-response.
Except when the purpose of survey is to collect information from a special group (e.g. school children or a particular occupational group), the household is the most common sampling unit.
The sample size is calculated based on the measurement being taken and the degree of precision desired.
Data must be gathered under standardized conditions with quality control.
The collection of data should be restricted to those items for which there is a clearly defined use.
The fact that data might be of interest or use to someone, someday, somewhere is not a valid reason for collecting them.
The collected data needs to be transformed into information by summarizing them and adjusting them for variation in the age and sex composition of the population (so that comparisons over time and place are possible)
• Regular health surveys in India:
– Registrar General of India (RGI) – Annual Health Survey
– National Sample Survey Office’s (NSSO)
– MoHFW – NFHS and DLHS (nodal agency - International Institute of Population Sciences)

14. Other routine statistics related to health
• Sources of data other than health are also required by the epidemiologists to study health, disease, risk factors etc.
• The following list, which is not comprehensive, merely serves to give examples of such sources of data
o Demographic: In addition to routine census data, statistics on such other demographic phenomena as population density, movement and educational level
o Economic: consumption of tobacco, dietary fats and domestic coal; sales of drugs and remedies; information concerning per capita income; employment and unemployment data.
o Social security schemes: medical insurance schemes make it possible to study the occurrence of illnesses in the insured population.
o Other data that may be needed comprise sickness absence, sickness and disability benefit rates.

15. Non-quantifiable information
• Health planners and decision makers require a lot of non-quantifiable information too e.g.
o Information on health policies,
o Health legislation,
o Public attitudes,
o Program procedures and technology.
• In other words, a health information system has multi-disciplinary inputs.

• Ministry of Statistics and Programme Implementation, GOI, 2010; Available at: downloaded on 3 July 2016
• WHO, 2009, FAQ in IHR; available at: accessed on 6 July 2016
• Park’s Textbook of Preventive and Social Medicine
• National Inst of Health: List of Registries: available at: Accessed on 9 July 2016
• ICMR: National Center for Disease Informatics and Research. Available on: Accessed on 9 July 2016
• Aerial et. al; 2014 Scientific paper: Record Linkage in Health Data: a simulation study Available at: accessed on 7th July 2016

Lecture on 'sources of health information - part 1:
Lecture on 'sources of health information - part 2: