Choose Report Type
Publication Date
Report Upload
Download
(2.5 MB)
vertical
Health & Family Welfare
PDF Text
0
HEALTH SURVEYS IN INDIA:
REVIEW AND RECOMMENDATIONS
Final Report
(Strictly confidential)
ACKNOWLEDGEMENT
This study was carried out with the financial support of NITI Aayog,
Government of India, and conducted by the Indian Council for Research on
International Economic Relations (ICRIER), New Delhi
INDIAN COUNCIL FOR RESEARCH ON INTERNATIONAL ECONOMIC RELATIONS (ICRIER)
RESEARCH PROJECT TITLE
Health Surveys & Research Studies – MIS component of the
Umbrella Scheme for Family Welfare & Other Health Interventions
(NITI Aayog’s Letter of Approval No. 0-15012/3/19-Research, 14 May 2019)
Final Report
(June 16, 2020)
Ali Mehdi
Principal Investigator
Senior Fellow, ICRIER
Disclaimer
The Indian Council for Research on International Economic Relations (ICRIER) has received the
financial assistance under the Research Scheme of NITI Aayog (RSNA-2018) to prepare this
report. While due care has been exercised to prepare the report using the data from various
sources, NITI Aayog does not confirm the authenticity of data and accuracy of the methodology
to prepare the report. NITI Aayog shall not be held responsible for findings or opinions expressed
in the document. This responsibility completely rests with ICRIER (contributors).
Acknowledgments
I am extremely thankful to the NITI Aayog – particularly Dr Rajiv Kumar, Vice Chairman; Dr Vinod
K Paul, Member; Mr Alok Kumar, Adviser (Health & Nutrition) and Ms Nina Badgaiyan, Consultant
– for inviting us to conduct this very important and interesting study.
I am also thankful to all those who kindly agreed to meet us during our field interactions (Annexure
A) and shared valuable insights with us. I would like to mention Ms Nivedita Gupta, Chief Director
(Statistics), Statistics Division, Ministry of Health & Family Welfare, Government of India particularly
for her time and insights on several occasions.
At ICRIER, I would like to thank Dr Rajat Kathuria, Director and Chief Executive and all those who
provided research assistance for the conduct and completion of the study – Priyanka Tomar, Divya
Chaudhry and Nilanjana Gupta, former Research Associates, ICRIER Health Policy Initiative (HPI);
Rituparna Sengupta, External Consultant, HPI; Hrishita Sharma, Aishwarya Choudhary and Vrinda
Sharma – HPI interns, pursuing their BA Economics at different colleges of the University of Delhi.
I would like to single out Priyanka for my special gratitude for her extraordinary research support.
This work would not have been possible without the overwhelming support of all those mentioned
above. Any errors that remain are entirely mine.
Ali Mehdi
Principal Investigator
Senior Fellow, ICRIER
1
1
Kindly share any comments and suggestions that you may have at – amehdi@icrier.res.in; mehdi2604@gmail.com.
“We recognise that a reliable evidence base is essential for
informing the planning, implementation and monitoring of health
programmes and of systems strengthening. Data are also critical
to facilitate timely responses to health risks. Yet health data often
remains unavailable, inaccessible, of poor quality, fragmented, not
well protected and are often not used appropriately. We therefore
encourage support for data system strengthening to improve
health data availability, disaggregation, quality, systems
interoperability, data transparency, sharing and the protection of
personal data, and the use of data on a national, regional and
international level, while respecting privacy and other human
rights with regard to all collected health data.”
Berlin Declaration of G20 Health Ministers (2017)
Table of contents
Abstract .......................................................................................................................... 1
Key terms ....................................................................................................................... 2
Introduction ................................................................................................................... 6
Study objectives ........................................................................................................................ 8
Study methodology .................................................................................................................. 8
Organization of the report ..................................................................................................... 10
1. Conceptual framework ............................................................................................ 11
The notion and determinants of health ................................................................................ 11
The dichotomy of complex vision and RCH-oriented action in India .............................. 14
What data should the government collect? ........................................................................ 15
The policy context of health systems and health information systems thinking in India
................................................................................................................................................... 17
WHO’s health systems approach to monitoring and evaluation ...................................... 18
Health surveys – quantitative or mixed methods? ............................................................. 22
Recommendations .................................................................................................................. 24
SECTION 1:
HEALTH SURVEYS AND THE SCOPE OF NFHS AT THE NATIONAL LEVEL
2. Health surveys in India ............................................................................................ 30
Heath Survey and Development / Bhore Committee (1943-46) ...................................... 30
National Planning Committee (NPC) – National Health Sub-Committee Report (1948)
................................................................................................................................................... 30
National Sample Survey (NSS, 1952-) ................................................................................ 32
Health Survey and Planning Committee (1959-61) ........................................................... 34
Model Registration System (MRS, 1965-81), Survey of Causes of Death (SCD), Rural
(1981-98).................................................................................................................................. 35
Sample Registration System (SRS, 1971 –) ....................................................................... 35
SRS CoD (VA) surveys .......................................................................................................... 38
National Family Health Survey (NFHS, 1992 –) ................................................................. 46
District Level Household and Facility Survey (DLHS, 1998-2013) .................................. 62
Global Youth Tobacco Survey (GYTS, 2000-) ................................................................... 63
National Behavioural Surveillance Survey (NBSS, 2001-06) ........................................... 64
Global Adult Tobacco Survey (GATS, 2009-) .................................................................... 64
Annual Health Survey (AHS, 2010-13) ................................................................................ 64
National Anti-TB Drug Resistance Survey (NATDRS, 2014-) .......................................... 65
National Mental Health Survey (NMHS, 2015-16) ............................................................. 66
Comprehensive National Nutrition Survey (CNNS, 2016) ................................................ 67
Longitudinal Ageing Study in India (LASI, 2016-) .............................................................. 67
National NCD Monitoring Survey (NNMS, 2017-) .............................................................. 68
Recommendations .................................................................................................................. 79
3. Scope of NFHS vis-à-vis MoHFW’s policies ........................................................... 82
4. Scope of NFHS vis-à-vis MoHFW’s schemes ......................................................... 94
5. Scope of NFHS vis-à-vis health-related SDGs ..................................................... 104
Global context ....................................................................................................................... 104
Indian context ........................................................................................................................ 106
Recommendations ................................................................................................................ 108
6. Data for Health Technology Assessment (HTA) .................................................. 110
SECTION 2:
PERSPECTIVES FROM THE FIELD
7. Overview of selected states .................................................................................. 112
8. Perspectives on the NFHS and other surveys ..................................................... 116
Objective (PRCs) .................................................................................................................. 116
Thematic scope ..................................................................................................................... 117
Geographical scope ............................................................................................................. 119
Periodicity .............................................................................................................................. 121
Sample design ....................................................................................................................... 122
Questionnaires ...................................................................................................................... 123
Biomarkers ............................................................................................................................. 124
Field staff ................................................................................................................................ 124
Field work ............................................................................................................................... 125
Data quality ............................................................................................................................ 126
Stakeholder involvement – National ................................................................................... 127
Stakeholder involvement – International ........................................................................... 128
Data dissemination ............................................................................................................... 128
Data analysis .......................................................................................................................... 128
Data use ................................................................................................................................. 129
Other health surveys ............................................................................................................ 132
Miscellaneous ........................................................................................................................ 134
SECTION 3:
HEALTH SURVEYS IN UNITED STATES, CANADA AND UNITED KINGDOM
9. United States .......................................................................................................... 135
National Vital Statistics System (NVSS) ................................................................... 137
National Health Interview Survey (NHIS) .......................................................................... 142
National Health and Nutrition Examination Survey (NHANES) ...................................... 146
National Survey of Family Growth (NSFG) ....................................................................... 151
Lessons for India ...................................................................................................... 155
10. Canada ................................................................................................................. 157
StatCan’s Health Statistics Program (HSP) ...................................................................... 158
Canadian Community Health Survey (CCHS) – Annual Component ........................... 165
Canadian Health Measures Survey (CHMS) .................................................................... 170
Canadian Health Survey on Children and Youth (CHSCY) ............................................ 177
Survey on Maternal Health (SMH) ..................................................................................... 178
Canadian Health Survey on Seniors (CHSS) ................................................................... 179
Canadian Survey on Disability (CSD) ................................................................................ 180
Lessons for India ................................................................................................................... 184
11. United Kingdom ................................................................................................... 186
Health statistics in the UK .................................................................................................... 186
Health Survey for England (HSE) ....................................................................................... 192
Health Survey Northern Ireland (HSNI) ............................................................................. 196
Scottish Health Survey (SHeS) ........................................................................................... 197
National Survey for Wales (NSW) ...................................................................................... 200
Lessons for India ................................................................................................................... 203
12. Conclusions ......................................................................................................... 206
Proposed surveys ................................................................................................................. 209
Annexure A – List of interviewees ............................................................................ 221 1
Abstract
As part of a country’s health information system (HIS), health surveys cater to a variety of national
and international data needs on a periodic basis. In the context of weak administrative health data
systems, the independent, population-based estimates provided by health surveys become all the
more significant. Nevertheless, it is also important to periodically review these surveys – themselves
instruments for reviewing a country’s health policies and programs – to ensure that they continue
to cater to a country’s evolving health data requirements in a smart, efficient and coordinated way.
This study undertakes a review of major health surveys in India, with a special focus on the National
Family Health Survey (NFHS), based on extensive desk research – covering major ongoing health
surveys in 3 countries as well (US, Canada, UK) – and key stakeholder / expert interactions in New
Delhi as well as 6 states, covering various geographical regions of the country. Based on this desk
and field research, it offers a set of recommendations for India’s health survey strategy as well as
a thematic and methodological framework for 6 health surveys it proposes as part of the strategy
– the India Health Survey (HIS), the India Health Measures Survey (IHMS), the Maternal and Child
Health Survey (MCHS), the ongoing Sample Registration System (SRS) and Cause of Death Survey
(CDS), and a COVID-19 Impact Survey (CIS). Kindly refer to the table in the conclusions for details.
The report is organized thus. Chapter 1 sets the conceptual context for the discussions on health
surveys that follow. It briefly refers to the notion and determinants of health, WHO’s health systems
approach to monitoring and evaluation, the policy context of health systems and health information
systems thinking in India, and whether health surveys should pursue a purely quantitative or mixed
methodology. The chapter is followed by 3 sections.
Section 1 has 5 chapters. Chapter 2 provides an overview of major health surveys in India since
independence, with a special focus on the NFHS and SRS, and discusses their comparative scope.
Chapters 3 to 6 assess the scope of NFHS vis-à-vis health policies, programs, health-related SDGs
and the emerging area of health technology assessment (HTA) respectively at the national level.
Section 2 shares the perspectives of respondents vis-à-vis NFHS as well as other health surveys.
Section 3 provides an overview of major health surveys in selected countries (US, Canada and UK)
and draws lessons for India. Several chapters end with recommendations, marked with ®. The key
recommendations, emerging from our desk and field research, are listed in the concluding chapter.
The chapter also includes a table which characterizes the potential respective features and themes
of the 6 national health surveys that we propose.
2
Key terms
Term Definition / description Source
Health
Health is a state of complete physical, mental and
social well-being and not merely the absence of
disease or infirmity.
WHO 1946 / 2020
Epidemiology
The study of the distribution and determinants of
health-related states or events (including disease).
WHO
2
Verbal autopsy
(VA)
An established health surveillance method that
provides information on levels and causes of death
in populations where medical death certification is
weak or absent.
Thomas,
D’Ambruoso and
Balabanova 2018:
1
Health system
All the organizations, institutions, resources and
people whose primary purpose is to improve health.
WHO 2010: vi
Health
information
system (HIS)
A well-functioning health information system is one
that ensures the production, analysis, dissemination
and use of reliable and timely health information by
decision-makers at different levels of the health
system, both on a regular basis and in emergencies.
It involves three domains of health information: on
health determinants; on health systems
performance; and on health status. To achieve this,
a health information system must:
• Generate population and facility based data: from
censuses, household surveys, civil registration data,
public health surveillance, medical records, data on
health services and health system resources (e.g.
human resources, health infrastructure and
financing);
WHO 2007: 18-19
2
https://www.who.int/topics/epidemiology/en/ (15/10/2019, 12:40 hours).
3
Term Definition / description Source
• Have the capacity to detect, investigate,
communicate and contain events that threaten
public health security at the place they occur, and
as soon as they occur;
• Have the capacity to synthesize information and
promote the availability and application of this
knowledge.
Health
management
information
system (HMIS)
An information system specially designed to assist
in the management and planning of health
programmes, as opposed to delivery of care.
WHO 2004: 3
Administrative
data
Data from the records maintained by agencies,
institutions, commercial entities and governments,
where the records are used for administrative
purposes or for providing services. Examples
include hospital and other health facility data, claims
data, occupational injuries data, and police data.
Global Health Data
Exchange (GHDx)
3
Civil registration
and vital statistics
(CRVS)
A well-functioning CRVS system registers all births
and deaths, issues birth and death certificates, and
compiles and disseminates vital statistics, including
cause of death information. It may also record
marriages and divorces.
WHO
4
Public health
surveillance
The continuous, systematic collection, analysis and
interpretation of health-related data needed for the
planning, implementation, and evaluation of public
health practice. Such surveillance can:
▪ serve as an early warning system for impending
public health emergencies;
WHO
5
3
http://ghdx.healthdata.org/about-ghdx/data-type-definitions (8/3/2020, 1:01 hours).
4
https://www.who.int/healthinfo/civil_registration/en/ (8/3/2020, 1:01 hours).
5
https://www.who.int/topics/public_health_surveillance/en/ (9/10/2019, 10:55 hours).
4
Term Definition / description Source
▪ document the impact of an intervention, or track
progress towards specified goals; and
▪ monitor and clarify the epidemiology of health
problems, to allow priorities to be set and to
inform public health policy and strategies.
Surveillance
system
The critical components in the definition of a
surveillance system include the ongoing collection,
analysis, and use of health data. Demographic or
health information systems (for example, registration
of births and deaths, routine abstraction of hospital
records, health surveys in a population) that are not
linked to specific prevention and control programs,
do not constitute a surveillance system.
WHO 2003: 3
Surveillance
sources
Surveys (population-based data), disease registries
(incidence and case fatality), hospital activity data
(morbidity and health service use indicators),
administrative data (births, deaths, insurance claims,
medication use, health systems performance,
hospital audits), aggregate consumption data (per
capita consumption) and economic activity data
(economic indicators).
WHO 2003: 4
Interoperability
The ability to access and process data from multiple
sources without losing meaning and then integrate
that data for mapping, visualization, and other forms
of representation and analysis. Interoperability
enables people to find, explore, and understand the
structure and content of data sets. In essence, it is
the ability to ‘join-up’ data from different sources to
help create more holistic and contextual information
for simpler, and sometimes automated analysis,
Collaborative on
SDG Data
Interoperability,
Global Partnership
for Sustainable
Development
Data
6
6
http://www.data4sdgs.org/initiatives/interoperability-data-collaborative (17/10/2019, 11:29 hours).
5
Term Definition / description Source
better decision-making, and accountability
purposes.
Health technology
The application of organized knowledge and skills in
the form of medicines, medical devices, vaccines,
procedures and systems developed to solve a
health problem and improve quality of life.
WHO
7
Health technology
assessment
(HTA)
The systematic evaluation of properties, effects,
and/or impacts of health technology. It is a
multidisciplinary process to evaluate the social,
economic, organizational and ethical issues of a
health intervention or health technology. The main
purpose of conducting an assessment is to inform a
policy decision making.
Bibliography
Thomas, Lisa-Marie, Lucia D’Ambruoso and Dina Balabanova. 2018. ‘Verbal autopsy in health
policy and systems: A literature review’. BMJ Global Health 3(2): 1-10.
WHO. 1946 / 2020. “Constitution of the World Health Organization”. 1946 – https://bit.ly/38zrROW
(8/3/2020, 11:08 hours). 2020 – https://bit.ly/3aBjJPb (8/3/2020, 11:08 hours).
WHO. 2003. “STEPS: A framework for surveillance. The WHO STEPwise approach to surveillance
of noncommunicable diseases (STEPS)”. World Health Organization, Geneva.
WHO. 2004. “Developing health management information systems: A practical guide for
developing countries”. World Health Organization, Regional Office for the Western Pacific, Manila.
WHO. 2007. “Everybody’s business: Strengthening health systems to improve health outcomes:
WHO’s framework for action”. World Health Organization, Geneva.
WHO. 2010. “Monitoring the building blocks of health systems: A handbook of indicators and their
measurement strategies”. World Health Organization, Geneva.
7
https://www.who.int/medical_devices/assessment/en/ (14/10/2019, 12:57 hours).
6
Introduction
India’s National Family Health Survey (NFHS) has been an enormously valuable source of reliable,
representative and richly disaggregated population-level data on family planning, selected aspects
of health, nutrition as well as related determinants. It has been all the more valuable in the context
of a weak health information system (HIS), including the civil registration and vital statistics (CRVS)
systems. With its round 1 (1992-93) hailed as a ‘landmark in the history of collection of demographic
data through surveys’ (Visaria and Rajan 1999: 3002), 4 more NFHS rounds have been conducted
so far – 1998-99, 2005-06, 2015-16 and 2019-20 – under the stewardship of India’s Union Ministry
of Health & Family Welfare (MoHFW), with the International Institute for Population Sciences (IIPS),
Mumbai, as the nodal agency for its conduct and coordination. Modelled on the lines of USAID’s
Demographic and Health Surveys (DHS)
8
– 400 rounds of which have been conducted in over 90
developing countries since 1984
9
– NFHS has received international financial support from USAID
and other agencies, and international technical support from the ICF International. As such, not only
has the NFHS itself been instituted as a widely reputed and referenced health survey, it has instituted
what we could refer to as a ‘health survey culture’ in the country, nationally, if not yet in the states.
However, even at the national level, the institution / progress of NFHS has not been linear or smooth.
Over time, there have been certain growing realizations in central / state governments that have had
implications for the scope of NFHS, other health surveys as well as the broader HIS landscape. The
foremost realization is that ‘severe fragmentation, compounded by market failures and governance
challenges, is the key driver of India’s underperforming health system’ (Kumar 2019: 6-7). This, inter
alia, has not only meant ‘fragmented data capture by multiple stakeholders’, but that ‘governments,
central and state alike, do not have reliable and complete data for policy analysis and evidence-
based interventions’ (MoHFW 2020: 6). More generally, there is a realization that ‘certain systemic
deficiencies continue to exist in the statistical system’ (MoSPI 2018: 32), and that it ‘needs to be
revamped, modernised and get aligned with the statistical system in the world’.
10
Secondly, there
has been a growing realization that, despite ‘an irreversible process of fertility decline’ even around
the NFHS-2 period (Visaria and Rajan 1999: 3007), which is now ‘close to the replacement level
at the national level and well below it in many states’ (Kulkarni 2020: 70) – and non-communicable
diseases (NCDs) emerging as the leading cause of death, disease and disability across the country
– our health programs and data systems continue to excessively focus on reproductive and child
health (RCH). Thirdly, with India’s growing economic and political influence at the global level, the
country has become more assertive and less accepting vis-à-vis international agencies. Likewise,
8
https://bit.ly/2w1adpK (25/2/2020, 11:58 hours).
9
https://dhsprogram.com/Who-We-Are/About-Us.cfm (25/2/2020, 11:58 hours).
10
https://bit.ly/2Vy0PEQ (29/2/2020, 13:13 hours).
7
external donor funding for health declined from 2.3% of total health expenditure during 2004-05
to an insignificant 0.3% by 2013-14 (NHA 2018: 10).
What have been the implications of these broad realizations / trends for the NFHS as well as other
health surveys in the country? As far as fragmentation is concerned, there have been calls for an
integrated health survey. The GoI reportedly decided to even discontinue the NFHS in April 2012,
as it wanted to ‘integrate all the existing surveys into one’, and obtain ‘data at the district level, so
that action can be taken accordingly’. A National Health Survey (NHS) was proposed instead, to be
fully funded by MoHFW.
11
With backlash from several quarters, the decision was shortly reversed,
and it was decided that the District Level Household and Facility Survey (DLHS) would instead be
subsumed under the NFHS and the latter would provide district-level data with 3-year periodicity.
12
The Annual Health Survey (AHS) was also discontinued in 2013, paving way for NFHS-4 to be
conducted as an integrated health survey and serve as the benchmark for the fifth and subsequent
rounds.
13
The Office of the Registrar General (ORGI), Ministry of Home Affairs (MHA) – responsible
for conducting AHS – was asked to provide district-level infant and under-five morality rates (IMR
and U5MR) and maternal mortality ratio (MMR) with a 3-year periodicity.
14
This is yet to materialize.
Last year, in a meeting of the Cabinet Committee on Economic Affairs (2/1/2019), chaired by the
Prime Minister, it was mentioned that ‘the NFHS provides valuable data for policy and programmes
right up to the district level’ and, for the future, ‘the target for NFHS is to provide reliable data on
all health indicators’. Schematically, NFHS is one of the 2 sub-components of the ‘Health Surveys
and Research Studies’ (HSRS) component – the other sub-component being Rural Health Statistics
(RHS) – of MoHFW’s ‘Umbrella scheme for family welfare and other health interventions’ (Umbrella
scheme, also referred to as ‘Family welfare schemes’ in the Union Budget 2019-20). The scheme
is seen as ‘crucial for attaining the goals and objectives’ of National Health Policy (NHP) 2017 and
international commitments under Sustainable Development Goals (SDGs). Its HSRS component
is expected to ‘assist in keeping a tab on the progress of entire set of health programmes/schemes
run by Government of India, assisting in timely course corrections’.
15
As far as the second realization is concerned, while NFHS is still predominantly RCH-oriented, it
now covers NCDs too in some ways (table 2.4). Finally, as far as the third realization is concerned,
one of the key respondents during our field research, who has been involved with the NFHS data
collection in several states, shared that USAID and ICF International will not be involved with NFHS
sixth round onwards, and it will be completely funded and managed domestically by the MoHFW.
11
Bhattacharya, Pramit. ‘Government to discontinue National Family Health Survey’. Mint, 11/4/2012.
12
Shrinivasan, Rukmini. ‘National health survey not nixed’. The Times of India, 30/7/2012.
13
‘Conducting the NFHS/Annual Health Surveys’. PIB, 2/7/2019. https://bit.ly/2T9X6dE (25/2/2020, 17:05 hours).
14
Krishnan, Vidya. ‘Govt discontinues annual health survey’. Mint, 25/7/2013.
15
‘Cabinet approves Continuation of Umbrella scheme for “Family Welfare and Other Health Interventions” during 2017-
18 to 2019-20’. PIB, 2/1/2019. https://bit.ly/2AHbNMz (6/10/2019, 00:34 hours).
8
Study objectives
Since no independent assessment of the HSRS had been envisioned or commissioned so far vis-
à-vis its objectives, the GoI mandated the NITI Aayog to get it done through a third-party institution.
ICRIER was selected for this purpose. In this context, the present study aims to –
1) Assess the scope of NFHS in the light of the data requirements for the fulfilment of objectives,
achievement of goals, monitoring of progress vis-à-vis – 1) NHP 2017 and other health policies,
2) health programs, 3) NITI Aayog’s dashboard for monitoring of health outcomes,
16
4) SDGs;
17
2) Assess the scope of NFHS vis-à-vis other health surveys;
3) Assess the NIMS (ICMR)-WHO study to validate Verbal Autopsy (VA) tools;
18
4) Assess health surveys of selected countries to draw lessons for India; and
5) Provide recommendations based on above assessments.
Study methodology
The study is based on extensive desk and field research involving key informant interviews in New
Delhi, selected states and districts between 2/12/2019 and 25/1/2020. Kindly refer to Annexure A
for the list of interviewees.
One state was selected from each NFHS region on the basis of IMR (NFHS-4) – Rajasthan from
North, Uttar Pradesh from Central, Bihar from East, Assam from the Northeast, Maharashtra from
West and Kerala from South India. Union territories (UTs) were not considered. The first 4 had the
worst IMR in their respective regions, the latter 2 the best. Such a selection strategy was adopted
not only to capture perspectives on the NFHS from various geographical regions of the country,
but also from both good- and bad-performing states vis-à-vis an outcome (IMR) that has long been
the focus of health policy and programs in the country since independence.
Within these states, we also tried to meet district-level health and statistical authorities in the state
capitals as well as in cities with a Population Research Centre (PRC) outside of the state capital –
in Udaipur (Rajasthan) and Pune (Maharashtra). Like NFHS, PRCs not only fall under the ambit of
the Umbrella scheme and are managed by the Statistics Division of MoHFW, they were involved
16
The dashboard – http://social.niti.gov.in/hlt-ranking (19/10/2019, 11:56 hours) – presents data from NITI Aayog’s Health
Index (NAHI). As such, we will look at the scope of NFHS vis-à-vis NAHI.
17
The Union Ministry of Statistics and Programme Implementation (MoSPI) developed the National Indicator Framework
(NIF) for SDGs, and brought out a report identifying the data sources of various indicators in 2019 and version 2.0 in 2020.
On its part, the NITI Aayog has brought out 2 SDG India Index reports so far, a baseline report in 2018 and one in 2019.
We have referred to both versions of their reports in the assessment of the scope of NFHS vis-à-vis the SDGs.
18
Despite several efforts, we were not able to obtain this study and could not, therefore, include its assessment here.
9
with the NFHS earlier, and many of them are now involved with the monitoring of annual district
Program Implementation Plans (PIPs) of the National Health Mission (NHM). Strengthening survey
research capabilities of the PRCs was one of the objectives of the NFHS (NFHS-1 national report).
As part of desk research, we –
1) Analyzed health surveys and institutional data requirements based on government documents
and websites –
a. Health surveys conducted / commissioned by the MoHFW, MHA and MoSPI;
b. Institutional health data requirements were assessed vis-à-vis MoHFW’s health-related
policies and schemes, health-related SDGs, NITI Aayog’s Health Index, etc. The scope
of NFHS vis-à-vis the above was tabulated;
2) Conducted thematic mapping of health survey indicators to assess their comparative scope;
3) Reviewed health surveys in selected countries (US, Canada and UK) to draw lessons for India.
As part of field research, we tried to –
1) Fill in the information gaps identified during desk research;
2) Understand institutional health data requirements at the state and district levels, and the extent
to which NFHS and other health surveys helped in meeting those requirements;
3) Elicit the views and suggestions of respondents regarding the NFHS and other health surveys.
Challenges
Let alone any assessment of institutional health data requirements, one of the biggest challenges
was the availability of basic information about health policies and programs, especially in the states.
Either the list of health policies and programs is not available in the public domain or not presented
systematically – the output and outcome indicators of schemes included in the Union budget being
the only exception that came to our rescue. Despite health being a state subject, states tend not
to have their own health policies. If they did – as in the case of Kerala (2013, 2018 and 2019 in the
recent past) and UP (‘The Uttar Pradesh Health Policy, 2018’ (draft), which was somehow available
on the Government of Madhya Pradesh’s rather than Government of UP’s website)
19
– quantifiable
targets against which the relevance of NFHS could be assessed were not available. In most cases,
states also did not have their own health schemes; instead, they had either modified some aspects
or renamed the central ones. And once again, their quantitative targets were not available against
19
https://bit.ly/2A3wG81 (8/6/2020, 19:40 hours).
10
which to assess the relevance of NFHS. Officials were generally very supportive, but tend to have
piecemeal information, and that too only vis-à-vis their work profile. They keep getting transferred,
and institutional memory is generally very weak. Given these challenges, we decided to confine
ourselves to the central government for assessing the scope of NFHS vis-à-vis institutional health
data requirements.
Organization of the report
Chapter 1 sets the conceptual context for the discussions on health surveys that follow. It briefly
refers to the notion and determinants of health, the dichotomy of complex vision, but RCH-oriented
action in India, data that the government should collect, the policy context of health systems and
health information systems thinking in India, WHO’s health systems approach to monitoring and
evaluation and whether health surveys should pursue a purely quantitative or mixed methodology.
Section 1 has 5 chapters. Chapter 2 provides an overview of major health surveys in India since
independence, with a special focus on the NFHS and SRS, and discusses their comparative scope.
Chapters 3 to 6 assess the scope of NFHS vis-à-vis health policies, programs, health-related SDGs
and the emerging area of health technology assessment (HTA) respectively at the national level.
Section 2 shares the perspectives of respondents vis-à-vis NFHS as well as other health surveys.
Section 3 provides an overview of major health surveys in selected countries (US, Canada and UK)
and draws lessons for India. Several chapters end with recommendations, marked with ®. The key
recommendations, emerging from our desk and field research, are listed in the concluding chapter.
The chapter also includes a table which characterizes the potential respective features and themes
of the 6 national health surveys that we propose.
Bibliography
Kulkarni, P M. 2020. ‘Stagnancy in the unmet need for family planning in India’. EPW 55(6): 65-
70.
Kumar, Alok. 2019. ‘Synthesis: Transforming the Indian health system in the 21st century’. In: NITI
Aayog. “Health system for a New India”. https://bit.ly/2Vvk0yP (29/2/2020, 12:58 hours).
MoHFW. 2020. “National Health Digital Blueprint”. https://bit.ly/2VvRpJV (29/2/2020, 13:06 hours).
MoSPI. 2018. “National Policy on Official Statistics (Draft)”. https://bit.ly/2IumXrY (7/3/2020, 13:48
hours).
NHA. 2018. “National Health Accounts estimates for India”. MoHFW, GoI.
Visaria, Pravin and Irudaya S Rajan. 1999. ‘National Family Health Survey: A landmark in Indian
surveys’. EPW 34(42/43): 3002-3007.
11
1. Conceptual framework
Before we move into a dense discussion of health surveys at national, state and international levels,
it is important to briefly discuss some of the key concepts from the perspective of health statistics.
Statistics is usually considered the domain of statisticians – and increasingly of IT experts who are
expected to help organize statistics from a wide range of sources into an integrated, user-friendly
framework to support decision-making. In fact, there is so much fascination with leveraging IT now
in the Central government – thanks to the ‘Digital India’ initiative – as well as in states like Rajasthan
that statisticians might worry that they no longer hold a position of privilege as they once did. The
private sector – not least India’s global success in IT – have played their part in stoking the public
sector’s fascination with IT. A top leader of the Indian pharmaceutical industry even argued that –
‘I truly believe that technology can solve many of the daunting challenges of poverty and primitive
living standards that we face as a country’ (Shaw 2018: 49). In a meeting with one of us (Ali Mehdi)
in 2016, senior officials from the US Centers for Disease Control and Prevention (CDC) exhorted
that India should lead the world in health surveillance, given its globally recognized IT credentials.
However, if you talk to any leading statistician or IT expert, she will tell you that conceptual clarity
and domain expertise are foremost, and statistical or IT tools are what they are – ‘tools’. It seems
that concepts are rarely invoked, while it is largely bureaucrats that we have in the name of domain
(health) expertise in the context of health data ecosystems in the country. It is, at best, rare to find
any public health academic / epidemiologist – let alone philosopher / social scientist with expertise
on health issues – involved even in any of the surveys, not to talk of administrative data systems.
For example, NFHS-4 had 4 committees – Technical Advisory, Project Management, Administrative
& Financial Management and a Steering Committee. There were also several project coordinators,
officers and consultants involved with the survey. While many of them are reputed domain experts,
only one member of its Steering Committee (Dr A K Shiva Kumar) was a development economist,
who has been writing on health from a broader social scientific and philosophical perspective. He
too was there, not as an independent expert, but as member of the Mission Steering Group (MSG)
of the National Rural Health Mission (NRHM) and that of the erstwhile National Advisory Council.
The notion and determinants of health
Health, generally, is a highly complex notion in terms of its dimensions (physical, mental and social
– as defined by the WHO), determinants and distribution in a population (by individual, household,
social, political, economic and physical environmental characteristics). It acquires added layers of
complexity in a country as comprehensively diverse as India, with a federal structure of general /
12
health governance,
20
fragmented system of health care financing (figure 1.10) and, not least, multiple
medicinal systems with their own notions of health, disease classifications and service providers.
21
Like health more generally, maternal and child health (MCH) – one of the key focus areas of health
policies and programs in India – is determined by a wider set of factors than by health care alone.
Figure 1.1 provides a conceptual framework, figures 1.2 and 1.3 empirical evidence from 146 low-
and middle-income countries on socioeconomic determinants of child and maternal mortality over
two decades. One important lesson here is – data collection as well as monitoring and evaluation
(M&E) have to be sensitive to this framework and evidence. In other words, we need interoperable
data on these factors from health / related sectors to analyze the determinants of health. Program
M&E has to keep such wider factors and data into consideration while analyzing the precise impact
of health care policies and programs, which have a limited role to play vis-à-vis health outcomes.
Figure 1.1: Determinants and consequences of maternal and child undernutrition
Source: UNICEF 2013: 4.
20
According to the NHP 2017, ‘one of the most important strengths and at the same time challenges of governance in
health is the distribution of responsibility and accountability between the Centre and the States’ (27).
21
Ayurveda, Siddha and Unani have their own morbidity codes. Refer to the NAMASTE (National AYUSH Morbidity and
Standardized Terminologies Electronic) portal. http://namstp.ayush.gov.in/#/index (27/2/2020, 16:35 hours).
13
Figure 1.2: Contribution of changes in the levels of determinates of health (health interventions,
social and environmental determinants) to reductions in U5MR, in 146 LMICs,1990-2010
Source: Bishai et al 2016: 9.
Figure 1.3: Contribution of changes in the levels of determinates of health (health interventions,
social and environmental determinants) to reductions in MMR, in 146 LMICs,1990-2010
Source: Bishai et al 2016: 10.
14
The dichotomy of complex vision and RCH-oriented action in India
There has been due recognition of the general complexity of health and the willingness to address
it as such in the national discourse. Around the same time as WHO’s constitution and definition of
health (1946), the Bhore Committee Report (BCR) argued that ‘the term health implies more than
an absence of sickness in the individual and indicates a state of harmonious functioning of the
body and mind in relation to his physical and social environment, so us [sic] to enable him to enjoy
life to the fullest possible extent and to reach his maximum level of productive capacity’. However,
it felt that ‘data regarding positive health are more-difficult to collect than those relating to sickness
and mortality’ (Vol. 1: 7). The first NHP 1983 started out with a reference to the constitutional vision
for ‘the establishment of a new social order based on equality, freedom, justice and the dignity …
the elimination of poverty, ignorance and ill-health … raising the level of nutrition and the standard
of living … improvement of public health’. The NHP 2017 ‘envisages as its goal the attainment of
the highest possible level of health and well-being for all at all ages, through a preventive and
promotive health care orientation in all developmental policies’. It calls for addressing the ‘social
determinants of health’ through ‘an empowered public health cadre’ and ‘developmental action in
all sectors’; ‘achieving convergence among the wider determinants of health’ for urban health; the
strengthening of Panchayati Raj Institutions (PRIs) so they can ‘play an enhanced role at different
levels for health governance, including the social determinants of health’. It argues that ‘maternal
and child health is a mirror that reflects the entire spectrum of social development’; and makes a
case for ‘research on social determinants of health’. From a programmatic perspective as well,
‘intersectoral convergence’ has been seen as a ‘key to the success of the NHM’,
22
India’s flagship
health program. ‘The thrust of the mission is on establishing a fully functional, community owned,
decentralized health delivery system with inter-sectoral convergence at all levels, to ensure
simultaneous action on a wide range of determinants of health such as water, sanitation,
education, nutrition, social and gender equality’. To achieve it, ‘the District/City Health Action Plan
is an important institutional structure for enabling decentralization, convergence, and integration,
and is also the vehicle for promoting equity and prioritizing the needs of the most socially and
economically vulnerable groups in a district’.
23
On its part, the NFHS provides richly disaggregated
data on the wider determinants and distribution of MCH as well as some aspects of general health.
Yet, in terms of program operationalization, financing and actual practice, the integration of ‘health’
and ‘family welfare’ (HFW) has meant that the focus of HFW has predominantly been on population
control / stabilization / FW / RCH / MCH rather than on health per se, even on its physical, let alone
its mental and social dimensions. India’s population has been viewed as problematic since colonial
22
https://nhm.gov.in/index1.php?lang=1&level=2&sublinkid=1084&lid=149 (28/2/2020, 13:40 hours).
23
https://nhm.gov.in/images/pdf/NHM/NHM_more_information.pdf (28/2/2020, 14:21 hours).
15
to present times.
24
‘India was the first country in the world to have launched a National Programme
for Family Planning in 1952’ (MoHFW Annual Report 2015-16: 81). Despite starting out on a highly
visionary note, NHP 1983 argued – ‘irrespective of the changes, no matter how fundamental, that
may be brought about in the over-all approach to health care and the restructuring of the health
services, not much headway is likely to be achieved in improving the health status of the people
unless success is achieved in securing the small family norm’. Despite being one of the most
visionary programs as far as conceptualization is concerned, 63.1% of total expenditure under
NHM between 2005-06 and 2015-16 was on its RCH component (NRHM-RCH flexible pool) vis-à-
vis 4.5% under its flexible pool for communicable disease control programmes and 1.4% under
its flexible pool for non-communicable disease (NCD) programmes (figure 2.8), despite the fact
that during the same period (2005-16), 6.7% of total deaths in the country were due to maternal
and neonatal disorders, 26.8% were due to communicable diseases and 55.6% due to NCDs –
the remaining being due to nutritional deficiencies (0.7%) and injuries (10.2%) (Global Burden of
Disease / GBD).
25
Yet, despite all the prioritized focus and allocations for RCH, and all the progress,
India continues to be the world’s leading contributor to child deaths since 1960, and managed to
become the second largest contributor to maternal deaths in 2008 (World Development Indicators
/ WDI, The World Bank).
26
And, despite NFHS providing richly disaggregated data, its focus (RCH)
seems to have been inspired more by health program funding and practice than the broad vision
of health which inspired health policy and program documents. As recently as 20/12/2019, NITI
Aayog held a national consultation, ‘Realizing the vision of population stabilization: Leaving no one
behind’.
27
At least, RCH is not going to be left behind any time soon, so it seems. External funding
for health may be miniscule, influence at least vis-à-vis population concerns does not seem to be.
What data should the government collect?
As is evident from the above discussion, there has been a disconnect between the government’s
understanding and vision for health (complex) on the one hand, and its programs and funding for
health (RCH-oriented) on the other. The relevant question is – should its data collection be guided
by the former or the latter?
24
‘India came to be imagined, both by Indians and in the West, as an overpopulated place. Beginning in the nineteenth
century and continuing into the twentieth, fears of overpopulation haunted Indian political culture, thereby shaping state
policy and civil society debates’, and inspiring ‘the colonial management of famine in the nineteenth century; debates
on contraception and reproductive technologies in the early twentieth century; and policies of population control in the
post-independence period’ (Sreenivas 2010). ‘In more recent years, some in the United States and Europe have argued
that this large population poses a global threat, as Indians consume an ever-increasing portion of the world’s resources’
(Sreenivas 2009).
25
http://ghdx.healthdata.org/gbd-results-tool (27/2/2020, 16:17 hours).
26
A total of 936,338 under-five child and 35,000 maternal deaths were recorded in the year 2017 (WDI).
27
https://pib.gov.in/newsite/PrintRelease.aspx?relid=195978 (1/3/2020, 11:34 hours).
16
In its “Three-year Action Agenda (2017-18 to 2019-20)”, the NITI Aayog suggested a ‘stewardship
role’ for the government, which involves – 1) ‘setting and enforcing rules / incentives to guide the
behaviour of the health system’, 2) ‘a data-driven and more decentralised approach to designing
health systems’, 3) increases in government health expenditure ‘to cover screenings for the entire
population, active case detection and disease surveillance including from the private sector’, 4)
‘availability of credible population-level data on the prevalence of risk factors and complete health
outcome data at frequent intervals’, and 5) ‘evidence-based preventive health interventions’ (NITI
Aayog 2017: 144-145). In a more recent document of NITI Aayog, there is a reference to ‘systems
approach to health’, ‘the stewardship function of the health system’ as ‘typically the core mission
of the national health authorities’, and efforts to take ‘a comprehensive view, impacting the multiple
determinants of health’ (Kumar 2019: 3,4,12). Figure 1.4 depicts the proposed system stewardship
function for the central government, with ‘intelligence’, notably, being the first of its 4 components.
Figure 1.4: Proposed health systems framework
Source: Kumar 2019: 12.
Clearly, the NITI Aayog’s thinking as well as general efforts toward data integration, as referred to
earlier, make it clear that GoI as well as some states like Rajasthan wish to go beyond their specific
health funding and programs, and take a comprehensive health systems’ view based on data from
various sources to increasingly play a stewardship role vis-à-vis the entire range of processes and
outcomes than merely be funders and providers of selective health care for selective populations.
17
This would also imply redefining the scope of government-sponsored administrative data sources
as well as surveys. More population-based data is being collected through administrative sources
through door-to-door surveys and mass screening of eligible persons. The scope of NFHS too has
broadened over the years. We will discuss this in detail in the next section. However, let us quickly
highlight that a health systems thinking is not new and has been reflected in various health policies.
The policy context of health systems and health information systems thinking in India
A health systems’ thinking has been there since India’s first National Health Policy. The NHP 1983
called for the establishment of a decentralized, ‘well dispersed network of comprehensive primary
health care services’, together with ‘a well worked out referral system’, and ‘services and support'
of the private health sector to be ‘utilised and intermeshed with the governmental efforts, in an
integrated manner’. It had a dedicated section on ‘Management information system’ (MIS), where
it argued that ‘appropriate decision making and programme planning in the health and related
fields is not possible without establishing an effective health information system. A nation-wide
organisational set-up should be established to procure essential health information. Such
information is required not only for assisting in planning and decision making but to also provide
timely warnings about emerging health problems and for reviewing, monitoring and evaluating the
various on-going health programmes. The building up of a well-conceived health information
system is also necessary for assessing medical and health manpower requirements and taking
timely decisions, on a continuing basis, regarding the manpower requirements in the future’. What
we are trying to do now had been proposed in a broad sense 37 years earlier itself.
NHP 2002 recognized that ‘unsatisfactory health indices’ were ‘an indication of the limited success
of the public health system’ (PHS). Noting ‘distortion’ in PHS of vertical implementation structures,
it envisaged ‘gradual convergence of all health programmes under a single field administration’,
‘full operationalization of an integrated disease control network’, a ‘public health surveillance
network’, which will include ‘information from private health care institutions and practitioners’.
With reliable data on ‘incidence of various diseases, the public health system would move closer
to the objective of evidence-based policy-making’. In its aftermath, NRHM was launched in 2005
‘to strengthen the Rural Public Health System’,
28
with a dedicated component of ‘health systems
strengthening’,
29
getting the highest share within NHM budget – 39% of total approved NHM outlay
in Union budget 2018-19 and 29% in 2019-20. ‘Strengthening data capturing, validity / triangulation’
is one of the imperatives under the program, involving complete registration of births and deaths
under the CRS, recording births in private facilities, data collection on key performance indicators,
rationalization of HMIS indicators and ensuring reliability of health data by means of triangulation.
28
https://nhm.gov.in/index1.php?lang=1&level=2&sublinkid=971&lid=154 (18/10/2019, 16:42 hours).
29
http://164.100.154.238/nrhm-components/health-systems-strengthening.html (18/10/2019, 16:42 hours).
18
A lesser known fact is the collection of cause of death (CoD) data as part of NHM’s maternal and
infant death reviews at facility and community levels.
30
We will discuss this in some detail shortly.
NHP 2017 has taken a step further compared to NHPs 1983 and 2002 by focusing more explicitly
on HIS. It has specific quantitative goals and objectives under 3 broad themes, one of them being
health systems strengthening (HSS). There are 3 indicators under each of the 3 sub-components
of HSS, one of the latter being health management information (HMI), whose goals are – 1) ensure
district-level electronic database of information on HS components by 2020; 2) strengthen ‘health
surveillance system and establish registries for diseases of public health importance by 2020’; 3)
‘establish federated integrated health information architecture, Health Information Exchanges and
National Health Information Network by 2025’. It talks of ‘an integrated health information system’
with ‘private sector participation in developing and linking systems into a common network/grid
which can be accessed by both public and private healthcare providers’. It talks of strengthening
post-marketing surveillance (PMS) for drugs, etc. as well. And most relevant from the perspective
of the present study, it calls for extending ‘the scope of health, demographic and epidemiological
surveys to capture information regarding costs of care, financial protection’. It also suggests ‘rapid
programme appraisals and periodic disease specific surveys to monitor the impact of public health
and disease interventions using digital tools for epidemiological surveys’. Another unique aspect
of NHP 2017 is its commitment to ‘development of institutional framework and capacity for Health
Technology Assessment’ (HTA), with its own set of implications for data collection in the country.
The Health Technology Assessment in India (HTAIn) in MoHFW’s Department of Health Research
has already been established. We will discuss it in some detail in chapter 6.
This overview demonstrates that a health system and health information system thinking has been
there at least at the policy level. There has been piecemeal operationalization of the vision of NHPs
in this regard, and it is now time to do so in an organized and integrated manner. However, most
importantly, we need to keep in mind that statistical and IT experts can only offer the tools for this
endeavor; a comprehensive vision and conceptual framework needs to be developed beforehand.
WHO’s health systems approach to monitoring and evaluation
In the context of the broadening scope of administrative data sources and surveys, it has become
imperative for GoI as well as state governments to develop / adopt a health systems approach to
monitoring and evaluation (M&E) too. Let us quickly discuss the work of World Health Organization
(WHO) in this regard and draw some lessons for the Indian context.
In 2007, the WHO presented its health system framework with 6 ‘building blocks’ (BBs), including
one on information. Focused on health systems’ performance, health determinants and outcomes,
30
http://www.nrhmhp.gov.in/content/reporting-formats-child-health (2/3/2020, 20:23 hours).
19
a well-functioning information BB or HIS ‘ensures the production, analysis, dissemination and use
of reliable and timely health information by decision-makers at different levels of the health system’
(WHO 2007: 18). It is important to note the word ‘use’ here – without it, the collected data becomes
useless. Data ‘collection and analysis should not be allowed to consume resources if action does
not follow’ (Foege, Hogan and Newton 1976: 29-30). In a 1956 address, titled ‘Statistics must have
purpose’, P C Mahalanobis, the architect of modern statistical methods in the Indian sub-continent
(Ghosh et al 1999), argued that ‘before starting to collect any new statistics it is useful to pause
and enquire, “for what purpose?” When a statistician is requested to collect some statistics his first
responsibility is to ask the person or authority making the request to explain as clearly as possible
the purposes for which the information would be used’ (Mahalanobis 1956: 3). Figure 1.5 highlights
the data cycle with a data use hemisphere, involving data interpretation and subsequent response
from a public health perspective. When we discuss the scope of NFHS, we would also refer to its
data use hemisphere because if it is not actually used for some reason, its scope gets limited in a
practical sense, which would also have implications for its data collection in the first place. Another
critical thing to note in WHO’s description of information BB is ‘production, analysis, dissemination
and use’ at ‘different levels of the health system’. Those below are not just supposed to collect
and share statistics upwardly – the entire data cycle is supposed to be operationalized at various
levels of the health system. This is also an integral part of the data use hemisphere.
Figure 1.5: Data cycle
Source: World Bank 2006: 1001.
In 2010, WHO put forth a data strategy for health system M&E (figure 1.6), involving a variety of data
collection sources pertaining to different health system BBs. The idea that administrative sources
/ surveys / registries / Electronic Health Records (EHR) would be sufficient seems naïve in the light
of this M&E framework. Figure 1.7 makes a further distinction between internally-driven monitoring
and externally-driven evaluation – monitoring should ‘be carried out internally by the implementing
20
agency and focus on linkages between inputs, processes and outputs’, while ‘evaluation efforts to
determine an intervention’s effect on health outcomes and impact may be conducted’ externally
by an independent agency and / or program’s ‘intended clients or beneficiaries’ (WHO 2016: 10).
Figure 1.8 is WHO’s Global Reference List (GRL) of 100 core indicators for harmonized monitoring
of health systems from the perspective of various international agencies.
Figure 1.6: WHO’s health systems monitoring and evaluation framework
Source: WHO 2010: viii.
Figure 1.7: Internal and external monitoring and evaluation
Source: WHO 2016: 10.
21
Figure 1.8: WHO’s Global Reference List of 100 core health indicators (including health-related SDGs)
Source: WHO 2018: 16.
22
Health surveys – quantitative or mixed methods?
Finally, a note on the desirable methodology of health surveys.
We tend to think of surveys in particular, data collection activities in general, in quantitative terms.
Among the dictionary meanings of ‘survey’ are – 1) ‘a general view, examination, or description
of someone or something’; 2) ‘an investigation of the opinions or experience of a group of people,
based on a series of questions’ (Oxford);
31
3) ‘an examination of opinions, behaviour, etc., made
by asking people questions’ (Cambridge).
32
None of these definitions seem to restrict the meaning
of the term ‘survey’ to the quantitative. On the contrary, one could argue that they convey a much
broader analytical as well as descriptive impression. The pre-independence ‘Heath Survey and
Development Committee (1943-46) – widely known as the Bhore Committee – as well as the post-
independence ‘Health Survey and Planning Committee’ (1959-61) – widely known as the Mudaliar
Committee – both established by GoI, were of this nature, as we shall briefly highlight, before we
move on to discuss the better known health surveys. Both these surveys as well as the National
Mental Health Survey (NMHS), conducted during NFHS-4 period (2015-16), used a mixed method
approach. The former two included detailed analytical reports on health and health care situations
of their times and discussion of wider determinants of health; the NMHS had a sociodemographic
information proforma and included socioeconomic impact assessment of the mental health burden.
Even the DHS Program – in which global DHS surveys, including NFHS, are anchored – also designs
and supports qualitative / mixed method surveys and research for understanding social and cultural
dynamics related to health, population and nutrition. ‘By using a qualitative approach to examine the
social and cultural contexts of daily life, The DHS Program works to increase the validity and
reliability of its surveys, to expand the information available for monitoring and evaluation, and to
contribute original qualitative research in the fields of anthropology, demography, and public health.
The capability to collect data through qualitative as well as quantitative methods provides a unique
opportunity to learn and demonstrate how quantitative and qualitative approaches can be linked to
expand our understanding of social and cultural dynamics related to health, population and nutrition
around the world’. RCH, child nutrition and HIV/AIDS are among the topics covered by DHS through
qualitative research, using ‘observations, participation, rapid assessment procedures, various types
of individual or group interviews, personal narratives, focus group discussions, and content analysis
of medical records and other documents’.
33
Let us conclude this discussion here by a reference to the famous libertarian economist, Friedrich
August von Hayek. In his lecture, ‘The pretence of knowledge’, delivered on receiving 1974 Nobel
31
https://www.lexico.com/definition/survey (11/3/2020, 9:54 hours).
32
https://dictionary.cambridge.org/dictionary/english/survey (11/3/2020, 9:55 hours).
33
https://dhsprogram.com/What-We-Do/Survey-Types/Qualitative-Research.cfm (11/3/2020, 12:02 hours).
23
Memorial Prize in Economic Science, he argued that ‘unlike the position that exists in the physical
sciences, in economics and other disciplines that deal with essentially complex phenomena, the
aspects of the events to be accounted for about which we can get quantitative data are necessarily
limited and may not include the important ones. … social sciences … have to deal with structures
of essential complexity … whose characteristic properties can be exhibited only by models made
up of relatively large numbers of variables. …. the superstition that only measurable magnitudes
can be important has done positive harm in the economic field’ (Hayek 1974). Health is surely one
of the ‘essentially complex phenomena’ – as borne out by WHO’s definition of health as well as
references to ‘well-being’ in our own NHP 2017. Likewise, aspects of health ‘about which we can
get quantitative data are necessarily limited and may not include the important ones’. In this
context, the obsession with ‘measurable magnitudes’ of health surveys is actually quite paradoxical,
given that their comparative disadvantage lies in not being able to assess physical health beyond
a limited scope (self-reported morbidity / biomarkers), while they tend to ignore their comparative
advantage in focusing on more subjective dimensions of health, given their opportunity to interact
with people, an opportunity they seem oblivious of. A definition and operationalization of the notion
of health may help in reorienting health surveys as well as health policy and practice more broadly
towards health per se and closer to people’s day-to-day sufferings. Health, suffering, pain, etc. are
more subjective regular issues, and rarely one-time events (birth, death, hospitalization, etc.). The
shift in focus toward chronic diseases has been taken to mean biomarkers, etc., while the fact that
‘chronic’ also means prolonged suffering, with several subjective health issues (related to mental
health or more regular aches – lower back, neck, etc.) and a major bearing on one’s quality of life,
seems to have been largely ignored. And it is precisely in these subjective contexts of chronic
diseases that traditional and complementary systems of medicine (TCSM), especially Yoga, have
gained enormous worldwide popularity. According to US National Health Expenditure 2007 data,
Americans spent USD 33.9 billion out-of-pocket on TCSM. According to the NIH’s National Cancer
Institute, ‘just as cancer affects your physical health, it can bring up a wide range of feelings you’re
not used to dealing with. It can also make existing feelings seem more intense. They may change
daily, hourly, or even minute to minute’.
34
These are not random, hypothetical issues which do not
merit attention – they are at the core of what we as human beings feel on a daily basis. The health
narrative in general, India’s in particular, has been sorely deficient from this perspective. If health
surveys capture such themes, they may not necessarily remain so and make more sense to their
primary stakeholders – the people of India.
34
https://www.cancer.gov/about-cancer/coping/feelings (26/10/2019, 13:16 hours).
24
Recommendations
® In the context of India’s health transition, the GoI should adopt / develop a definition of health
35
which should guide the design and assessment of all health-related activities, including health
information systems. Although this may seem basic / academic / unnecessary, one could argue
that it is, inter alia, due to the absence of a publicly articulated / guiding definition that health
has historically come to mean population stabilization / fertility / mortality / RCH in India. Efforts
should also be made to operationalize a positive notion of health beyond the negative death /
disease / disability oriented notion, which should not be difficult in a country where local health
traditions / traditional systems of medicine have adopted a positive, holistic approach to health.
® In line with India’s health transition, health data collection should also shift from a demographic
to a predominantly health orientation – according due importance to the population dimension.
At the same time, we need to ensure that the emergent health orientation is not exclusively /
predominantly biomedical, and is sufficiently focused on the wider determinants of health. As
such, GoI should involve experts from all health-related disciplines, including social scientists,
ethicists and others working on health issues for administrative data as well as health surveys.
® None of the NFHS committees has independent experts or state representatives. Health survey
committees should also include independent experts from public health, epidemiology, social
sciences, etc. working on health issues to enhance the domain and conceptual underpinnings
of the survey as well as representatives from state / UT governments to enhance the sense of
ownership at the subnational level. For instance, NHM’s Mission Steering Group (MSG) has 9
public health professionals as well as health secretaries of the high-focus states as members.
36
® GoI should develop a National Health Data Policy (NHDP) and a National Health Data Advisory
Committee (NHDAC) with members from relevant ministries / departments of central and state
/ UT governments (health, statistics, planning); national organizations like IIPS, ICMR, National
Institute of Mental Health and Neuro-Sciences (NIMHANS) and Indian Council of Social Science
Research (ICSSR); international organizations like the UN Statistical Commission, WHO, UNDP,
UNFPA, UNICEF and the World Bank; leading international health statistics agencies like NCHS
from the US, Statistics Canada and NHS Digital from the UK; leading national and international
health scholars; industry and civil society representatives.
® The NHDAC should develop a health systems framework and health-related goals, targets and
indicators with timelines like SDGs – a National Reference List (NRL) of core health indicators,
35
India is a signatory to WHO’s constitution, and it could be argued that it affirms the definition of health enshrined in it.
https://bit.ly/32mZWjR (24/2/2020, 14:55 hours). If that is the case, it too should be spelt out clearly rather than assumed.
36
https://nhm.gov.in/index1.php?lang=1&level=1&sublinkid=1293&lid=193 (28/2/2020, 13:06 hours).
25
like the WHO’s GRL, which is periodically revised to incorporate emerging concerns. For every
indicator, there should be a rationale, standardized definition, numerator, denominator, method
of measurement and estimation, disaggregation, frequency, preferred and other data sources,
baseline value, etc. State / UT governments should, likewise, develop SHDPs, SHDACs and
SRLs. NRLs and SRLs should guide interoperable data collection through a variety of sources.
® The NRL / SRL should be developed vis-à-vis core indicators of national / state health policies
and programs, international data reporting requirements (including health-related SDGs) and
WHO’s Family of International Classifications (WHO-FIC). The entire health information system
should be revised according to the above.
® Figure 1.6 shows that various data sources are required for monitoring various health system
building blocks, and there is a preferred respective role for each of them. In India, surveys like
SRS and NFHS are seen as compensating for weaknesses of administrative sources, including
the civil registration system. The latter need to be strengthened. However, strengthening them
would not imply that surveys are no longer required – as a high-ranking health official in a state
argued. Surveys have their own role to play, in periodically monitoring ‘outcomes’ and ‘impact’
of not just specific policies and programs, but more generically. It is not necessary that surveys
provide data in real-time or much more frequently, as some policymakers and experts expect,
although their periodicity should be annual / biennial at the most (as in US, Canada and UK –
kindly refer to section 3 for details).
® Health surveys should focus on monitoring the vision / goals / objectives of health policies and
programs to periodically ensure that they are being fulfilled. Program MIS / other mechanisms
(ground assessments by DGHS Regional Offices, PRCs, review missions, local communities,
etc.) should be strengthened for regular program monitoring and evaluation. Surveys should
not be expected to help in MIS data validation beyond a few core indicators. Where there is an
expectation from surveys to help validate MIS data, indicator definitions, population coverage,
etc. should be harmonized. At the moment, their definitions, numerators, denominators, etc.
do not necessarily match and validation cannot be done in a strict sense. In fact, data from 2
surveys also does not match precisely due to these as well as several methodological reasons.
That does not automatically make one data source inferior, the other superior, based on overall
perceptions rather than case-specific assessments of data quality of the concerned sources.
® No independent, systematic and exhaustive review of India’s public health surveillance system
has been conducted. GoI should commission an independent review urgently, especially given
the widespread prevalence – and, therefore, particularly from the perspective – of COVID-19.
® The division of labor between the 4 relevant ministries could be the following. MHA looks after
all population-related indicators through census (decennial), CRVS (continuous) and the SRS
26
(annual enumeration-cum-survey) – as it already does. MoHFW should look after public health
surveillance – as it already does. However, it should be Department of Health Research (DoHR)
in MoHFW, rather than its Department of Health and Family Welfare (DoHFW) – as is presently
the case – which should lead and coordinate all public health surveillance activities, with the
exception of policy- and program-based MIS. DoHFW, being the operational wing of MoHFW,
should manage various MIS in an integrated, consistent and coordinated manner. The MHA and
MoHFW could collaborate for a cause of death survey, given that it requires domain expertise
which the MHA lacks. DoHR / ICMR institutions should be involved in this case. MoSPI should
oversee all health surveys in consultation with MoHFW and MHA. All statistical activities should
strictly be conducted under its statistical guidance, coordination and supervision – and, in the
case of health, under the domain-related guidance of the DoHR (MoHFW). MoSPI, in its turn,
should play a more proactive rather than a passive role as far as statistical coordination and
regulation is concerned – both at the Central and state / UT levels. For this, MoSPI needs to
be independent, both from political interference and IAS-led bureaucracy. There is a serious
conflict of interest that the DoHFW, which manages various health schemes, also commissions
and manages the independent NFHS, conducted by an agency (IIPS) which itself ‘is under the
administrative control’ of MoHFW.
37
Not only this, the MoHFW’s Statistics Department, which
manages the HMIS, also manages the NFHS. These are very serious conflicts of interest which
should be addressed immediately. If need be, the DoHR should be renamed as the Department
of Health Research and Surveillance (DoHRS) – research and surveillance go hand in hand –
and all health surveillance activities, including surveys, should be carried out under its domain
supervision and MoSPI’s statistical supervision. Budgets and staff in both these organizations
need to be enhanced accordingly. The DoHR / ICMR already has a network of leading centers
across the country, which can be leveraged for this purpose. However, at the same time, DoHR
needs to go beyond its ‘biomedical’
38
approach – with ICMR as only one network of institutions
– and adopt a much more broad-based approach to health – with another network of public
health and health-related social scientific institutes developed / supported by it. COVID-19 has
already put DoHR in the lead. It is time that its role be expanded as we reorient for the future.
® While data collection is important, analysis is also part of data generation hemisphere, followed
by interpretation and response under the data use hemisphere (figure 1.5). All four data-related
frameworks need to be strengthened at the central / state / local-most levels – it cannot be the
exclusive prerogative of researchers / statisticians on sidelines (DES, NIHFW / SIHFWs, PRCs,
etc.) or at the top (ICMR, IHME, etc.) to analyze / interpret data. Central / state / local capacities
need to be strengthened and IT tools leveraged for the entire data life-cycle. In fact, those who
37
http://iipsindia.org/about.htm (15/6/2020, 11:53 hours).
38
https://dhr.gov.in/about-us/about-department and https://main.icmr.nic.in/ (15/6/2020, 11:58 hours).
27
collect data at local levels can sometimes contextualize and contextually analyze it better than
those who do not know / understand the local context in which the data was collected. This is
also in keeping with the spirit of decentralization inherent in the conceptualization of the NHM
– India’s leading public health program.
® There has to be a clearly defined framework for data collection, processing, synthesis, analysis
and use for the design and assessment of policies and programs as well as course-correction.
In the absence of such a framework – despite humongous data collection and ‘reporting’ within
the system as part of MIS and accountability of various functionaries – data ‘use’ for the design
and assessment of policies and programs as well as course-correction is not seen as important
and becomes an arbitrary / whimsical activity. Statisticians and IT can provide the tools, but it
is eventually the domain officials who have to use the data from a policy / program perspective.
This is seriously missing across the country – from the national to the local levels, including in
states like Kerala (field interactions).
® Ease of data use should be facilitated for policymakers as well as other stakeholders. This is a
huge challenge at the moment. The STATcompiler customization tool for DHS surveys
39
/ the
visualization hub of GBD data on causes of death are 2 excellent examples.
40
A Kolkata-based
organization, Riddhi, has developed spatial visualization tools for NFHS-4 and Census data.
41
A senior health official in Kerala found the NFHS-4 tool helpful for promptly accessing data –
he said reports are bulky and one needs to collate data from different reports for comparisons.
There are a few efforts in this direction within the government system as well. For e.g. the NITI
Aayog’s National Data and Analytics Platform (NDAP) ‘aims to democratize access to publicly
available government data’.
42
However, we need a dedicated platform for ease of data use for
policy purposes for various levels of officials, which goes beyond visualization dashboards and
helps in using data to develop policies and programs and monitor / course-correct the latter.
® A mixed methods approach should be adopted to health surveys in the country. For guidance,
we could refer to health surveys conducted in India pre- and post-independence and The DHS
Program,
43
of which the NFHS is a part, for instance. The richness of the notion and experiences
39
https://www.statcompiler.com/en/ (7/3/2020, 12:50 hours).
40
https://vizhub.healthdata.org/gbd-compare/ for global data / https://vizhub.healthdata.org/gbd-compare/india for India
and state data (7/3/2020, 12:53 hours).
41
http://nfhs4.indiagis.org/nfhs4/ and http://www.censusgis.org/india/ (7/3/2020, 13:09 hours).
42
https://pib.gov.in/PressReleaseIframePage.aspx?PRID=1600370 (7/3/2020, 13:12 hours).
43
‘The DHS Program supports qualitative research to produce informed answers to questions that lie outside the purview
of a standard survey approach to understanding issues in health, population, and nutrition … to increase the validity
and reliability of its surveys, to expand the information available for monitoring and evaluation … The capability to collect
data through qualitative as well as quantitative methods provides a unique opportunity to learn and demonstrate how
28
of health, well-being and health care utilization and satisfaction can best be captured by means
of qualitative methods. Qualitative information could also imbue the otherwise dry quantitative
data with an intimate, human sense and help in making health systems more people-oriented.
® The DHS Program also has several types of surveys and not just the standard DHS, according
to which the NFHS has been modelled. Beyond the standard DHS surveys – with large sample
sizes, typically conducted every 5 years, to allow comparisons over time – ‘interim DHS’ focus
on select indicators, are conducted between standard DHS, have shorter questionnaires and
sample sizes, but are nationally representative. There is also ‘continuous DHS’, as part of which
data is collected and reported annually by a permanent DHS office and field staff.
44
There are
‘in-depth DHS’ and ‘mini DHS’
45
too. India should adopt a dynamic approach to health surveys,
and consider the various options available in the light of its requirements.
Bibliography
BCR. 1946. “Report of the health survey and development committee”. Volume 1. The Manager
of Publications, Delhi. https://bit.ly/2TzSjUq (8/3/2020, 11:28 hours).
Bishai, David et al. 2016. ‘Factors contributing to maternal and child mortality reductions in 146
low- and middle-income countries between 1990 and 2010’. PLoS ONE 11(1): 1-13.
Foege, W, Robert Hogan and Ladene Newton. 1976. ‘Surveillance projects for selected diseases’.
International Journal of Epidemiology 5(1): 29-37.
Ghosh, J K, P Maiti, T J Rao and B K Sinha. 1999. ‘Evolution of statistics in India’. International
Statistical Review 67(1): 13-34.
Hayek, Friedrich. 1974. ‘The pretence of knowledge’. Lecture to the memory of Alfred Nobel, 11
December 2017. https://bit.ly/2Ul50SV (8/6/2020, 11:41 hours).
Kumar, Alok. 2019. ‘Synthesis: Transforming the Indian health system in the 21st century’. In: NITI
Aayog. “Health system for a New India: Building Blocks. Potential pathways to reform”.
https://bit.ly/2Vvk0yP (29/2/2020, 12:58 hours).
Mahalanobis, P C. 1956. ‘Statistics must have purpose’. Presidential Address, Third Pakistan
Statistical Conference, Lahore.
NITI Aayog. 2017. “India: Three-year Action Agenda (2017-18 to 2019-20)”. https://bit.ly/3aEz2H3
(7/3/2020, 13:50 hours).
quantitative and qualitative approaches can be linked to expand our understanding of social and cultural dynamics
related to health, population and nutrition around the world’. https://bit.ly/30IXsgE (15/6/2020, 11:22 hours).
44
https://bit.ly/3cZvrUB and https://bit.ly/2Bd0TBq (15/6/2020, 12:21 hours).
45
For a list of various types of DHS conducted so far, kindly refer to – https://bit.ly/3e43qwr (15/6/2020, 12:21 hours).
29
Shaw, Kiran Mazumdar. 2018. ‘Leveraging affordable innovation to tackle India’s healthcare
challenge’. IIMB Management Review 30(1): 37-50.
Sreenivas, Mytheli. 2009. ‘Population bomb? The debate over Indian population’. Origins 3(2).
https://origins.osu.edu/article/population-bomb-debate-over-indian-population (5/2/2020, 18:25).
Sreenivas, Mytheli. 2010. ‘Population Policy in India: The Role of the Rockefeller Foundation and
the Population Council’. https://www.issuelab.org/resources/27945/27945.pdf (5/2/2020, 21:14).
UNICEF. 2013. “Improving child nutrition: The achievable imperative for global progress”. New
York.
WHO. 2007. “Everybody’s business: Strengthening health systems to improve health outcomes:
WHO’s framework for action”. World Health Organization, Geneva.
WHO. 2010. “Monitoring the building blocks of health systems: A handbook of indicators and their
measurement strategies”. World Health Organization, Geneva.
WHO. 2016. “Monitoring and evaluating digital health interventions: A practical guide to
conducting research and assessment”. World Health Organization, Geneva.
WHO. 2018. “Global Reference List of 100 core health indicators (plus health-related SDGs)”.
World Health Organization, Geneva.
World Bank. 2006. “Disease control priorities in developing countries”. Second edition. The World
Bank and Oxford University Press, Washington DC and New York.
SECTION 1 –
HEALTH SURVEYS AND THE SCOPE OF
NFHS AT THE NATIONAL LEVEL 30
2. Health surveys in India
In this chapter, we will discuss health surveys in India in a chronological order, with a special focus
on the NFHS and SRS. Towards the end, we will highlight their comparative / respective coverage
and offer recommendations towards India’s health survey strategy.
Heath Survey and Development / Bhore Committee (1943-46)
This is the first official ‘survey of the whole field of public health and medical relief’, ‘a broad survey
of the present position in regard to health conditions and health organisation in British India and
to make recommendations for future development’, commissioned by the GoI in 1943 (BCR 1946,
Vol. 1: 1). Given its broad mandate, it not only provided data on life expectancy, death rate, infant
and under-10 child deaths, maternal mortality, deaths due to epidemic diseases and other causes,
but also on ‘the social background of ill-health’, on the ‘causes of the low level of health in India’
– e.g. insanitary conditions, social customs, people’s outlook, nutrition, education, unemployment
and poverty. The authors of the report argued that ‘our survey of the causes of ill-health in India
will not be complete without drawing attention to the profound influence that these factors exert
on the health of the community’ (ibid.: 17). Six decades before the Michael Marmot-led 2005 WHO
Commission on Social Determinants of Health (CSDH) that held ‘a toxic combination of poor social
policies, unfair economics, and bad politics’ responsible for ill-health, the BCR highlighted the role
of the broader determinants of health,
46
unlike the narrow focus of health surveys of our time, with
limited exceptions, NFHS being one of them.
National Planning Committee (NPC) – National Health Sub-Committee Report (1948)
Appointed by Jawaharlal Nehru in 1938, under the chairmanship of Col S S Sokhey (also known,
therefore, as the Sokhey Committee Report), the NPC had 29 sub-committees, with one being on
population and another on health. ‘Investigation into the volume and causes of infant mortality, as
well as mortality among women’, ‘compilation of vital statistics’ were among the terms of reference
of the national health sub-committee. Noting that ‘India is very poor in statistical information of all
kinds’, it proposed ‘vital statistics accurately taken and maintained for statistics and research’ as
one of the ‘preventive measures’ and a village survey with the following scope and methodology,
inspired by a comprehensive view of health and its determinants like the Bhore Committee earlier.
46
In Britain, vast differences in life expectancy across occupational groups were reported by Edwin Chadwick in his
‘Report on the sanitary condition of the labouring population and on the means of its improvement’ way back in 1842,
which became the basis for the first Public Health Act there in 1848. So, there was a background even to what the Bhore
Committee was referring to. See Mehdi (2019) for further discussion on this. Likewise, universal health coverage, too,
had a background in pre-independence India and Britain, in terms of provisioning it for the working class, to begin with.
31
No definite programme can successfully be made with out a through survey of the local conditions.
This survey must include all the necessary data needed for future planning and expansion.
For the general health of a community is dependent on the topography, soil formation,
productivity, industry, water supply, drainage, population (man and cattle), weather conditions,
education and economical or cultural backgrounds of a community, therefore the survey must
include all these items.
Besides the survey of the society and its surroundings one must possess intimate knowledge of
the individual’s mental and physical health this makes it imperative to make study of not only each
family, but every individual.
This survey must be done on card system by school teachers who may be provided with
questionnaire; and they may not only survey once and be done with them, but keep them up-to-
date by checking and rechecking on them for many years. Students have proved useful in
gathering information of this kind of survey work.
This survey will give us information of the success or failure of a programme, therefore though
monotonous, dry and statistical in outlook, it has to be carried out with scientific precision.
The survey schedule, with the following components, was also included in the report –
1) Basic details (name, district, province, name of villages in north, south, east and west)
2) General topography, quality of soil, area
3) Sources of water supply, drainage, sewage, garbage
4) Household population characteristics (male, female, under 5 and 5-16 years, occupation)
5) Kinds of produce and trade
6) Dairy products
7) Animals, animal fodder and sheds
8) Tree surveys
9) Housing
10) Family survey (with details of the head of the household– for e.g. caste, race, gender, age,
occupation, income, economic and housing status and details, family diet)
11) Family health survey (name, gender, age, illness in past 12 months and duration, treatment
(home, dispensary, private), cost of medical care, details of physician, dispensary, hospital)
12) Morbidity survey (communicable and chronic diseases, disease names, number of cases,
average number of days / person, whether epidemic / endemic / quarantine, cases traced)
13) Vital statistics
14) Individual’s detailed medical history (with socioeconomic characteristics)
32
15) Detailed physical examination by physician (with socioeconomic characteristics)
16) School survey with attendance and health record
17) Population survey
Truly an inspiration on the scope of health surveys from the perspective of WHO’s definition and
social determinants of health.
National Sample Survey (NSS, 1952-)
Given the poor status of health statistics at the time of independence, health was one of the focus
areas for the NSS since its inception in 1950. IMR from as far back as its 2nd (1952), 14th (1958-
59) and 17th to 20th rounds (1961-62 to 1965-66) are available. Its estimates were, however, not
considered reliable (Chandrasekhar
47
1972: 141-142). As per RGI (1989), population surveys were
‘an integral part’ of annual rounds of the NSS during 1958-68, after which they were discontinued.
‘Comprehensive survey on population, fertility, family planning and mortality’ was reintroduced in
its 28th round (1973-74), with the second one conducted in its 39th round (1984). These ‘surveys
conducted by NSS since 1958 have provided comprehensive data on all aspects of population
covering size, structure and composition of population, marital status, births, deaths and migration’
(1). NSS has conducted 4 health surveys since the 1990s – during its 52nd (1995-96), 60th (2004),
71st (2014) and, most recently, 75th round (2017-18), covering 113,823 households (NSO 2019).
The NSS conducted ‘exploratory’ surveys on morbidity, starting with the 7th round (1953-1954),
followed by three subsequent rounds (11th-13th rounds, 1956-58). A pilot survey was conducted
in its 17th round (1961-1962) ‘to examine alternative approaches of morbidity reporting’. Based
on the experience of these surveys, a ‘full-scale’ survey on morbidity was conducted in the 28th
round (1973-1974). Subsequently, no separate morbidity surveys were undertaken, and morbidity
became part of ‘decennial surveys on social consumption’. The first national social consumption
survey (SCS) – household schedule 25.0 on ‘Social Consumption: Health’ – was carried out during
its 35th round (1980-81) – covering the public distribution system, health services (including mass
immunisation), family welfare, etc. The second SCS 25.0 was carried out during the 42nd round
(1986-1987), and included problems of the aged. The third SCS 25.0 was carried out during the
52nd round (1995-1996), dropping utilisation of the public distribution system and family planning
services since the former was covered during the 50th round (1993-94), the latter in NFHS-1. The
fourth SCS 25.0 was conducted in the 60th round (2004), covering morbidity, health care and the
problems of the aged on MoHFW’s request. The fifth SCS 25.0 was conducted after a gap of 10
years during the 71st round (2014), covering self-reported morbidity and hospitalisation, childbirth
and maternity health care services and condition of the aged, including the extent of use and cost
47
S Chandrasekhar was Health Minister in Indira Gandhi’s cabinet. https://bit.ly/2MLUeSr (25/10/2019, 13:04 hours).
33
of treatment of AYUSH for the first time. It also provides information on proportion of ailing persons
per 1,000 for acute and chronic ailments by age, residence, gender and quintile. However, unlike
other rounds, 60th and 71st rounds were 6-month surveys with smaller sample size and, therefore,
could not provide even nationally representative data (NSO 2019). The sixth SCS 25.0 survey was
conducted during the 75th round (2017-18). Data on outbreak of certain communicable diseases,
immunisation status and related expenditure was also collected. Table 2.9 offers a comparative
overview of major themes covered in the latest NSS and NFHS.
NSS has been conducting disability surveys too. The first attempt to gather information on physical
disability was during its 15th round (1959-60), and subsequently in the 16th (1960-61), 24th (1969-
70) and 28th (1973-74) rounds, all of which were exploratory in nature, and provided data only on
the number of persons suffering from type of physical disabilities in India. A comprehensive survey
on disability was carried out in the 36th round (1981), followed by the 47th round (1991), with the
objective of developing a database on the prevalence and incidence of disability in the country. It
provided data on all types as well as the particulars of physical disabilities, like degree of disability,
cause, age, etc. along with demographic characteristics of households. In the 58th round (2002),
along with physical, data on mental disability was also collected. The latest disability survey was
conducted during the 76th round (2018), covering 7 disability types – locomotor, visual, hearing,
speech and language, mental retardation / intellectual disability, mental illness other disabilities.
NSS has also been conducting quinquennial (5-yearly) household Consumer Expenditure Surveys
(CES) since its 27th round (1972-73) to generate the ‘estimates of household monthly per capita
consumer expenditure (MPCE), its distribution over households and persons, and its break-up by
commodity group, at national and State/UT level, and for different socio-economic groups’. These
estimates help assess the population’s standard of living, nutrition (energy, protein and fat intake),
poverty, inequality as well as certain indices related to the national economy. Latest available data
is from the 9th quinquennial survey in the series conducted in its 68th round (2011-12), that of the
10th CES survey, conducted during its 75th round (2017-18), was not released due to ‘data quality
issues’. Two back-to-back CES surveys were under consideration for 2020-21 and 2021-22.
48
We
should highlight that while these surveys provide data on nutrition, they are ‘not specially designed
to serve the needs of a nutrition survey’ (‘Nutritional Intake in India, 2011-12’, 68th round report).
It also needs to be noted that, over different rounds, sample size and selection, classifications and
the definitions of variables have changed. Until the 60th round, only ailments treated by medical
personnel were considered to be treated, which changed since the 71st round. With growing self-
medication in India, the 71st and 75th rounds consider use of over-the-counter (OTC) drugs too
as treatment. The definition of disability also changed since 71st round, with pre-existing disability
48
Seshadri, Suresh. ‘What is Consumer Expenditure Survey, and why was its 2017-2018 data withheld?’. The Hindu
(24/11/2019). https://bit.ly/2wVJ5sA (15/3/2020, 18:24 hours).
34
considered as chronic ailment only when it is under treatment. Further, classification of ailments
and nature of treatment has also changed with more categories being added. Therefore, caution
needs to be exercised while analyzing, comparing and interpreting data over the rounds. Further,
it is to be noted that the data pertaining to ailments is self-reported, and not diagnosed, except for
few chronic diseases like cancers, tuberculosis, HIV/AIDS, diabetes, etc. In all the other ailments,
prevalence was reported by respondents or diagnosed on the basis of symptoms, which may lead
to wrong reporting / classification of ailments. Another issue has been the presentation of data on
medical services received from modern and traditional systems of medicine (AYUSH) – the report
provides percentage of ailments treated by allopathy and AYUSH, which adds up to 100 percent,
despite the fact that people tend to use various systems of medicine in a ‘complementary’ rather
than an ‘alternative’ manner, and so their use should not add up to 100 percent. This is one of the
classic examples of how the lack of domain expertise can lead to errors in providing estimates. Of
course, people can use raw data and calculate absolute rather than relative use, but the ability to
use raw data is quite limited among policymakers as well as other relevant stakeholders. We have
pointed out later that in one of the Kerala reports, a statistician in the state Department of Health
Services totaled the IMR for different districts to provide state IMR! Another case which highlights
need for domain and statistical experts to work together for data collection, tabulation and analysis.
Another limitation is that NSS ‘still does not give reliable district-level estimates for most indicators.
… experts fear that over the last two decades, developments like cuts in staff strength, change in
recruitment practices, and the creation of a ‘centralised’ Subordinate Statistical Service have
resulted in a shortfall of experienced field investigators, which may have compromised the quality
of data being collected by the NSSO. … considerable resources are also spent to collect what is
known as the ‘state sample’, which covers at least a matching number of respondents in each
state. Pooling of central and state samples will enhance the policy relevance of NSS data, as these
can then be used to arrive at district-level estimates’ (Kurian 2016: 10-11). The SRS was instituted
‘when the inadequacy of the National Sample Survey became obvious’, (Jeffery 1988: 23), when
NSS proved not ‘to be reliable due to the effect of non-sampling errors, which are only too obvious
in the results’ (Chandrasekhar 1972: 74).
Health Survey and Planning Committee (1959-61)
Sixteen years later, the government of independent India appointed a ‘Health Survey and Planning
Committee’ in June 1959 to review the ‘developments that have taken place since the publication’
of the Bhore Committee Report, and ‘formulate further health programmes for the country in the
third and subsequent five-year plan periods’ (Mudaliar Committee Report 1962,
49
Vol. 1: 7). The
49
Subsequent committee reports – Chadha (1963), Jungalwalla (1967), Mukerji (1968), Kartar Singh (1973), Shrivastav
(1975) and Bajaj (1986) – were sectoral in nature, focusing on specific aspects, rather than a comprehensive survey.
35
Mudaliar Committee adopted a mixed methods strategy – eliciting ‘information and views’ through
907 questionnaires (receiving 665 responses), visit to ‘representative institutions’, ‘interviews with
representatives of organizations’ as well as with other individuals, listening to ‘views, experiences
and suggestions’ of foreign participants at the 1961 World Health Assembly in India, and ‘scrutiny
of memoranda received from various sources’ (ibid.: 11). Like the Bhore Committee Report, it also
referred to the social determinants of health – ‘the most essential conditions for promotion of health
are good housing, adequate and whole some food, abundant supply of potable water, proper
disposal of sewage, free perflation of air, prevention of overcrowding and clearing of slum areas,
supply of pure milk, particularly to children’, not least, ‘the development of sanitary conscience in
the community’ (ibid: 64). Its report referred to NSS data on the distribution of doctors and hospital
services in rural areas and out-of-pocket expenditure on health care (ibid: 74, 79).
Model Registration System (MRS, 1965-81), Survey of Causes of Death (SCD), Rural (1981-
98)
50
India introduced the world’s first VA-based CoD reporting in 1965 to get CoD data from rural areas
as part of the MRS, and subsequently the SCD. In January 1999, the SCD was also discontinued,
and a CoD component was added to the SRS. MRS and SCD have been discussed in some detail
elsewhere (RGI 1988; Mahapatra and Rao 2001; Mahapatra 2002; Chattopadhyay and Agnihotram
2004; Mahal, Karan and Engelgau 2010), so we will leave it here.
51
However, let us point out that
both the MRS and SCD were designed to get rural CoD data from a sample of villages, given lack
of a robust CRVS system. Like the SRS, both of them were based on continuous enumeration as
well as half-yearly verification by a PHC statistician. It seems that because of the similarity in their
approaches, the SCD was discontinued and CoD for the entire country subsumed under the SRS.
Sample Registration System (SRS, 1971 –)
Given the challenges of the CRVS system, the ORGI initiated the sample registration of births and
deaths scheme as ‘an interim measure’ in the 1960s (Mahapatra 2017: 2) so as to provide regular,
reliable and representative vital statistics (birth and death rates). It was piloted in a few states during
1964-65, launched across the country during 1969-70 and has regularly been providing data since
1971. Table 2.1 shows the indicators and background characteristics for which it currently provides
data at national level, state level for smaller states and NSS natural division level
52
for major states.
50
We could find the years of these surveys only in Colaco (2016).
51
For a robust discussion on CoD in India, kindly also refer to http://www.ihs.org.in/BurdenOfDisease/CauseofDeath.htm
(13/3/2020, 19:11 hours).
52
NSS natural divisions are ‘classified group of contiguous administrative districts with distinct geographical and other
natural characteristics’ (SRS Statistical Report 2017: 2).
36
Table 2.1: SRS indicators and background characteristics, 2017
Indicators Disaggregation
Annual estimates
Percentage distribution of estimated population Age-group, residence, gender, marital status
Fertility Age-group, residence, mother’s educational level
Percentage distribution of live births Birth order, residence, gender, birth interval
Percentage distribution of deaths Age-group, residence, gender
Death rate Age-group, residence, gender
Child mortality (all levels from SBR to U5MR) Residence, gender
Medical attention received at delivery (mothers)
and before death at public / private facilities
Residence
Three-yearly estimates
Sex ratio at birth (3-yearly) Residence
Cause of death
2 state groupings – EAG + Assam and other states
By major ICD codes at the national level (by age-
group, residence, gender)
By 4 major ICD cause groups and top 10 causes
for state groupings (by residence, gender)
By top 10 causes for age-groups (by state
groupings, residence, gender)
Maternal mortality ratio and rate, lifetime risk Age-group
Five-yearly estimates
Life expectancy Age-group, residence, gender
Source: Developed by author from various SRS reports.
SRS is based on a dual record system, beginning with a baseline survey, involving preparation of
a notional map of areas to be surveyed, house numbering / listing and filling of household schedule
with the residential status and other demographic characteristics of each individual residing in the
household – identification code, name, sex, date of birth, age, educational / marital status, etc. List
37
of all women in reproductive age-group, along with their pregnancy status, is also prepared during
the baseline survey. The second phase involves continuous enumeration of births and deaths in
sample villages and urban blocks by resident part-time enumerators – usually Anganwadi workers
and teachers. The third phase is independent and retrospective half-yearly surveys (HYS) by SRS
supervisors belonging to the statistical cadre of State Census Directorates. While carrying out the
HYS survey, supervisors do not have access to birth and death records of enumerators for same
periods, which are withdrawn from the field before supervisor’s visit for the HYS survey. The fourth
phase involves matching data from two sources – i.e. continuous enumeration and HYS survey –
while the fifth phase involves verification in the field of the unmatched and partially matched events
‘to get an unduplicated count of correct events’. In the sixth phase, verbal autopsy (VA) forms are
filled for finalized deaths to obtain CoD estimates. Table 2.2 lists the SRS forms and the information
collected as part of them, while figure 2.1 depicts the entire SRS process.
It is interesting to note that ‘for ensuring complete netting, the enumerator uses different sources
to get information of the occurrence of vital events in the sample unit. These include the help of
the village priest, barber, village headman, midwife and such other functionaries. The enumerators
maintain contact with these informants at frequent intervals and collects information about the
occurrence of births and deaths. On being informed about the occurrence of an event, the
enumerator visits the concerned household and records the prescribed particulars. The
enumerator also keeps in touch with other socially important persons and visits local or nearby
hospitals, nursing homes, cremation or burial grounds, at frequent intervals to keep updated about
the occurrence of events. Besides, enumerator maintains and updates a list of all women in the
reproductive span along with their pregnancy status, which helps in better netting of all the births.
Despite all these efforts, the enumerator may miss information about some of the events and is,
therefore, required to visit all the households once a month in urban area and once in a quarter
in rural areas so as to ensure that all the events have been recorded’.
53
The infant mortality rate is taken as the decisive indicator for estimation of sample size. HYS 2017
was conducted in 8,853 sample units, covering about 7.9 million sample population. SRS sampling
frame is revised every 10 years based on results of the latest Census. While changing the sample,
modifications in sampling design, wider representation of population, overcoming limitations in the
existing scheme, meeting additional requirements, etc. are taken into account. First sampling frame
replacement was done in 1977-78, the latest in 2014. Given its sample design, the SRS is the only
panel household health-related survey in the country that can potentially help in tracking changes
in vital statistics at the individual / household level. Given its sample size, it claims to be the world’s
‘largest demographic survey’. To add more feathers to its cap, SRS data has been used to conduct
Census Evaluation Studies (CES) ‘to estimate the extent of misreporting of age at younger ages
53
http://censuskarnataka.gov.in/SRS%20Introduction.pdf (11/3/2020, 13:35 hours).
38
in the census’, with ‘ages recorded’ in the SRS ‘taken to be the true ages’. The CES, along with
the census PES, have highlighted ‘the constraints in data collection in a large operation like the
census’, helping improve ‘future census operations’ (CES 1996: iii). At the same time, its data has
also been used to check under-enumeration in CRS. SRS vital rates are much more widely used
than those of the NFHS or other sources.
The NFHS has a major rival in the SRS – its sample design and size, continuous enumeration and
bi-annual independent retrospective survey methodology, periodicity, validation potential vis-à-vis
the census and CRS, ability to canvass CoD surveys and provide data not only on a higher range
of vital statistics (for e.g. MMR and life expectancy as well), but also on the whole range of diseases
as per ICD-10, being an independent source of data from within the government system, etc. give
SRS a massive advantage over the NFHS. However, the NFHS also has an edge over the SRS on
certain parameters – it is coordinated by a group of professional and independent demographers
with the best international support as far as vital statistics are concerned; provides data on a wide
range of background characteristics that SRS usually does not (for e.g. religion, caste/tribe, wealth
index, etc.); is able to provide at least some data up to the district level (though not vital statistics).
However, as far as background characteristics are concerned, the ORGI did commission 3 special
SRS surveys – in 1978, 1984 and 1997 – ‘to throw light on inter-relationship of fertility and mortality
indicators with various socio-economic factors’ (RGI 1989), providing data by caste, religion, etc.
As far as district level data is concerned, the ORGI did contemplate in 2012 ‘a massive increase
by adding 46,000 sample units for the year 2015-16’. Given the ‘prohibitive costs and operational
difficulties’, however, the idea was not taken any further (Mahapatra 2017: 6). The sample size of
NFHS-4 was instead expanded so that it could provide data at the district level.
SRS CoD (VA) surveys
The SCD was integrated with the SRS on 1st of January 1999 (MHA Annual Report 2003-04: 118).
Four rounds of SRS CoD surveys have been conducted ever since – 2001-03, 2004-06, 2007-09
and 2010-13 – covering 455,460 deaths.
54
SRS forms 10A to 10D are meant for these surveys.
Figure 2.1 illustrates the process. However, given the limited sample size, only 3-yearly estimates
are possible; that too, only for 2 groupings of all states; and, more importantly, even at the national
level, we do not have data for each ICD-10 code, rather several codes have to be lumped together
to provide estimates for broader categories (table 2.1). So, lumping of samples, states as well as
ICD codes. And the latest available estimates are almost a decade old. Although SRS CoD surveys
have been a silver lining, there is still a lot of darkness out there as far as regular, representative
data on causes of death in the country is concerned. In this context, there have been attempts to
fill in the gaps through the Global Burden of Disease and National Burden Estimates (NBE, Menon
54
https://bit.ly/2WgRH5J (26/10/2019, 13:58 hours).
39
et al 2019) by researchers affiliated with the ICMR. The ORGI has been making efforts not only to
revamp the CRVS system, but also to boost up the VA-CoD statistics.
55
Both are urgently needed.
‘Recent studies suggest that VA can provide cause of death information that, at the population
level, is similar to death certification in high-quality hospitals’, argues Dr Chattopadhyay, Director
of All India Institute of Hygiene and Public Health.
56
However, it has also been argued that the VA
method assumes that ‘most causes of death can be recognized by trained physicians based on
descriptions of signs and symptoms provided by a close relative of the deceased. Identification is
simple for causes of death with distinct symptoms such as tetanus and injuries but can be difficult
to distinguish in cases which have symptoms common to many diseases, such as of the fever
during malaria’. Furthermore, VA is ‘not very useful in identifying causes of death in persons over
70 years of age’.
57
Strengthening of MCCD should be a top priority for improved CoD data quality.
Table 2.2: SRS forms and their respective data points
Form No. Form Name Data points
1 House list
Building number, name and status of household head, house
(residential, etc.) and household (migration, etc.) over 4 HYS
2 Household schedule
Name and identification code (used in most forms) of household
members, relationship to head, gender, DOB, date of first marriage,
total number of children (born alive, surviving), status over 5 HYS
(age, educational status, residential status, marital status, reason of
migration – work, education, marriage, etc.)
2A
Morbidity, personal
habits and socio-
economic status
Identification details, self-reported morbidity (name of the disease –
codes 1 to 9 are given, the last being ‘Others (specify)’ – duration
of disease, source of treatment), behavioural risk factors for
persons aged 15+ years (tobacco, smoking, alcohol)
2B
Maternity history and
family planning
practices schedule
(currently married
women aged up to
49 years)
Identification details
Woman – age, age at marriage, number of children born and
surviving, awareness and use of contraceptives by type
Children – gender, DOB, surviving / dead
55
https://causeofdeathindia.com/ (13/3/2020, 19:39 hours). An MoU was signed between RGI and AIIMS Delhi in March
2017 to provide technical support to RGI for SRS-based VA. AIIMS developed an online platform, the Minerva (Mortality
in India established through Verbal Autopsies), for this purpose. https://bit.ly/32HLXUW (25/10/2019, 13:37 hours). Future
data on CoD using VA will come from this source. ‘Assigning most-probable underlying cause of death for nearly 50,000
deaths identified annually under the SRS all over the country’ is its target.
56
https://bit.ly/32HLXUW (25/10/2019, 13:37 hours).
57
https://causeofdeathindia.com/background/ (25/10/2019, 13:53 hours).
40
Form No. Form Name Data points
3
Pregnancy status of
women
Identification details, pregnancy status, expected month and
outcome of pregnancy (live births / still births / abortions), remarks
4
Outcome of
pregnancy recorded
by enumerator
Identification details, residential status, age, details of outcome of
pregnancy (place of delivery, outcome and date, birth multiplicity,
gender, birth weight), type of attention at delivery / abortion, details
of sterilization of husband and wife
5
Deaths recorded by
enumerator
Identification details, particulars of the deceased (identification
details, place of death, residential status, date of death, gender)
age at death, type of medical attention at the time of death
6
Monthly report of
outcome of
pregnancy
Identification details of household head and pregnant woman,
details of outcome of pregnancy (place and date of live births / still
births / abortions, gender)
7
Monthly report of
deaths
Identification details of household head and the deceased, place
and date of death, residential status, gender, age at death
8
Unit-wise
consolidated monthly
report
Sample unit code, population as on 1st January / July, live births,
still births, abortions, deaths, infant deaths, remarks
9
Outcome of
pregnancy recorded
by supervisor
Identification details (including age and level of education of the
pregnant woman), details of outcome of pregnancy, details of
previous child, order of current live birth, birth interval, type of
attention at delivery / abortion, sterilization, matching remarks
(code: fully matched – 1, partially matched – 2, unmatched – 3),
reverification remarks
10
Deaths recorded by
supervisor
Identification details, place and date of death, residential status,
gender, death registration (yes – 1, no – 2, not known – 3), age at
death, order of birth in case of infant death, type of medical
attention at the time of death, matching and reverification remarks
11
Finalized list of
outcome of
pregnancy
Identification details, residential status, age, level of education, type
of attention at delivery / abortion, details of outcome of pregnancy
(including birth registration status), details of previous birth, birth
interval, sterilization
12
Finalized list of
deaths
Identification details and particulars of the deceased, type of
medical attention at the time of death
13
Results of the Half
Yearly Survey on
Gender, live births, still births, abortions, remarks
41
Form No. Form Name Data points
outcome of
pregnancy
14
Results of the Half
Yearly Survey for
deaths
Gender, deaths, infant deaths
15
Distribution of usual
resident population
of the sample unit by
age, sex and marital
status
Age-group, marital status by gender
16
Distribution of total
female population by
broad age-group,
sex and level of
education
Age-group, level of education
17
Number of females
who got married by
age at effective
marriage
Age at effective marriage, number of females
SRS – Verbal Autopsy Form (2011 versions)
10A
Neonatal death (28
days or less of age)
Name of the head of the household, full name of deceased, name
of mother of the deceased, identification codes of the head,
deceased and mother of the deceased, unique form number, SRS
unit number, year
Section 1
Respondent – name, relationship with deceased, lived with the
deceased during the events that led to death, age, gender,
education, religion of the head of the household, identification code
of the respondent
Details of deceased – gender, age, relationship of the deceased
with the head of the household, date of birth, date of death, house
address of the deceased, PIN, place of death, HH-reported cause
of death
Section 2
Death details –
• Death due to injury or accident
42
Form No. Form Name Data points
• Details of pregnancy and delivery – duration of the pregnancy,
immunization during pregnancy, complications, single /
multiple birth, place of delivery, medical attention at birth
• Details of baby after birth – signs of a live birth, birth weight,
child’s size at birth vis-à-vis other children in that area, breast-
feeding details
• Details of sickness at the time of death – fever, diarrhoea,
breathing issues, cough, spasms or fits
o Physical symptoms – for instance, yellow eyes or skin, cold
body, discolored hands / legs / lips, redness around
umbilical cord stump
Section 3
Written narrative in local language – description of symptoms in
order of appearance, doctor consulted or hospitalization and
history of similar episodes
10B
Child death (29 days
to 14 years)
Name of the head of the household …
Section 1
Respondent – name, relationship with deceased, lived with the
deceased during the events that led to death, age, gender,
education, religion of the head of the household, identification code
of the respondent
Details of deceased – gender, age, relationship of the deceased
with the head of the household, date of birth, date of death, house
address of the deceased, PIN, place of death, HH-reported cause
of death
Section 2
Death details –
• Death due to injury or accident
• Details of baby after birth – premature, duration of pregnancy,
child’s size at birth vis-à-vis other children in that area,
breastfeeding details, child’s size at birth, birth weight
• Details of sickness at the time of death – multiple illnesses
along with symptoms, consumption of antibiotics, immunization
received (BCG, DPT, polio, measles), growth of children vis-à-
vis other children of the same age, details of fever, diarrhoea
o Disease symptoms – for instance, stiffness of body / neck,
breathing issues, convulsions or fits, blood in stools, cough,
abdominal pain, vomit, skin rash, red eyes, yellow eyes /
43
Form No. Form Name Data points
skin, weight loss, swelling of hands / feet / abdomen, lack
of blood
Section 3
Written narrative in local language – description of symptoms in
order of appearance, doctor consulted or hospitalization and
history of similar episodes
10C
Adult death (15
years or older)
Name of the head of the household …
Section 1
Respondent – name, relationship with deceased, lived with the
deceased during the events that led to death, age, gender,
education, religion of the head of the household, identification code
of the respondent
Details of deceased – gender, age, relationship of the deceased
with the head of the household, education, occupation, date of
death, house address of the deceased, PIN, years deceased lived
at this address, place of death, HH-reported cause of death
Section 2
Past medical history of the deceased – hypertension, heart disease,
stroke, cholesterol, diabetes, tuberculosis, HIV/AIDS, cancer,
asthma and other chronic illness, medications taken regularly by
the deceased during the last five years
Behavioural risk factors (tobacco, smoking, alcohol, dietary habit)
of the deceased and the respondent
In case of death of a female aged 15-49 years: known or suspected
to be pregnant or within 42 days of delivery or abortion.
Description of key symptoms – fever, breathlessness, cough,
diarrhoea, chest pain, paralysis, urinary problems, jaundice, etc.
Section 3
Written narrative in local language – description of symptoms in
order of appearance, doctor consulted or hospitalization and
history of similar episodes
10D
Maternal death
(females aged 15-49
years)
Name of the head of the household …
Section 1
Duration of pregnancy, pregnancy history, antenatal care, duration
between delivery / abortion and death, place of delivery / abortion,
medical attention at delivery, caesarean delivery, complications
during delivery – prolonged labour, excessive bleeding at the
44
Form No. Form Name Data points
beginning of labour pain / after delivering the baby, difficulty
delivering the placenta, fever after birth, etc.
Section 2
Written narrative in local language – description of symptoms in
order of appearance, doctor consulted or hospitalization and
history of similar episodes
Source: Forms 10A-10D – https://bit.ly/2IF86ef; rest – https://bit.ly/2TZHfi5 (11/3/2020, 20:34 hours). Developed
by Priyanka Tomar. 45
Figure 2.1: The SRS process, including VA CoD survey
Source: Adapted by Priyanka Tomar, Divya Chaudhry and Rajesh Chaudhry from SRS Statistical Report 2017, RGI and AIIMS 2017.
Baseline survey
(By supervisor with the help of enumerator)
Continuous enumeration by part-
time enumerator (PTE)
Houselist
(Form 1)
Household Schedule
(Form 2)
List of pregnant women
(Form 3)
Retrospective Half-Yearly
Survey / HYS by supervisor
Partially matched & unmatched deaths re-verified through
field visit for obtaining an unduplicated count of deaths
Form 12: Finalized list of deaths used by SRS supervisor for
conducting VA
VA conducted by supervisor for deaths listed in Form 12
using appropriate VA forms
Complete VA forms received at ORGI
Scanning of VA forms at ORGI after review, corrections as required
QC of scanned VA forms, sent to 2 physicians, after review, correction as required
Reverification of
randomly selected
VA forms
CoD assigned by
physicians
Updation of
Forms 1, 2 & 3
BIRTHS
Netting of births
(Form 9)
DEATHS
Netting of deaths
(Form 10)
BIRTHS
Netting of births
(Form 4)
Monthly report
(Form 6)
DEATHS
Netting of deaths
(Form 5)
Monthly report
(Form 7)
Matching
Births: Form 4 with Form 9
Deaths: Form 5 with Form 10
Partially matched or unmatched Completely matched
RE-VERIFICATION
(Independently by another supervisor)
Form 11: Finalised list of births
Form 12: Finalised list of deaths
Form 13: Result of HYS for births
Form 14: Result of HYS for deaths
Transmission of
forms to ORGI
Correct birth &
death 46
National Family Health Survey (NFHS, 1992 –)
We had offered a brief overview of the NFHS at the beginning. Let us now discuss it in some detail.
Objectives
Why exactly was the NFHS launched, and what is the purpose that it was expected to serve earlier
and now? Since SRS was already providing data on selected vital statistics for two decades before
the NFHS was launched, some justification for the latter should have been provided in the context
of the former. However, no such justification vis-à-vis SRS in particular / health information system
in general has been provided (table 2.3). This is particularly surprising given the demographic / RCH
focus of both the SRS and the NFHS. One could ask – couldn’t the additionally required indicators
been included in the SRS instead of launching a separate survey altogether, especially given the
relative strengths of SRS vis-à-vis NFHS, some of which we highlighted in the write-up on SRS?
Nevertheless, let us discuss the objectives of the NFHS, as put forth on NFHS website and national
reports of its various rounds (table 2.3). Despite some commonalities, it seems the objectives of
NFHS have also changed over the rounds. Let us start with the commonalities. The NFHS website
as well as most national reports highlight that providing data for policy and program purposes was
one of its key reasons. In chapters 3 and 4, we will assess the scope of NFHS vis-à-vis all health-
related policies and programs of MoHFW, and in section 2 vis-à-vis that of the selected states. For
now, let us look at 2 related email responses from the IIPS Director (16/10/2019, with some others
from IIPS and ICF International involved with NFHS copied on the email – henceforth ‘IIPS email’) –
1) ‘Before each round of NFHS, there is detailed consultation with different programme divisions
of MoHFW and other ministries on their data requirements. Accordingly, indicators are
finalised and questions are added or aligned to SDG health indicators’. We will also review the
scope of NFHS vis-à-vis SDG 3 (health) in chapter 5.
2) ‘There are enough evidence of using NFHS data for policy and programme in India, starting
from using it as inputs in different Five year plans to recently launched national nutrition
mission. Ministry of Health and Family Welfare has used NFHS results as scientific evidence
for various policy decisions including adopting target free approach in 1996, setting goals for
the national population policy 2000, framing of different national health policies, etc. Notable
policy or programme changes have also been based on NFHS results in areas such as
domestic violence, child marriage, menstrual protection, sanitation, and C-section deliveries.
In addition, various states have brought out state-specific population and health policies and
programmes based on state level findings of various rounds of NFHS’.
This is quite remarkable as far as inputs for ‘design’ of a series of policies and programs at national
and state levels is concerned. However, several references in table 2.3 indicate that NFHS is also
47
expected to help in their M&E. MoHFW’s Annual Report 2017-18 states that ‘the Ministry has been
conducting large scale survey periodically to assess the level and impact of health interventions -
National Family Health Survey (NFHS)’ (22). There is also reference to ‘effective management’ of
HFW programmes and monitoring of their ‘key process indicators’ using NFHS data in table 2.3.
Assessment of a few ‘impact’ and ‘process indicators’ within a broad framework through NFHS is
fine, but there are serious challenges if we try to get more nuanced from a program perspective
or try to use NFHS data to validate that of the MIS. For instance –
1) With a few exceptions, NFHS and MIS data are not comparable largely because –
a. Their data points / indicators have not been harmonized;
b. NFHS offers population-level data, while MIS largely has facility / beneficiary data, which
does not include data from the private sector and is, therefore, not representative. There
is also no claim of population-level completeness / representativeness vis-à-vis MIS data,
although the severely over-worked and statistically under-competent ASHAs and ANMs
are expected to collect community-based data as well, conduct surveys, etc. Comparing
data collected by frontline health workers with data from a professional survey like NFHS
is asking for too much. With all its weaknesses, the MIS data helps the service delivery
system to function and fulfil its reporting requirements at various levels. But that is it;
2) MIS data is available on real-time / daily / monthly basis and is used as such for M&E purposes.
NFHS national reports have been released with a gap of 10 to 22 months (table 2.4), although
preliminary data / fact sheets have been made available earlier to the programme managers.
58
Nevertheless, it is not clear how a programme manager could hold functionaries accountable
based on data which is several months old or plan for the future precisely for the same reason.
Health outcome data could be relevant, but the same cannot be said about process indicators.
However, even vis-à-vis outcome indicators (fertility, mortality, etc.), both the Centre and the
states tend to use SRS data anyways rather than the NFHS due to some of the reasons spelt
out earlier (periodicity, from the government system, etc.). Furthermore, the Centre (MoHFW
and NITI Aayog) is conducting annual assessments of state performance, states like Rajasthan
are doing so for their districts on a monthly basis. There is still some room for annual estimates
of the SRS in such assessments. NFHS does not seem too helpful for regular M&E purposes.
58
Vaccination coverage is one of the key and most widely used indicators of the NFHS. However, 2 different figures on
‘all age appropriate vaccinations’ are available from 2 versions of the NFHS-4 national report. According to one version
– https://bit.ly/2TO8iyg – it is 27.4% for India; according to another – https://bit.ly/39UxVmI (DHS), https://bit.ly/2TWvlWc
(NFHS) (13/3/2020, 20:40 hours) – it is 42.9% (a difference of 15.5%)! And both reports have the December 2017 date!
If any revision was made subsequently, the dates of the reports should have been changed and this should have been
indicated in the initial pages itself. Policymakers, researchers, media and others tend to predominantly use the reports
rather than the raw data, and such changes in key indicators can lead to general confusion. In a bid to make the reports
available quickly, data accuracy should not be compromised. We have only pointed out one instance. One might assume
that there could potentially be other instances too, which may impinge on the overall credibility of published NFHS data.
48
Let us clarify that this is not necessarily a reflection on NFHS per se, and even as far as periodicity
is concerned, it is a matter of debate whether an independent survey should be conducted once
in 3 or 5 years. After all, there is a government survey (SRS) which provides annual estimates on
RCH / demographic indicators. Why do we need an independent survey also to provide data with
a similar periodicity? Let us also clarify that NFHS data is used at times by policymakers for regular
M&E as well, but to hold their subordinates accountable – they do not usually like it when they are
held accountable using it, hence, the outrage of some senior state officials against NFHS in some
of the not-so-well performing states.
Let us now turn to another NFHS objective, which was listed in the first place in its round 1 national
report and in the NFHS-5 bid document too – strengthening ‘survey research capabilities’ of PRCs
(round 1) and ‘Indian institutions’ (NFHS-5). As far as PRCs are concerned, all 18 of them were
involved in NFHS-1, then only 6 in NFHS-2 and -3, 3 in NFHS-4 and, eventually, none in NFHS-5.
On the other side, ‘increased involvement of commercial agencies for data collection’ has been a
matter of concern from the perspective of data quality (Srinivasan and Mishra 2020: 40). Neither
the survey research capabilities of PRCs seem to have neither been leveraged nor strengthened
in any significant way vis-à-vis the NFHS. As far as ‘Indian institutions’ are concerned, we do not
know for which institutions or in what ways have their capabilities been strengthened by the NFHS.
As far as use of NFHS data by researchers is concerned, it has been argued that ‘the four rounds
of the NFHS have produced an enormous quantity of data, which regrettably have been subjected
to only inadequate critical scrutiny by Indian scholars’, even as ‘the opposite is true of researchers
outside the country’ (Rajan 2020: 39). This is not necessarily a reflection on the NFHS itself, but it
should be explored as to what could be done to enhance the use of NFHS data for policymakers,
researchers and other stakeholders. On its website, NFHS highlights ‘worldwide media coverage’
of NFHS-4. The DHS Program website lists 24 analytical publications on the NFHS over a 20-year
period (May 1999 to July 2019).
59
This seems to be a gross under-estimation. On a quick check,
the Google Scholar shows around 21,600 results (search: “National Family Health Survey NFHS”).
Let us end this section by highlighting that its objective of providing ‘information in the context of
related socioeconomic and cultural factors’ (NFHS-2 introduction) seems to have been well-served
– even if from the limited perspective of RCH and a few other indicators – since NFHS is the only
data source in India providing richly disaggregated data by various background characteristics as
well as for households, status of women, etc. It is also the only source which provides data in such
a professional and transparent manner – government data sources, with the limited exception of
the NSS – are seriously lacking on this front. Even SRS forms were accidentally found online, and
we have never seen SRS’ raw data, for instance. None of this would, probably, have been possible
without regular involvement – both technically and financially – of relevant international agencies.
59
https://bit.ly/3cS3r6p (12/3/2020, 21:16 hours).
49
Table 2.3: Objectives of NFHS as outlined on NFHS website and in national reports of various rounds
Round Source(s) Objectives
NFHS website
60
‘Each successive round of the NFHS has had two specific goals’ –
a) ‘To provide essential data on health and family welfare needed by’
MoHFW ‘and other agencies
61
for policy and programme purposes’;
b) ‘To provide information on important emerging health and family welfare
issues’.
1
Multiple, in the
national report
Foreword by K B Pathak (Director, IIPS) –
a) The NFHS is ‘an important component’ to strengthen survey research
capabilities of the PRCs (the PRC project of MoHFW);
b) ‘Undertaken with the principal objective of providing state-level and
national-level estimates of fertility, infant and child mortality, the practice
of family planning, maternal and child health care and the utilization of
services provided for mothers and children’;
c) ‘Another important objective of the NFHS was to provide high quality
data to academicians and researchers for undertaking analytical
research on various population and health topics’;
d) ‘I do hope that it will contribute to the knowledge of researchers and
analysts in India and that programme administrators and policymakers
will find it useful for policy development and implementation of the family
welfare programme’.
Chapter 2 (survey design and implementation) –
a) ‘The primary objective of the NFHS is to provide national-level and state-
level data on fertility …’;
b) ‘This information is intended to assist policymakers, administrators and
researchers in assessing and evaluating population and family welfare
programmes and strategies’.
2
Multiple, in the
national report
Preface by T K Roy (Director, IIPS) –
a) NFHS-1 created ‘an important demographic and health database in
India’. NFHS-2 ‘is designed to strengthen the database further and
facilitate implementation and monitoring of population and health
programmes in the country’;
b) ‘As in the earlier survey, the principal objective of NFHS-2 is to provide
state and national estimates of fertility …’;
c) ‘We hope that the report will provide helpful insights into the changes
that are taking place in the country and will provide policymakers and
60
http://rchiips.org/nfhs/ (12/3/2020, 14:09 hours).
61
The NFHS-4 steering committee had representations from MoSPI, RGI, Ministries of Women & Child Development
and Drinking Water & Sanitation in addition to MoHFW and its erstwhile Department (now Ministry) of AYUSH from GoI.
50
programme managers with up-to-date estimates of indicators that can
be used for effective management of health and family welfare
programmes, with an emphasis on reproductive health dimensions’;
d) The report should also contribute to the knowledge of researchers and
analysts in the fields of population, health, and nutrition’.
Introduction –
e) ‘Another important objective is to examine this information in the context
of related socioeconomic and cultural factors’.
3
Multiple, in the
national report
Foreword by Naresh Dayal (Secretary, MoHFW) –
a) NFHS has provided ‘newer set of evidences of the ground realities’ to
help in policy- and program-making;
b) NFHS-3 has provided baseline information on RCH, nutrition, lifestyle
and HIV/AIDS related indicators in the context of RCH-2 and NACP-3;
c) ‘I hope’ NFHS-3 ‘would further strengthen India’s demographic and
health database’.
Preface by S Lahiri (Officiating Director, IIPS) –
a) NFHS is designed to be / ‘has emerged as a nationally important source
of data on population, health, and nutrition for India and its states’;
b) ‘The basic objective of releasing fact sheets within a very short period …
was to provide immediate feedback to planners and programme
managers on key process indicators’;
c) ‘We hope that the report will provide helpful insights into the changes
that are taking place in the country and will provide policymakers and
programme managers with up-to-date estimates of indicators that can
be used for effective management of health and family welfare
programmes, with an emphasis on both the reproductive and nutritional
health of the population’;
d) ‘The report should also contribute to the knowledge of researchers and
analysts in the fields of population, health, and nutrition’.
Introduction –
a) The MoHFW ‘initiated the NFHS surveys to provide high quality data on
population and health indicators. The three NFHS surveys conducted to
date are a major landmark in the development of a demographic and
health data base for India’;
b) NFHS-2 ‘was an important step in strengthening the database for
implementation of the Reproductive and Child Health (RCH) approach
adopted by India after the International Conference on Population and
Development (ICPD) in 1994 in Cairo’.
4
Multiple, in the
national report
Message by Preeti Sudan (Secretary, MoHFW) –
51
a) NFHS ‘indicators are more needed now than ever before to monitor the
progress of a number of flagship programs launched by the Government
of India in the recent past’;
b) ‘The NFHS-4 also brings to the forefront a number of emerging issues
that will occupy central place in the near future’;
c) ‘I hope’ NFHS data ‘will immensely help policy makers and programme
managers in planning focused policies and programmes’;
d) ‘I also hope that this report will be of great help to all those who are
working in the area of population and health’.
Foreword by Manoj Jhalani (Additional Secretary & Mission Director, NHM) –
a) NFHS has ‘played a crucial role in providing the Government of India
with reliable evidence on the success of its flagship programmes as
envisioned in the National Health Policy that aim to improve’ RCH ‘and
the health care delivery system in the country’;
b) NFHS-4 ‘will serve as a benchmark’ for government initiatives to achieve
SDGs by 2030;
c) ‘Over the years, the NFHS has expanded its scope and coverage to fill
the gap in the data required by the government, NGOs, and researchers
in the field of population and health’;
Message from Laishram Ladusingh (Officiating Director, IIPS) –
a) ‘These indicators available at the national, state and district levels shall
serve not only as benchmark for guiding the trajectory of health for all
but also as process indicators for a number of ongoing health
programmes’.
Introduction –
a) The main objective of NFHS-4 ‘is to provide essential data on health and
family welfare, as well as data on emerging issues in these areas’;
b) It is ‘intended to assist policymakers and programme managers in
setting benchmarks and examining progress over time in India’s health
sector. Besides providing evidence on the effectiveness of ongoing
programmes, NFHS-4 data will help to identify the need for new
programmes in specific health areas’.
5
Request for
Proposal: Bid
document
62
a) ‘The main objectives of the NFHS programme have been to strengthen
India’s demographic and health database by providing information that
is both reliable and relied upon’;
b) ‘To strengthen the survey research capabilities of Indian institutions to
provide, analyse, and disseminate high quality data’;
c) ‘To anticipate and meet the country’s needs for data on emerging health
and family welfare issues’.
Source: Developed by author.
62
http://rchiips.org/NFHS/NFHS5/pdf/Final%20RFP-Packer%20&%20Movers-NFHS-5.pdf (12/3/2020, 14:12 hours).
52
Scope
Table 2.4 highlights the logistical, respondent and thematic scope of various rounds of the NFHS.
Expectations from it have increased considerably over the years, and so has its scope and sample
size. The expectation to provide district-level data led to five- / six-fold increases in its sample size,
which has only been enough to develop district fact sheets with 93 indicators, covering population
and household profile, family planning, maternal and child health (including maternal and delivery
care, female examinations (breast, cervix, oral cavity), child immunization / disease prevalence /
treatment / feeding practices), nutritional status of children and adults, blood pressure and glucose
among adults. However, crucial RCH indicators such as IMR and TFR cannot be calculated, given
small NFHS sample size at the district level. Obviously, no district reports can be developed. The
increase in sample size has, nevertheless, meant that the duration from fieldwork to national report
increased from close to 2 years in the case of NFHS-2 and -3 to 3 years for NFHS-4, when district
level data was expected. Most importantly, it is not clear whether any assessment was done as to
what sort of data is required at the district, or central and state, levels before deciding the sampling
strategy. We will come to more on this in the section on methodology. The use of CAPI and geo-
referencing is said to have helped in monitoring of data collection and reducing turn-around time.
As far as the thematic scope of NFHS is concerned, it has been ‘designed to provide information
on sexual behaviour; husband’s background and women’s work; HIV/AIDS knowledge, attitudes,
and behaviour; and domestic violence’ (NFHS-4 national report: 1) and other RCH-related themes.
Men were included NFHS-3 onwards – but their sample size has gone down from 60% of female
sample size in NFHS-3 to 16% in -4 and 14% in -5. So much for gender equality, on a lighter note!
So, although the proportion of MCH questions has gone down from 99.2% in NFHS-1 to 92.3% in
-4, the MCH orientation of NFHS in terms of its sample size seems to have become stronger over
time. Yet, NFHS-4 ‘did not capture data for most known risk factors and commonly recommended
interventions for the improvement of newborn and maternal outcomes throughout the continuum
of care from pregnancy to the neonatal period, due to which, it was inadequate to inform relevant
policy and program (Dandona and Kumar 2019: 563-564). Further, maternal and neonatal causes
have only been one of the many health issues for women aged 15-49 years – their scope vis-à-vis
causes of death for this group declined from 16 to 6% during NFHS rounds (figure 2.2). In 2040,
according to a forecast, maternal and neonatal disorders would only be responsible for 0.79% of
all deaths in the country (Foreman et al 2018). True, that NFHS now covers their NCD risk factors
and self-reported status as well, but the proportion of NCD-related questions has only increased
from 0.8% in NFHS-1 to 4.7% in -4. But why blame NFHS alone for a predominant RCH orientation
– only 1.9% of the central release and 1.4% of total national expenditure under NHM was for NCDs
between 2005-06 and 2015-16 (figure 2.3). Communicable diseases weren’t treated much better.
It has been proposed that, with declining fertility rates, NFHS should focus on issues like infertility,
early menopause, rise in hysterectomies, children, adolescents and senior citizens (Rajan 2020).
53
Figure 2.2: Causes of death among females 15-49 years in India during NFHS rounds (1992-2017)
Source: GBD. Developed by author.
Figure 2.3: Central Release and expenditure under NHM, India, 2005-06 to 2015-16 (% of total)
Source: MoHFW. https://bit.ly/2O9ctBl (29/1/2020, 12:06 hours). Developed by Divya Chaudhry and Ali Mehdi.
16%
14%
11%
6%
6%
13%
13%
12%
18%
18%
8%
8%
9%
13%
14%
12%
12%
11%
11%
12%
14%
13%
12%
10%
10%
0% 20% 40% 60% 80% 100%
1992
1998
2005
2015
2017
Maternal disorders
Cardiovascular diseases
Neoplasms
Self-harm and interpersonal
violence
Respiratory infections and
tuberculosis
Unintentional injuries
Enteric infections
Digestive diseases
Chronic respiratory diseases
Diabetes and kidney diseases
Other infectious diseases
Transport injuries
HIV/AIDS and sexually transmitted
infections
Other non-communicable diseases
60.5
29.0
6.6
1.9
2.0
63.1
30.1
4.5
1.4
0.9
NRHM-RCH flexible pool
Infrastructure maintenance
Flexible pool for communicable disease control
programmes
Flexible pool for non communicable disease
programmes
National Urban Health Mission - flexible pool
ExpenditureCentral release
54
Let us end the discussion here by pointing out that although questions are added or removed over
time as per the prevailing requirements, it is important to ensure that comparability over its rounds
vis-à-vis core indicators in particular is not compromised so that we are able to track progress on
them. NFHS has had problems in this regard. Single NFHS reports provide comparable data, but
users need to be cautious while collating data from reports of various rounds.
63
This is a challenge
since policymakers, researchers, media-persons and others may not have the statistical capacity
to use raw data. Thankfully, the DHS Program offers online data visualization tools – for instance,
the STATcompiler – that help get comparable data over various rounds, and even other countries.
Cross-country comparisons is one of the major distinguishing features of the NFHS, which is rarely
used within India, with the exception of a few researchers. However, the STATcompiler does not
provide data on India-specific characteristics – for instance, caste / tribe and religion – or districts.
Neither the IIPS nor MoHFW have developed any user-friendly, online visualization tool to facilitate
access to comparable NFHS data. We had mentioned an online tool privately developed for NFHS
data at the national / state / district level, which one of the NHM officials in Kerala uses for ease of
access. He also said that it is very cumbersome to search for data in bulky NFHS reports.
Table 2.4: Selected characteristics of various rounds of NFHS
Details NFHS-1 NFHS-2 NFHS-3 NFHS-4 NFHS-5
Logistics
Geographical coverage
States / UTs /
Districts (marked ‘D’)
25 26 29
36,
640 (D)
36,
707 (D)
Fieldwork duration
Month / Year
04/1992-
09/1993
11/1998-
12/1999
11/2005-
08/2006
01/2015-
12/2016
06/2019-
06/2020
Months (numbers) 18 13 10 24 13
National report dated Month / Year 08/1995 10/2000 09/2007 12/2017 N/A
Time taken from fieldwork to
national report
Months (numbers) 24 10 13 12 N/A
Survey mode
Paper Yes Yes Yes
CAPI Yes Yes
63
For instance, U5MR went ‘up’ for ‘others’ in the caste / tribe category from 63.4 in NFHS-2 to 70.4 in -3, while it went
down from 81.5 to 46.6 during this period for SCs, in West Bengal (state reports). West Bengal was the only state in the
country where SCs had a lower U5MR than others as per the NFHS-2 and -3 state reports, which is difficult to fathom,
and could lead to skewed priority-setting in the state. It seems there are issues with NFHS data presentation in reports.
55
Details NFHS-1 NFHS-2 NFHS-3 NFHS-4 NFHS-5
GPS / geo-referenced Yes Yes
Respondents
Households Sample size 88,562 92,486 109,041 601,509 609,120
Females
Group category Ever married All
Age-group 13-49
64
15-49
Sample size 89,777 90,303 124,385 699,686 668,622
Males
Age-group
15-54 (married / unmarried)
Sample size 74,369 112,122 91,200
Themes
Disease- / condition-specific
questions
Number 246 294 694 868
N/A
Maternal and child
health (%)
99.2 97.3 93.3 92.3
NCDs (%) 0.8 1.7 3.2 4.7
Injuries (%) 0 1 3.5 3
Abortion Yes Yes Yes
Alcohol consumption Yes Yes Yes
Anemia
Question (Q) Yes Yes Yes
Test (T) – child (C),
men (M), women (W)
Yes Yes Yes Yes
Anthropometry (waist and hip
circumference also in NFHS-5)
Measurement (MT)
– C, M, W
Yes Yes Yes Yes Yes
Asthma Yes Yes Yes
64
The DHS Program website mentions that although data was collected for women aged 13-49 years during NFHS-1,
indicators were calculated for women aged 15-49 years, like the later rounds.
56
Details NFHS-1 NFHS-2 NFHS-3 NFHS-4 NFHS-5
Birth registration Yes Yes
Blood glucose (random testing
in NFHS-4, Hba1c in NFHS-5
through DBS)
Q Yes Yes
T – M, W Yes Yes
Blood pressure
MT – M, W Yes Yes
Q Yes
Causes of death Yes Yes
Child labor Yes Yes
Cooking fuel Yes Yes Yes Yes
Disability Yes
Domestic violence Yes Yes Yes Yes
Health expenditure Yes Yes
Health insurance Yes Yes
HIV
T (DBS / dried blood
samples) – M, W
Yes Yes
HIV behavior Yes Yes
HIV knowledge Yes Yes Yes Yes
Iodine salt test Yes Yes Yes
Lead testing Yes
Malaria (testing to help assess
the burden of malaria and
antimalarial drug resistance)
T (DBS) Yes
Q Yes Yes
Maternal mortality Yes Yes
Micronutrients Yes Yes Yes Yes
57
Details NFHS-1 NFHS-2 NFHS-3 NFHS-4 NFHS-5
Service availability Yes
Social marketing (contraception) Yes Yes
Tobacco use Yes Yes Yes
Tuberculosis Q Yes Yes Yes Yes
Verbal autopsy Yes
Vitamin A Q Yes Yes Yes Yes
Vitamin D3 T Yes
Women's examination (cervix,
breast, oral cavity)
Q Yes
Women's status Yes Yes Yes
Source: The DHS program website; Dandona, Pandey and Dandona 2016. Developed by author.
Table 2.5: NFHS-5 schedules
Schedule Themes
Household
1) All members of the household
2) Household characteristics
a. Drinking water
b. Sanitation
c. Ownership of assets
d. Cooking fuel
e. Mosquito net ownership and use
f. Hand washing facilities
g. Salt iodization
3) Socio-economic characteristics
a. Type of residence
b. Caste / tribe
c. Religion
58
d. Literacy / education
e. Employment / occupation
Women
1) Background characteristics
2) Gender issues, including domestic violence
3) Marriage and sexual activity
4) Family planning knowledge and use
5) Fertility and fertility preferences
6) Media exposure
7) Information on reproductive outcomes in calendar
8) Maternal / reproductive health (antenatal, delivery and postnatal care)
9) Nutrition (infant and young child feeding practices, micronutrient intake)
10) Beneficiaries of national GoI programs (for e.g. ICDS, JSY and JSSK)
11) Infant and child mortality
12) Child health (immunizations, prevalence of diarrhoea, fever, ARI and their treatment
seeking behaviour)
13) HIV/AIDS knowledge, stigma and discrimination, previous HIV testing
14) Tuberculosis
15) Non-communicable diseases
16) Medical injections
17) Smoking / drinking
18) Health insurance coverage
Men
1) Background characteristics
2) Gender issues, including domestic violence
3) Marriage and sexual activity
4) Family planning knowledge and use
5) Fertility and fertility preferences
6) Reproductive health
7) Infant and child mortality
8) HIV/AIDS knowledge, stigma and discrimination, previous HIV testing
9) Tuberculosis
10) Non-communicable diseases
11) Medical injections
59
12) Smoking / drinking
13) Health insurance coverage
Biomarkers
Measurement –
1) Height / length, weight, waist and hip circumference
2) Blood pressure
Testing –
3) Anemia
4) Blood glucose / HbA1C
5) Malaria (parasites and antimalarial drug resistance)
6) Vitamin D3
Source: NFHS-5 bid document.
Sample design
The NFHS follows a two-stage stratified sampling method. The Census 2011 formed the sampling
framework for selection of Primary Sampling Units (PSUs) during its 4th round, with the Census
Enumeration Block (CEB) in urban areas and villages in rural areas forming the PSUs. Each district
was stratified into rural and urban stratum. Rural substratum was further divided into smaller strata,
taking into account the village population as well as the percentage of scheduled population in the
village. Within each explicit rural sampling stratum, a sample of villages were chosen as the PSUs.
These PSUs were sorted according to the literacy rate of women aged above 6 years before their
selection. In urban areas, CEBs were sorted according to the percentage of SC / ST population,
and, thereafter, a sample of CEBs were selected through Probability Proportional to Size sampling.
In the second stage, 22 households per cluster were selected from the newly created list of
households living in the selected PSUs.
Sampling and non-sampling errors
We have highlighted sampling and non-sampling challenges of NFHS raised by our respondents
in the relevant sections, but let us discuss some of them here, starting with non-sampling errors.
NFHS includes sensitive themes (sexual activity / health, domestic violence, etc.), getting reliable
information on which is difficult. Interviewer are supposed to maintain privacy while interviewing
the respondents on such questions, but it has been noticed that household head as well as several
others are present during interviews. This is particularly a problem in rural areas, from where 71%
of respondents were in NFHS-4. Further, recall error and age-heaping are some of the major non-
sampling errors (Srinivasan and Mishra 2020; Rajan and James 2008), which is a serious problem,
60
once again, in rural areas. One of us was in one of the villages in Western Uttar Pradesh, wherein
he asked for the age of an old man. After ‘negotiation’, the old man settled on 70 years, down from
140 years! Respondents with higher educational levels are more likely to provide correct answers
compared to those with little or no education (Rajan and James 2004). As per NFHS-4, only 13.7%
of female and 20% of male respondents had 12 or more years of education at the national level.
These figures were lower in BIMARU states that have a much higher proportion of NFHS sample.
This also critically highlights the limits to capturing self-reported information in population surveys
in a predominantly rural and uneducated country like India, which is a particularly serious problem
from the perspective of health. Some questions on MCH pertain to 5 years preceding the survey,
increasing the possibility of recall errors. Answers are also often found to be affected by individual
/ social bias of the respondents or wrong interpretation of questions by respondents / interviewers.
It is difficult to statistically assess non-sampling errors or their effect on data quality. Due caution
needs to be exercised in the type of information that we wish to obtain through population surveys,
given the characteristics of potential respondents. It is not clear whether this was a consideration
for schedule / sampling design of the NFHS. One of our respondents said that during an interaction
with the NFHS team, the Chief Statistician of India said that he can understand that the team would
have taken due precautions vis-à-vis potential sampling errors, but what the non-sampling errors?
It is the latter that he said he is more concerned about as far as quality of NFHS data is concerned.
Another related concern has been the continuous disengagement of PRCs from the NFHS sphere,
despite its foremost objective being to strengthen their survey research capabilities, as highlighted
earlier, and the implications it potentially had on the quality of the data collected. ‘Assigning data
collection to consulting agencies that had no capacity building agenda by trained demographers
resulted in making the survey more of a money-making exercise than one focusing on delivering
reliable data. The biggest shortcoming in hiring private consulting agencies is the presence of
insufficient number of poorly trained and poorly paid field agents to collect data who receive little
logistical support and work under harsh conditions that lead to the violation of labour laws. Many
consulting agencies were operating without a local base in the states. This affects the quality of
data collection’ (Rajan 2020: 38). 4 field investigators also died (Karpagam and Sathyamala, 2015).
Sampling errors can be assessed statistically, involving complex formulas. The variance estimates
for important variables are provided by NFHS to evaluate the reliability of data. As its key focus is
to generate estimates on MCH, the sample selection criteria is based on characteristics of women,
which can lead to errors in information collected about men / other indicators. Further, NFHS also
collects information on alcohol, smoking, NCDs, etc., but its sample does not include institutional
populations in hostels, hospitals, etc. which are more prone to them. Some experts we interacted
with argued that a different sample strategy is required for different health concerns, while NFHS’
sampling is largely done from an RCH perspective. The discussion in the scope of NFHS section
is relevant here as well.
61
Length of the questionnaire
With each round, along with increase in sample size, the length of the questionnaire also increased
due to the expansion in the scope of the survey. There is a general consensus among researchers
that lengthy questionnaires result in poor quality of data due to hastening of the process to reduce
the time (Rajan 2020). From 610 questions in NFHS-1 (women schedule), the number of questions
has risen to more than 948 questions in NFHS-5. Inclusion of new areas along with elaboration of
particular topics have led to increase in its length. Further, wide variations in time taken to interview
respondents was also noticed across the states. While the average time taken was 86 minutes in
Tamil Nadu, it was 45 minutes in Haryana, for NFHS-3 women’s schedule (Rajan and James 2008).
NDQF has also argued that lengthy questionnaires and sensitive questions leading to non-response
or skipping of questions are important factors affecting data quality. Further, it is to be noted that
many of the questions may not be applicable to all the household as it depends on the occurrence
of events, like births, immunizations, etc. during five years preceding the survey. The involvement
of too many ministries / stakeholders has meant that everyone wants their questions included, and
there is little scope for independent assessment regarding the feasibility / rationality of questions
to be included – everyone has to be satisfied. Some of the same set of stakeholders will also pull
up NFHS organizers if the data it throws up does not match their ‘expectations’ or is inconvenient.
Data validation
Let us end the discussion of the NFHS here with a comparison of data on some of the indicators
from NFHS-4 and HMIS (2015-16), which matches with the periodicity of NFHS-4. Table 2.6 offers
data on some of the rare NFHS-4 and HMIS indicators which are broadly comparable. Once again,
Kerala’s figures match perfectly as far as home and institutional deliveries are concerned, hinting
toward a potential correlation between the robustness of administrative data and health outcomes.
These are major divergences as far as other, particularly health-backward, states are concerned,
adding more weightage to the correlation. In the case of C-section deliveries, however, while India
and Rajasthan figures match quite closely, there are substantial divergences in the case of other
states, including Kerala. In the case of TT injections, divergence in the case of Kerala is the highest,
while figures for India, Rajasthan, Uttar Pradesh and Maharashtra are close by. Does the potential
correlation stand nullified? Given some of the challenges vis-à-vis quality of NFHS data highlighted
above, it is difficult to say. Nevertheless, harmonizing the definitions of some of the core indicators
of NFHS and HMIS would help in more robust and extensive comparisons and may help both data
sources and their organizers to be more vigilant vis-à-vis data quality.
62
Table 2.6: Data comparison of selected indicators from HMIS (2015-16) and NFHS-4 (2015-16)
Source: HMIS and NFHS reports. Developed by Priyanka Tomar.
District Level Household and Facility Survey (DLHS, 1998-2013)
DLHS was initiated by MoHFW – with IIPS as the nodal agency – to provide RCH data at the district
level. After 4 rounds – 1998-99, 2002-04, 2007-08 and 2012-13 – it was discontinued due to reasons
highlighted in the introduction. It also generated data on utilization of health services and people’s
perception about the quality of the services provided. It was the only survey which provided data
on the quality of government health facilities.
The 1st round was conducted in 2 phases and the key objective was to gather information on ANC
and immunization services, deliveries, contraceptive use, awareness about RTI/STI, HIV/AIDS,
family planning, utilization of government health services, user satisfaction, incidence / prevalence
of malaria, leprosy, tuberculosis, information on childbirth, maternal health, unmarried adolescents’
counselling by ANMs on reproductive health issues and the management of anaemia. It covered
529,817 households, 474,463 currently married women aged 15-44 years and 257,245 men aged
20-54 years in 504 districts. Providing district-level data and covering men were both firsts for an
RCH survey in the country, both of which were incorporated in the NFHS subsequently.
DLHS-2 was conducted across 593 districts, covering 620,107 households, 507,622 women aged
15-44 years and 330,820 husbands of eligible women. Beyond DLHS-1, it also collected data on
iodine intake through testing of cooking salt used by households, biometric and anthropometric
measures (measurement of weight of children, assessment of anaemia levels through blood tests
of children, adolescents and pregnant women), and had 3 additional questionnaires – husband’s,
village and health – in addition to household and women’s questionnaires canvassed in DLHS-1.
Inputs from DLHS-2 helped in the design of NRHM, launched in 2005-06. In order to monitor and
assess its impact, the third round of DLHS was launched in 2007-08 to collect data on utilization
of various health care services, accessibility to health services, effectiveness of ASHA and JSY in SourceIndicatorIndia Rajasthan Uttar Pradesh Bihar Assam Maharashtra Kerala
HMIS Home deliveries to total reported deliveries (%)11.1 3.8 22.3 22.1 14.1 1.3 0.2
NFHS-4Home deliveries (%)20.8 15.8 31.8 35.9 29.2 9.6 0.1
HMIS Institutional deliveries to total reported deliveries (%) 88.9 96.2 77.7 77.9 85.9 98.7 99.8
NFHS-4Births delivered in a health facility (%)78.9 84 67.8 63.8 70.6 90.3 99.8
HMIS C-section deliveries to reported institutional deliveries (%) 17.3 9.9 4.2 2.9 18.5 14.8 41.4
NFHS-4Births delivered by caesarean section (%)17.2 8.6 9.4 6.2 13.4 20.1 35.8
HMIS Women received TT2+ TT Booster to total ANC registration (%) 82.8 82.3 83 88 90.8 84 81.3
NFHS-4Women received two or more TT injections during the pregnancy (%) 83 81.9 81.4 81.5 83.6 81.4 94.8
63
improving the health scenario in the country and condition of health infrastructure along with usual
information. It also tried to assess the linkages between RCH indicators and health facilities. Some
major changes were also introduced in the third round. In the last two rounds, the survey covered
currently married women and men, while in the third round, ever married women in the age group
of 15-49 years and unmarried women in the age group of 15-24 years were interviewed. A total
of 720,320 households, 643,944 ever-married women and 166,620 unmarried women across 601
districts 34 states / UTs were covered in the third round using 5 questionnaires – household, ever
married women’s, unmarried women’s, village and health facility questionnaires, the latter covering
all CHCs and district hospitals at the district level and all SCs and PHCs expected to serve selected
primary sampling unit (PSU) populations.
The fourth and final round did not cover Empowered Action Group (EAG) states (Bihar, Rajasthan,
Madhya Pradesh, Jharkhand, Chhattisgarh, Uttarakhand and Uttar Pradesh) and Assam since they
were covered under RGI’s Annual Health Survey. The discontinuation of DLHS in 9 states meant
it could not provide national estimates. At the state level as well, it only provided fact sheets rather
than reports as earlier. DLHS was discontinued and the scope of NFHS-4 was expanded to provide
RCH data up to the district level.
Global Youth Tobacco Survey (GYTS, 2000-)
Three rounds of GYTS have been carried out in the country as part of Global Tobacco Surveillance
System (GTSS) from 2000-05 at the state level, and in 2006 and 2009 at the national level. IIPS
was appointed as the nodal agency by MoHFW for GYTS-4, using the Unified-District Information
on School Education (U-DISE) 2017-18 database for its sampling. A school-based survey, GYTS
uses a standardized instrument – developed by WHO, UNICEF and CDC – to collect information
on tobacco use among school going children aged 13-15 years, studying in grades 8 to 10.
65
During 2003-05, individual surveys were carried out in 28 states / UTs. Data was compiled from
all the states, applying weights to independent samples to produce a weighted national estimate.
For the 2006 phase, while the sampling procedure remained the same, samples were drawn from
6 independent geographical regions consisting of contiguous states, to save time and budget. The
2009 survey also followed the same sampling method with minor modifications. Uttar Pradesh and
Rajasthan were included in the Central region to mirror the composition of Global Adult Tobacco
Survey, which was also carried out in the same year. The sample size of the survey was 11,768
students. Questionnaire for all the three surveys were self-administered and collected information
on prevalence of tobacco use (smoke and smokeless), access and availability of tobacco products,
perceptions and attitudes about tobacco, exposure to second-hand smoke and smoke-cessation.
65
https://bit.ly/2TPp0x3 (15/3/2020, 22:35 hours).
64
National Behavioural Surveillance Survey (NBSS, 2001-06)
Behavioural surveillance is one of the important tools to understand the levels of knowledge and
awareness about HIV/AIDS, sexual behaviours and attitudes. Therefore, to provide a baseline for
the interventions of National AIDS Control Program (NACP), a behavioral surveillance survey was
conducted by the National AIDS Control Organization (NACO) and UNICEF in 2001, covering the
15-49 year population. However, data for 15-24 years, which was disaggregated from the survey,
did not have a sufficient sample size for representative estimates and, therefore, NBSS-2 (2006)
had a separate sample of 15-24 year-olds beyond general population. NBSS-2 aimed to measure
changes in the knowledge and attitudes of the youth vis-à-vis NBSS-1 and help in the expansion
of interventions to reduce transmission. It covered a sample of 97,240 respondents (15-49 years)
from 2,434 PSUs across 25 states, with smaller states and UTs merged with the larger ones. In
each selected PSU, a sample of 40 respondents (20 male and 20 female) was interviewed for the
general population along with additional 20 respondents (10 male and 10 female) in the age group
of 15-24 years for the youth survey. The total sample covered for the youth was 78,916 – 30,791
from the general population survey and 48,125 additionally covered. With inclusion of knowledge
and awareness about HIV/AIDS in NFHS-3, NBSS was discontinued.
Global Adult Tobacco Survey (GATS, 2009-)
GATS is a nationally representative survey of persons aged 15 years and above, using a standard
protocol across countries to monitor adult tobacco use and key tobacco control indicators. GATS
is supposed to enhance technical capacity of countries in designing, implementing and assessing
the impact of tobacco control initiatives. It also captures socio-economic determinants influencing
tobacco use, and its data is considered useful for making projections about tobacco-related health
and economic consequences. Further, GATS has enabled countries in achieving their obligations
under WHO’s Framework Convention on Tobacco Control (FCTC) to generate comparable data
within and across countries. Such features have made GATS one of the most powerful instruments
that countries can deploy to support tobacco cessation programs and curtain tobacco use (GATS
2016-17, 2010). As India is world’s third largest tobacco producer and second largest consumer,
GYTS and GATS have been critical for the country. The first round of GATS was implemented in
2009-10 in 31 states / UTs, covering a sample of 69,296 adults, the second in 2016-17 in 32 states
/ UTs, with a sample size of 76,500 adults.
Annual Health Survey (AHS, 2010-13)
While DLHS was unique in covering districts, men and health facilities, the AHS was conceived in
2005 and launched in 2010-11 to provide annual estimates on RCH as well as other indicators for
284 districts of EAG states and Assam. These were high-priority states, which together accounted
65
for around – 50% of India’s population, 61% of births, 71% of infant deaths, 72% of under-5 deaths
and 62% of maternal deaths. AHS was the world’s largest household sample survey, covering 4.1
million households in its first round (2010-11), 4.2 million in the second (2011-12), 4.3 million in
the third (2012-13) and a population of nearly 18 million. Despite its RCH-orientation – presenting
a district-level index of maternal and child health deprivation – it collected data on the prevalence
of chronic and acute illness, disability, injury as well as health care utilization (AHS report, Vol. 1).
It also had a clinical, anthropometric and biochemical (CAB) component in a sample of 1.8 million
population in 360,000 households – height / weight of all members of the household, women, men
and children 1 month and above; Hb estimation of women, men and children aged 6 months and
above; fasting blood glucose and blood pressure of all members of household aged 18 years and
above and household salt testing for iodine content.
66
National Anti-TB Drug Resistance Survey (NATDRS, 2014-)
Although India has achieved significant gains in the treatment of TB, it contributes 27 percent to
the global burden of tuberculosis (TB). Roughly, 2.79 million new TB cases are reported annually.
Moreover, rapidly emerging evidence of drug-resistant (DR) strains of TB pathogen is jeopardizing
the national progress achieved in TB control. Patients with DR-TB fail to respond to rifampicin –
the most effective first-line drug therapy for TB. MDR-TB is another form of TB infection in which
the pathogen is resistant to at least two of most powerful anti-TB drugs (rifampicin and isoniazid).
Further, extensively drug-resistant TB (XDR-TB) is an extreme form of MDR-TB that is resistant to
isoniazid and rifampicin, at least one fluoroquinolone (antibiotics used to treat or prevent bacterial
infections) and at least one of three injectable second-line anti-TB drugs (amikacin, capreomycin,
kanamycin) (WHO 2017). While prevalence of XDR-TB is currently low worldwide, the estimated
prevalence of MDR-/rifampicin resistant (RR)-TB in India is 147,000 – accounting for a quarter of
the global burden of MDR-/RR-TB.
In order to investigate the epidemiology of DR-TB and estimate the prevalence of DR among TB
patients in India, the first ever NATDRS was conducted by MoHFW and WHO India between July
2014 and July 2015. This is the world’s largest DR survey ever conducted and first ever to include
drug-susceptibility testing for 13 anti-TB drugs, making use of the automated liquid culture system
and mycobacteria growth indicator tube (MGIT) 960. A total of 5,280 sputum smear+ pulmonary
TB patients diagnosed at designated microscopy centres (DMCs) of RNTCP between August 2014
to July 2015 were enrolled in the survey. Despite these merits, the NATDRS had a major limitation
– patients treated in the private sector who did not seek care in public health facilities during their
treatment course remained out of the scope of the survey, and hence, accurate prevalence rates
of DR-TB could not be estimated (MOHFW, WHO and USAID 2014-16).
66
https://bit.ly/2voCZAH (15/3/2020, 21:06 hours).
66
National Mental Health Survey (NMHS, 2015-16)
Given the historical focus on fertility and mortality, and recently on physical health / NCDs – not
unjustified in the context of poverty, population growth and the enormous burden of premature,
especially child, mortality in the country – mental health has long been a neglected area, and it is
only since 2014 that some serious, yet limited, efforts have been made to deal with it. The National
Mental Health Policy – themed ‘New Pathways, New Hope’ – was released in October 2014. Soon
thereafter, the first National Mental Health Survey, 2015-16 was commissioned to address the lack
of reliable and comprehensive data on mental health. The prestigious National Institute of Mental
Health and Neurosciences (NIMHANS), Bengaluru undertook the survey across 12 selected states
– 1) to estimate prevalence and burden of mental health disorders in a representative population,
2) to identify current treatment gap, health care seeking and service utilisation patterns, disability
status and impact of mental disorders, and 3) assess mental health care facilities, resources and
systems in the surveyed states for planning and strengthening of mental health services.
67
Its total
sample size included 39,532 individuals across 720 clusters from 80 talukas in 43 districts of 12
states. Employing both quantitative and qualitative methods, all individuals 18 years and above in
selected households were interviewed. The qualitative component of the survey included questions
related to drug use / abuse characteristics, region- / state- / area-specific mental health problems,
stigma towards mental health problems and mental health care-seeking patterns, etc. No second
round of the survey is being planned. While the sample size consists of individuals above the age
of 18 to ensure correctness of information, survey of adolescents (13-17 years) was also carried
out alongside in the states of Gujarat, Uttar Pradesh, Jharkhand and Tamil Nadu. The tools used
in the survey consisted of Mini International Neuropsychiatric Interview Schedule (MINI), jointly
developed by European and American psychiatrists, a socio-demographic questionnaire, tobacco
use and dependence questionnaire, screeners for epilepsy, intelligence deficiency and autism
spectrum disorders, pathways to care and disability assessment schedule (Gururaj & Collaborators
2016). . A salient feature of the NMHS is the collection of data according to ICD-10 classification
which renders its analysis useful in global context. Further, it included all types of mental disorders
(including epilepsy) which are of public health importance along with substance abuse. The NMHS
not only provides information about the extent and patterns of mental disorders / substance abuse,
but also service utilization patterns and gaps in infrastructure and manpower. While the first round
provides robust nationally representative data, the second phase should be conducted across all
states for representative data at the state level as well.
67
http://indianmhs.nimhans.ac.in/Docs/Report2.pdf (25/10/2019, 16:38 hours).
67
Comprehensive National Nutrition Survey (CNNS, 2016)
Launched in 2016 to collect reliable data on various domains of nutritional health for the age group
of 0-19 years, the CNNS was conducted in 30 states, covering a sample of 112,316 children and
adolescents under the stewardship of MoHFW, in collaboration with UNICEF and the Population
Council. The surveyed population was divided into three age groups of pre-schoolers (0-4 years),
school-going children (5-9 years) and adolescents (10-19 years) – with a sample size of 38,060,
38,405 and 35,856 respectively. For children below 10 years, the respondents were the heads of
the households or parents of the child; those aged 10-19 years were themselves the respondents.
Along with information on anthropometric status, anemia and iron deficiency, it also provided data
on micronutrient consumption, infant and young child feeding practices and risk factors for NCDs
– like glucose concentration, lipid profile and physical fitness – becoming the first survey to provide
such data for its age-groups. Unlike NFHS-4, in which random blood glucose samples were tested,
CNNS used fasting plasma glucose and glycosylated hemoglobin (HbA1c) (MoHFW, UNICEF and
Population Council 2019). The survey also has some limitations despite its comprehensive nature
and robust data quality. Data is available only at the state level. Furthermore, the most important
limitation of the survey is that disaggregated analysis cannot be conducted, limiting the ability to
understand the underlying causes. Overcoming these limitations in phase 2 should be considered.
Longitudinal Ageing Study in India (LASI, 2016-)
Even though adult health and ageing have garnered significant attention in international discourse,
there is a serious dearth of comprehensive and comparable survey data on the economic, social,
and public health implications of ageing in India. In order to address this lacuna, the pilot wave of
LASI was launched in India in 2010. The MoHFW appointed IIPS as nodal agency for conducting
LASI surveys in India. LASI team in IIPS has collaborated with the Harvard School of Public Health
(HSPH) and University of Southern California for technical support. The most important aspect of
LASI is its longitudinal character which will enable researchers to analyze the dynamics of India’s
ageing population and inform policy decisions. Like NFHS, despite India-specific characteristics,
LASI has been developed to be consistent with other international ageing surveys – for instance,
the Health and Retirement Study, the Chinese Health and Retirement Longitudinal Study, the
Japanese Study on Aging and Retirement and the Korean Longitudinal Study of Aging
68
– with the
expectation that ‘it will contribute to scientific insights and policy development in other countries’
as well and vice-versa. Its survey instrument is, therefore, internationally harmonized. The pilot
68
http://iipsindia.org/research_lasi.htm (25/10/2019; 13:00 hours).
68
phase of the survey was supported by a grant from the National Institute on Aging (NIA), National
Institutes of Health (NIH), Health and Human Services Department, US government.
69
The LASI pilot study was carried out in 4 Indian states – Punjab, Rajasthan, Karnataka and Kerala
– with a targeted sample of 1,600 non-institutionalized Indian residents aged 45 or older and their
spouses (irrespective of age). LASI survey instrument comprises of a household survey (one per
household collected by interviewing a selected key informant), an individual survey (one for every
age-eligible respondent of at least 45 years of age and their spouse) and a biomarker module (one
for every consenting age-eligible respondent and spouse). In addition to recording demographic
composition of the household, the household survey has questions on residential history, physical
and social characteristics of neighbourhood, household consumption, assets and debt, income of
all household members from all sources and coverage under public and private health insurance
schemes. The individual survey is more detailed, comprising elaborate modules on demographics,
family and social welfare networks, awareness and utilization of social security schemes, health
(overall health, specific diseases, functional health, family medical history, mental health, including
cognition and depression, etc.), health care access and utilization, work, retirement and pension.
There is a biomarker module in the health section of the individual survey that collects information
on anthropometrics, blood pressure, dried blood spots and performance measures like gait speed,
grip strength, balance, lung function and vision.
70
LASI’s first full-scale, nationally representative survey was launched in 2016, covering a sample
size of 60,250 eligible individuals in 36 states / UTs. The results of the survey are still awaited. The
LASI team will follow these individuals over time and survey them once every 2 years (Onur and
Velamuri 2018). It is aimed to continue the survey at this scale for the next 25 years. Although
LASI is the first of its kind, a number of challenges are yet to be addressed. For instance, problems
like lack of documentation, self-production and consumption, limited number of transactions in
market contexts, etc. make it virtually impossible to estimate income or assets. Furthermore, some
people may be reluctant to share certain information with surveyors. For instance, women often
hesitate to reveal information about their savings because they fear that their husband / children
/ sons-in-law might claim it.
71
National NCD Monitoring Survey (NNMS, 2017-)
With the passing of the World Health Assembly resolution 66.10, India became the first country to
develop its National NCD Monitoring Framework with country-specific 10 targets and 21 indicators
to be achieved by 2020 / 2025. In consultation with all the relevant stakeholders, MoHFW further
69
https://www.hsph.harvard.edu/pgda/major-projects/lasi-2/ (24/10/2019, 10:35 hours).
70
https://www.ncbi.nlm.nih.gov/books/NBK109220/; http://iipsindia.org/research_lasi.htm (25/10/2019, 13:00 hours).
71
https://www.ncbi.nlm.nih.gov/books/NBK109220/ (25/10/2019, 13:05 hours).
69
developed National Multisectoral Action Plan (NMAP) for prevention and control of common NCDs
(2017-22). To monitor progress on India’s NCD targets and indicators, NNMS has been conducted
by ICMR’s National Center for Disease Informatics and Research (NCDIR), Bengaluru with support
from MoHFW.
72
The first round of the survey was initiated in 27 states in October 2017, with 2010
treated as baseline to evaluate progress made. It covered primary and secondary health facilities,
and 1 adult (18-69 years) and all adolescents (15-17 years) from selected households. The survey
included questions on NCD risk factors – tobacco consumption (smokeless and smoking), harmful
consumption of alcohol, dietary habits, salt intake, physical measurements, activities, body mass
index, fasting blood sugar and blood pressure. Not much is known about it at this stage since its
results / report have not been made public so far – allegedly, pending government clearance. This
would be the first full-fledged NCD survey in the country.
72
http://ncdirindia.org/ncd_dashboard/documents/AboutNNMS_2017_18.pdf (25/10/2019, 16:35 hours). 70
Table 1.7: An overview of active population health surveys in India
Survey
(initiation year)
Coordinating
Ministry
Population covered and sample
size (latest round)
Representativeness
(latest round)
Disaggregation
(latest round)
Periodicity
Rounds
completed
National
Sample Survey
(1952)
MoSPI
Social Consumption: Health
(SCH) – 75th round (2017-18) –
113,823 households, 555,115
persons
Disability – 76th round (2018) –
118,152 households, 576,569
persons
Household Consumer
Expenditure Survey (CES) – 68th
round (2011-12) – 101,662
households
National, state
Gender, rural /
urban, age
groups, education,
religion, caste,
quintile class,
occupation,
marital status
SCH and
disability –
Non-periodic
CES –
Quinquennial
SCH – 6
Disability – 4
CES – 9 (data of
10th CES was
not released)
73
Sample
Registration
System (1971)
MHA 2017 – 7,925,000 persons
National, state
(major)
Gender, rural /
urban, broad age
groups, education,
marital status
Annual / 3-
yearly / 5-
yearly
N/A
National Family
Health Survey
(1992)
MoHFW
NFHS-4 (2015-16) – 601,509
households, 699,686 women
(15-49 years), 112,122 men (15-
54 years)
National, state,
district (selected)
Gender, rural /
urban, education,
religion, caste /
tribe, wealth
index, occupation,
marital status
Non-periodic 4
73
‘In view of the data quality issues, the Ministry has decided not to release the Consumer Expenditure Survey results of 2017-2018. The Ministry is
separately examining the feasibility of conducting the next Consumer Expenditure Survey in 2020-2021 and 2021-22 after incorporating all data quality
refinements in the survey process’. https://pib.gov.in/Pressreleaseshare.aspx?PRID=1591792 (1/6/2020, 11:52 hours).
71
Survey
(initiation year)
Coordinating
Ministry
Population covered and sample
size (latest round)
Representativeness
(latest round)
Disaggregation
(latest round)
Periodicity
Rounds
completed
Global Youth
Tobacco
Survey (2000)
MoHFW
2009 – 10,112 students (13-15
years, in grades 8-10)
National, state
Gender, age,
grade
Non-periodic
4 (4th round,
2017-18, data
not released)
Sample
Registration
System – cause
of death survey
(2001)
MHA 2010-13 – 182,827 deaths
National, 2 state
groupings
Gender, rural /
urban, broad age
groups, education,
religion,
occupation
Non-periodic 4
Global Adult
Tobacco
Survey (2009)
MoHFW
GATS-2 – 33,772 men and
40,265 women (15+ years)
National, state
Gender, rural /
urban, age
groups, education,
religion, caste /
tribe occupation,
marital status
Non-periodic 2 rounds
National Anti-
Tuberculosis
Drug
Resistance
Survey (2014)
MoHFW
5,280 sputum smear-positive
pulmonary TB patients
diagnosed at the designated
microscopy centres (DMCs) of
RNTCP between August 2014 to
July 2015
Unknown Gender, age Unknown
74
1
National Mental
Health Survey
(2015)
MoHFW
39,532 (18+ years, including 13-
17 years in 4 states)
National, 12 states –
Assam,
Chhattisgarh,
Gujarat, Jharkhand,
Kerala, Madhya
Pradesh, Manipur,
Gender, rural /
urban, age
groups, education,
occupation,
Unknown 1
74
We have written ‘non-periodic’ against those surveys for which more than 1 round has been conducted and the periodicity is not fixed / clear, and ‘unknown’ against
those where periodicity is not know since only 1 round has been conducted so far.
72
Survey
(initiation year)
Coordinating
Ministry
Population covered and sample
size (latest round)
Representativeness
(latest round)
Disaggregation
(latest round)
Periodicity
Rounds
completed
Punjab, Rajasthan,
Tamil Nadu, Uttar
Pradesh, West
Bengal
marital status,
income quintile
Comprehensive
National
Nutrition
Survey (2016)
MoHFW &
UNICEF
112,316 children and
adolescents (0-19 years)
National, state
Gender, rural /
urban, age
groups, mother’s
age and
schooling,
religion, caste /
tribe, wealth
index, occupation,
marital status
Unknown 1
Longitudinal
Aging Study in
India (2016)
MoHFW 60,250 individuals (45+ years) National, state
Gender, rural /
urban, education,
religion, caste,
wealth index,
occupation,
marital status
Unknown
1 (data not
released)
National Non-
communicable
Disease
Monitoring
Survey (2017)
MoHFW
300 urban and 300 rural PSUs –
1 adult (18-69 years) and all
adolescents (15-17 years)
eligible from a household
National, state Rural / urban Unknown
1 (data not
released)
Source: Developed by Ali Mehdi, Priyanka Tomar and Divya Chaudhry.
73
Table 2.8: Thematic mapping of selected population health surveys (latest rounds) in India
Themes
Population
covered
NFHS SRS
SRS
CoD
NNMS NMHS GATS CNNS LASI NSS NATDRS
Demographics
Birth record (live births)
Birth record (non-live births)
75
Fertility indicators
76
Death record
Child
77
Maternal
Others
Mortality indicators
Child
78
Maternal
Others
79
Disability
80
75
NFHS: non-live births include abortion, miscarriage or stillbirth. SRS: still birth and abortion.
76
NFHS: CBR, TFR, ASFR, age at birth of first child, wanted fertility rate, birth order and birth interval. SRS: CBR, general fertility rate, ASFR, TFR, gross reproduction
rate, general and total – marital fertility rates, mean age at effective marriage for females, birth order and birth interval.
77
SRS CoD: neonatal death (28 days or less), child death (29 days to 14 years).
78
All types of child mortality rates.
79
NFHS and SRS: CDR, ASDR. NFHS also provides adult mortality rates in two categories – deaths due to non-medical reasons (accidents, violence, poisoning, homicides
or suicides) and deaths due to other reasons.
80
NSS: locomotor, visual, hearing, speech and language, mental retardation / intellectual disability, mental illness, other. NFHS-5: hearing, speech, visual, mental,
locomotor, other. NMHS: intellectual, self-reported disability across work, social and family life. LASI: self-reported ‘difficulty with at least one activity of daily life (ADL)’.
74
Themes
Population
covered
NFHS SRS
SRS
CoD
NNMS NMHS GATS CNNS LASI NSS NATDRS
Cause of death
Verbal Autopsy
HH-reported
81
Morbidity
82
Communicable diseases
83
Children,
men, women
NCDs
84
Adolescents,
men, women
Mental health
Injuries
Maternal disorders
85
Determinants of health (A)
Residence (rural / urban)
81
NFHS: maternal, injuries.
82
SRS CoD captures morbidity of the deceased, as reported by the household respondent.
83
NFHS: under-5 children (self-reported – acute respiratory infections, fever, diarrhoea), men and women (testing – HIV, self-reported – tuberculosis). SRS: jaundice,
HIV/AIDS, tuberculosis (SRS Form No. 2A: morbidity, personal habits and socio-economic status). SRS CoD: diarrhoea, cough, measles and other sickness at the time
of death (SRS VA Form 10A and 10B for child), past history of HIV/AIDS and tuberculosis’ (SRS VA Form 10C for adult). NATDRS: MDR-TB/XDR-TB among new and
previously treated TB patients.
84
NFHS: diabetes, hypertension, chronic respiratory diseases including asthma, goitre or any thyroid disorder, any heart disease, cancer, any chronic kidney disease.
SRS: diabetes, asthma, hypertension, cancer, coronary heart diseases, others (SRS Form No. 2A: morbidity, personal habits and socio-economic status). SRS CoD: past
history of hypertension, heart disease, stroke, cholesterol, diabetes, cancer, asthma and other chronic illness (SRS VA Form 10C for adult).
85
NFHS: abortions, convulsions, vision problem, swelling, post-partum complications (self-reported). SRS CoD: excessive bleeding, prolonged labour, fits or loss of
consciousness during pregnancy / during labour or after labour, fever after birth, foul smelling discharge.
75
Themes
Population
covered
NFHS SRS
SRS
CoD
NNMS NMHS GATS CNNS LASI NSS NATDRS
Socioeconomic
determinants
(A1)
86
Education
87
Marital status
Religion
Caste / tribe
Wealth index / household
expenditure or income
quintiles
Drinking water facilities
Sanitation facilities
Occupation / employment
status
Ownership of agricultural
land, house and farm
animals
Health insurance
Out-of-pocket
88
Individual risk factors (A2)
86
CNNS also includes type of mother’s diet (vegetarian, vegetarian with egg, non-vegetarian).
87
SRS: women’s level of education.
88
NFHS: average out-of-pocket cost paid for delivery. NMHS: amount spent for care and treatment of mental disorders.
76
Themes
Population
covered
NFHS SRS
SRS
CoD
NNMS NMHS GATS CNNS LASI NSS NATDRS
Metabolic risk
factors (A2.1)
Anthropometric indicators
Child
Adolescents
Men, women
Haemoglobin
Child
Adolescents
Men, women
Blood pressure
Child
Adolescents
Men, women
Blood glucose
Child
Adolescents
Men, women
Behavioral risk
factors (A2.2)
Tobacco consumption /
smoking
Alcohol consumption
Level of physical activities
Dietary habits Child
77
Themes
Population
covered
NFHS SRS
SRS
CoD
NNMS NMHS GATS CNNS LASI NSS NATDRS
Overall
89
Access to health
care (A3)
RCH
Communicable diseases
NCDs
Mental health
Injuries
General
Source: Developed by Priyanka Tomar and Ali Mehdi.
89
SRS CoD: diet of deceased (pure vegetarian or not). 78
Table 2.9: Thematic mapping of NSS health survey (75th round, 2017-18) and NFHS-4 (2015-16)
Sn. Indicators
NSS (75th
round, 2017-18)
NFHS-4
(2015-16)
1 Proportion of ailing persons
2 Ailments
3 Treatment-seeking behaviour
4 Proportion of in-patient treatment
5 Nature and characteristics of in-patient treatment
6 Average out-of-pocket expenditure
a In-patient treatment
b Other ailments
7 Expenditure on childbirth by
a Type of hospital
b Nature of delivery
c Household expenditure class
9 Childbirths involving surgery
10 Children receiving any vaccination
11 Fully immunized children (0-5 years)
12 Average expenditure on immunization
13 Condition of the aged (living arrangements, physical mobility)
14 Health insurance
15 Household and socioeconomic characteristics
Notes: NFHS collects data on selected ailments, treatment-seeking behavior and out-of-pocket expenditure, while
NSS collects comprehensively.
Source: Developed by Priyanka Tomar.
79
Recommendations
® Even if we do not agree with the Bhore Committee Report’s self-description as ‘a broad survey’,
a similar report could be brought out annually / biennially by DoHR. The CBHI’s National Health
Profile does provide a broad range of statistics on an annual basis, but it has almost no analysis
and certainly no ‘recommendations for future development’ of the health system / HIS. DoHR
is capable of doing this. The fragmentation of the health system / HIS is reflected in discussions
as well. The proposed report might help develop a health systems approach, even if notionally,
to begin with. Like the Bhore and Sokhey Committee Reports, it should take a comprehensive
view of health and its determinants and go beyond health systems / HIS vis-à-vis its approach,
evidence and recommendations.
® NSSO was the pioneer in health statistics in the country. It still continues to provide some very
helpful health statistics (tables 2.7 and 2.8). Beyond the health-related data that it collects, the
NSO should focus on the coordination and consistency of health data collection at the national
and state / UT levels. This would be a far bigger service than the data that it produces. To lead
by example, it should not collect any data which is already being collected through any existing
source (tables 2.8 and 2.9). For health-related data that it still collects, it should consult health
experts to ensure that there are no conceptual or methodological issues with its data. Among
the examples where it has faltered is vis-à-vis its data on the utilization of different systems of
medicine as part of its 71st round (‘Key indicators of social consumption in India: Health’. NSS
2014). Since people use these systems of medicine more on a complementary rather than an
alternative basis, we should not ask about their utilization in exclusive terms. Not surprisingly,
AYUSH stakeholders have not been too happy with these statistics, especially because this is
the only national survey data that exists on this issue. MoSPI also refused to entertain a request
from Ministry of AYUSH to conduct a survey for it (respondent), and the latter was desperately
looking for alternatives. Both statistical and domain expertise needs to come together for high-
quality statistics in the country. MoSPI needs to play its part much more proactively.
® Table 2.8 highlights the thematic overlaps in the major health surveys in the country. The SRS
and NFHS have major overlaps in the sphere of demographic / population / vital statistics. And
there is preference for SRS data in this regard at the central and state levels primarily due to
its annual periodicity and being from within the government system. The major advantage that
NFHS has over the SRS in this regard is the background characteristics by which it offers this
data. However, table 2.2 shows the extensive nature of data and the background characteristics
that the SRS collects. The SRS should put its entire data in the public domain – at the moment,
it only puts out a fraction of what it collects. Secondly, it should do so in a more organized and
professional manner. These are 2 areas in which the SRS can learn from the NFHS. If it actually
80
does so, there will be no need for the NFHS to waste its respondents’ precious time and GoI’s
limited resources on duplication. We recommend that vital statistics should only be collected
by SRS by incorporating from the NFHS all that it presently lacks so that there is no net loss
of data in the system, duplication is avoided and resources are rationalized.
® RGI should give up the SRS cause of death survey, for which a more specialized agency like
the ICMR is better suited. It can support it with its death statistics from the CRVS, MCCD, SRS,
etc. The ICMR should enhance the sample size for the CoD survey to yield representative data
at the national and state / UT levels. The frequency of this survey, like others, should preferably
be annual (for better respondent recall) or biennial at the most. In some senses, this survey is
the backbone of the country’s health system / HIS inasmuch as it is / would be the only robust
source of data based on which the massive burden of premature mortality in the country can
be addressed. India has been the leading contributor to premature deaths at all levels in the
world, and reduction in premature mortality is part of both the national and international health
agenda. The GoI should no longer let this critical source of data remain underutilized with RGI.
According to the World Health Statistics 2019, ‘monitoring of 11 health-related SDG indicators
relies on good-quality cause-of-death data from countries’ (WHO 2019: 50).
® Not just survey schedules, but fact sheets, at least, should also be prepared in local languages.
They should be made available as well as painted on the walls of SCs, PHCs, CHCs and district
hospitals in local languages. This can also be done for key indicators from non-survey sources.
This would not only help in the democratization of official data, but also enhance accountability.
® As NFHS data is most professionally and transparently organized and disseminated, especially
as done through the DHS Program, the latter’s template of data dissemination and visualization
should be adopted by all surveys at the national and state / UT levels. The NHDAC should lay
down guidelines for the same, ensure that they are religiously followed and develop a common
visualization platform that should be accessible to policymakers as well as the public. A mobile
phone app should also be developed for the visualization. The support of DHS Program, WHO,
UNICEF and the World Bank could be sought in this regard.
® Kindly refer to the table in the concluding chapter for the proposed health survey strategy and
health surveys at the national level.
Bibliography
Chattopadhyay, Aparajita and Ramanakumar Agnihotram. 2004. ‘Burden of disease in rural India:
An analysis through causes of death’. The Internet Journal of Third World Medicine 2(2).
https://bit.ly/2A1NgFl (7/6/2020, 23:58 hours).
81
Colaco, Rhea. 2016. ‘Open health data in India: Finally a reality?’. Health Express. Observer
Research Foundation. https://bit.ly/2W8L7A3 (13/3/2020, 19:23 hours).
Dandona, Rakhi, Anamika Pandey and Lalit Dandona. 2016. ‘A review of national health surveys
in India’. Bulletin of the World Health Organization 94: 286-296A.
Mahal, Ajay, Anup Karan and Michael Engelgau. 2010. “The economic implications of non-
communicable disease for India”. Health, Nutrition and Population (HNP) Discussion Paper 52913.
The World Bank, Washington DC.
Mahapatra, Prasanta and P V Chalapati Rao. 2001. ‘Cause of death reporting systems in India: A
performance analysis’. The National Medical Journal of India 14(3): 154-162.
Mahapatra, Prasanta. 2002. ‘The Verbal Autopsy based cause of death reporting systems in rural
areas of India’. Demography India 31(2): 235-252.
Mahapatra, Prasanta. 2017. ‘The Sample Registration System (SRS) in India: An Overview, as of
2017’. Prepared for the Workshop on National CRVS Verbal Autopsy Sampling Strategies, jointly
organised by University of Melbourne Bloomberg Initiative for Civil Registration and Vital Statistics
and Swiss Tropical and Public Health Institute, Basel, 14-15 August 2017. https://bit.ly/2Y4ESwE
(7/6/2020, 23:55 hours).
Rajan, S and K James. 2004. ‘Respondents and quality of survey data’. EPW 39(7): 659-663.
Rajan, S and K James. 2008. ‘Third National Family Health Survey in India: Issues, problems and
prospects’. EPW 43(48): 33-38.
RGI and AIIMS. 2017. ‘Manual for conducting Verbal Autopsy (VA)’. https://bit.ly/2vZaObM
(12/3/2020, 12:18 hours).
RGI. 1988. “Handbook on civil registration”. ORGI. MHA, GoI. https://bit.ly/39OZ6z8 (13/3/2020,
18:55 hours).
Srinivasan, K and Rakesh Mishra. 2020. ‘Quality of data in NFHS-4 compared to earlier rounds:
An assessment’. EPW 55(6): 40-45.
WHO. 2019. “World Health Statistics 2019”. World Health Organization, Geneva.
82
3. Scope of NFHS vis-à-vis MoHFW’s policies
Tables 3.1 and 3.2 show how NFHS presently contributes to MoHFW’s vision, mission, objectives
and policies. It does not necessarily have to contribute to them completely – as we have pointed
out, surveys like the NNMS and NMHS have been conducted for NCD and mental health policies.
This and subsequent tables are quite detailed and self-explanatory, and we will only highlight some
of the key aspects in this and subsequent discussions.
There are certain aspects on which NFHS cannot contribute due to the fact of being a population
survey, and we have marked them ‘not applicable’ (N/A). ‘Yes’ is mentioned against themes where
it contributes at the moment and ‘no’ / ‘partially’ against those where it could potentially contribute.
Obviously, this is a matter of interpretation. Let us provide a few examples here, however, to justify
our case regarding potential coverage in a population health survey of themes which may not look
so on the surface. A thorough exercise is required to identify themes of various policies / programs
that could be covered in population health surveys.
1) NFHS contributes partially to certain aspects of the vision, mission and objectives of MoHFW,
and not at all to other aspects. One can argue about its potential to shed light on establishment
of ‘comprehensive primary healthcare delivery system’, and its well-functioning linkages with
secondary and tertiary care, very important from the perspective of the management of NCDs.
One could argue that this should be covered under a facility rather than a population survey.
Very true. However, a population survey can also ask beneficiaries whether this is actually the
case, which would be more important from the perspective of increasing focus of GoI on health
sector performance and strengthening, in comparison to the structural-operational information
that a facility survey or study would generate, which too is important. There are socioeconomic
inequities too in accessing health systems, and they would also come out through a population
rather than a facility survey.
2) UHC has emerged as, perhaps, the most important theme in both the national and international
health narrative and trajectory – it is MoHFW’s vision. Yet, NFHS is of little help in this regard,
either from the perspective of service coverage, financial risk protection or health technology
assessment. NSS is also a population survey, but has been providing OOP data for a long time.
Not that NFHS should necessarily collect it – just pointing out the possibilities.
3) With respect to Mission 4 of MoHFW, NFHS could shed light on the regulatory status of health
service delivery and rational use of pharmaceuticals. In fact, antimicrobial use (AMU) is widely
used as a proxy of antimicrobial resistance (AMR) in countries where data on the latter is weak.
4) Self-reported questions about specific diseases yield inaccurate information at the population
level and do not serve any major purpose, except awareness (if there is parallel clinical testing
and comparison being done). Instead, as per WHO’s definition of health as well as MoHFW’s
83
vision and NHP 2017’s goal related to well-being, the NFHS could ask questions about mental,
emotional and social well-being. These are huge issues generally, especially for women during
pregnancy and motherhood as well as for people suffering from chronic conditions, especially
pain, or even the more concrete ones like diabetes, cancer, etc. Mental and musculoskeletal
disorders are huge dimensions and drivers of disability and lack of well-being which are rarely
covered in health surveys despite their massive day-to-day implications for the living. From a
demographic point of view, the topics of interest are fertility, birth, death, marriage, migration,
etc. Regular life and its challenges do not mean much. From health and well-being perspective,
though, they are core concerns, and should be so for public policies, programs, surveys, etc.
Table 3.1: Summary of NFHS-4’s scope vis-à-vis MoHFW’s vision, mission and objective statement and
relevant policies
Institutional context
Themes /
indicators
Completely Partially Scope (%)
Vision, Mission and Objective
Statement for MoHFW (24.4.2019)
7 0 4 57
NHP 2017 31 16 5 68
NPP 2000 16 6 7 81
NVP 2011 0 0 0 N/A
NPC-AMR 2011 0 0 0 N/A
NAP-AMR 2017 2 0 0 0
National Multisectoral Action Plan for
NCDs (2017-22)
22 9 2 50
NMHP 2014 14 0 0 0
EHR 2016 2 0 1 50
Source: Developed by Priyanka Tomar and Divya Chaudhry. 84
Table 3.2: Assessment of NFHS-4’ scope vis-à-vis MoHFW’s vision, mission and objective statement and relevant policies (our comments on
NFHS coverage in bold within [] brackets)
Document Goals / Vision / Mission Objectives Related indicators
‘Vision, Mission
and Objective
Statement for
MoHFW’
(24.4.2019)
Vision: Attainment of highest possible
level of health and well-being for all,
through preventive and promotive
health care and universal access to
good quality health services without
anyone having to face financial
hardship as a consequence [Partially –
see NHP 2017 goal below for details]
Mission:
1. Ensure availability of quality
healthcare on equitable, accessible
and affordable basis across
regions and communities with
special focus on under-served
population and marginalized
groups [Partially]
2. Establish comprehensive primary
healthcare delivery system and
well-functioning linkages with
secondary and tertiary care health
delivery system [No]
3. Develop the training capacity for
providing HRH (medical,
paramedical and managerial) with
adequate skill mix at all levels
[N/A]
4. Regulate health service delivery
and promote rational use of
pharmaceuticals in the country
[No]
1. Improve health status of the people
through concerted action [Partially]
2. Expand preventive, promotive, curative,
palliative and rehabilitative services
provided through public health sector
with focus on quality [No]
3. Progressively achieve universal health
coverage [Partially]
Not available
85
Document Goals / Vision / Mission Objectives Related indicators
National Health
Policy 2017
1. Attainment of the highest possible
level of health and well-being for all
at all ages [Partially]
2. Pivotal importance of SDGs [See
SDG section in this chapter]
Improve health status through concerted
policy action in all sectors and expand
preventive, promotive, curative, palliative and
rehabilitative services provided through the
public health sector with focus on quality
1. Progressively achieve UHC
a. Assuring availability of free,
comprehensive primary health care
services (CPHCS) [Partially]
b. Ensuring improved access and
affordability of quality secondary and
tertiary care services [No]
c. Achieving a significant reduction in OOPE
[See 3.a.iii in next column]
2. Reinforcing trust in public health care
system [No]
3. Align the growth of private health care
sector with public health goals (including
enabling private sector contribution to
making health care systems more
effective, efficient, rational, safe,
affordable and ethical) [No]
Specific quantitative goals and objectives under
NHP 2017:
1. Health status and programme impact
a. Life expectancy and healthy life
i. Life expectancy at birth (LEB) [Can be
estimated]
90
ii. Disability-adjusted life years (DALY) [NFHS-5
has questions on disability; DALYs can be
calculated together with disability data
from SRS, NSS and Census]
iii. TFR [Yes]
b. Mortality by age and / or cause
i. Under-five mortality rate (U5MR) [Yes] and
maternal mortality ratio (MMR) [No]
ii. Infant mortality rate (IMR) [Yes]
iii. Neonatal mortality rate (NNMR) [Yes] and still
birth rate (SBR) [Yes]
c. Disease prevalence / incidence
i. HIV / AIDS [Yes – knowledge, prevalence;
No – sustained antiretroviral therapy (ART)
or viral suppression]
ii. Leprosy, Kala-Azar and Lymphatic Filariasis
[No]
iii. Tuberculosis (TB) – cure rate in new sputum
positive patients and incidence reduction [No
– only prevalence of TB covered]
iv. Blindness [No]
v. Premature mortality from cardiovascular
diseases, cancer, diabetes or chronic
respiratory diseases [No]
90
For instance, Mohanty and Ram (2010) estimated LEB among poor and non-poor by caste and religion in India using NFHS. More recently, Asaria et al (2019) estimated
socioeconomic disparities in LEB combining data from NFHS and SRS.
86
Document Goals / Vision / Mission Objectives Related indicators
2. Health systems performance
a. Coverage of health services
i. Utilization of public health facilities [Yes –
only for delivery / birth covered]
ii. ANC [Yes] and skilled attendance at birth
[Yes]
iii. Newborn fully immunized by one year of age
[Yes]
iv. Met need of family planning [Yes]
v. Known hypertensive and diabetic individuals
maintaining controlled disease status [No –
only prevalence and medicine intake
covered]
b. Cross-sectoral goals related to health
i. Prevalence of tobacco use [Yes]
ii. Prevalence of stunting of under-five children
[Yes]
iii. Access to safe water and sanitation [Yes]
iv. Occupational injury among agricultural
workers [No – only injury due to spousal
violence covered]
v. National / State level tracking of selected
health behaviour [Yes – covered only in the
context of HIV/AIDS and fertility]
3. Health system strengthening
a. Health finance
i. Government health expenditure as % of GDP
[Not applicable / N/A]
ii. State health spending as % of their budget
[N/A]
iii. Proportion of households facing catastrophic
health expenditure [Partially – only out-of-
pocket expenditure (OOPE) in the case of
delivery for the most recent live birth in
87
Document Goals / Vision / Mission Objectives Related indicators
public and private health facilities
covered]
b. Health infrastructure and human resource
[N/A – RHS, the other data source under
HSRS, provides information on these
aspects]
c. Health management information [Partially –
only a few MIS indicators can be validated
through NFHS data]
National
Population Policy
(NPP) 2000
1. Immediate objective: Address unmet
needs for contraception, health care
infrastructure, and health personnel and
to provide integrated service delivery for
basic RCH care [Partially – excluding
infrastructure and integrated service
delivery]
2. Medium-term objective: Bring the TFR to
replacement levels by 2010 through
vigorous implementation of inter-sectoral
operational strategies [Partially – TFR
status only]
3. Long term objective: To achieve a stable
population by 2045, at a level consistent
with the requirements of sustainable
economic growth, social development and
environmental protection [No]
Relevant indicators based on national socio-
demographic goals (NSDG) under NPP 2000:
1. Unmet needs for basic RCH services,
supplies and infrastructure [Partially –
excluding infrastructure]
2. Free and compulsory school up to age 14,
dropout reduction at primary and secondary
school levels [Partially – school
attendance, educational attainment,
reasons for dropout covered]
3. IMR [Yes]
4. MMR [No]
5. Universal immunization of children against all
vaccine preventable diseases [Yes]
6. Girls’ age at marriage [Yes]
7. Institutional deliveries, by trained persons
[Yes]
8. Access to information / counseling and
services for fertility regulation and
contraception with a wide basket of choices
[Yes]
9. Registration of births, deaths, marriage and
pregnancy [Partially – only birth
registration covered]
10. Spread of AIDS [No – HIV prevalence
covered]
88
Document Goals / Vision / Mission Objectives Related indicators
11. Prevention and control of communicable
diseases [Partially – see footnote 38
above]
12. Integrated Indian systems of medicine (ISM)
for reproductive and child health (RCH)
services and household outreach [Partially –
no household outreach]
13. Promotion of small family norm [Yes]
National Vaccine
Policy (NVP) 2011
The Policy document provides – within
the overall framework of NHP – broad
policy guidelines and framework to
guide the creation of evidence base to
justify the need for R&D, production,
procurement and quality assessment of
vaccines for the Universal Immunization
Program (UIP) in India. It also
addresses the broad issues of
strengthening the institutional
framework, processes, evidence base
and framework required for decision
making for new vaccine introduction,
addresses vaccine security and
program management, regulatory
issues and product development [N/A]
Not available
Not available
‘The overarching goal of vaccine use is to reduce
morbidity and mortality due to vaccine
preventable diseases (VPD)’ (NVP 2011: 5).
- While NFHS provides data on vaccination, it
could also include questions which can help
assess their impact on morbidity, mortality as
well as other relevant outcomes.
‘Presently, the efforts to collect data on childhood
infectious diseases of public health importance
are fragmented and there is a need for reliable
and comparable data to establish baseline
information, monitor trends of infectious
diseases, and monitoring the impact of existing
interventions’ (NVP 2011: 22).
- NFHS should consider tackling this problem.
Surveys like the NHFS should ‘create data sets
on baseline demography’, which are of ‘utmost
importance in interpreting disease burden data,
results of clinical trials or when an adverse
events following any intervention has to be
investigated and causal linkages established’
(NVP 2011: 23).
- NFHS should consider this. It can also ask
respondents about related adverse events.
National Policy for
Containment of
Antimicrobial
The Policy is the report of a task force
constituted with following terms of reference:
Not available
While hospital-based sentinel surveillance was
recommended in the report and more generally,
89
Document Goals / Vision / Mission Objectives Related indicators
Resistance (AMR)
2011
1. Review the current situation regarding
manufacture, use and misuse of
antibiotics in the country [N/A]
2. Recommend the design for creation of a
national surveillance system for antibiotic
resistance [N/A]
3. Initiate studies documenting prescription
patterns and establish a monitoring
system for the same [N/A]
4. Enforce and enhance regulatory
provisions for use of antibiotics in human,
veterinary and industrial use [N/A]
5. Recommend specific intervention
measures such as rational use of
antibiotics and antibiotic policies in
hospitals [N/A]
6. Diagnostic methods pertaining to
antimicrobial resistance monitoring [N/A]
NFHS can include questions on awareness and
attitudes towards AMR, antimicrobials and their
use (AMU), sources of information, prescription
behavior, prevalence and causes of self-
medication, antibiotic use in agriculture, etc.
National Action
Plan on
Antimicrobial
Resistance (NAP-
AMR) 2017
1. To effectively combat AMR in India,
and contribute towards the global
efforts to tackle this public health
threat [N/A]
2. To establish and strengthen
governance mechanisms as well as
the capacity of all stakeholders to
reduce the impact of AMR in India
[N/A]
1. Define the strategic priorities, key actions,
outputs, responsibilities and indicative
timeline and budget to slow the
emergence of AMR in India and
strengthen the organizational and
management structures to ensure intra
and inter sectoral coordination with a One
Health approach [N/A]
2. Combat AMR in India through better
understanding and awareness of AMR,
strengthened surveillance, prevention of
emergence and spread of resistant
bacteria through infection prevention and
control, optimized use of antibiotics in all
sectors and enhanced investments for
AMR activities, research and innovations
[No]
M&E indicators:
1. Level of awareness and knowledge about
AMR in different social and professional
groups [No]
2. Number of IEC resources developed for
awareness / behaviour change
communication campaign on AMR and AMU
[N/A]
3. Revised curricula for health professionals and
for professionals in animal health, food
industry and agriculture [N/A]
4. Terms of reference for National Coordinating
Centre, National Reference Laboratories,
Infection Prevention and Control Coordinating
Unit and Multidisciplinary Antimicrobial
Stewardship Committees [N/A]
5. Establishment of a quality management
system for the medicines supply chain [N/A]
90
Document Goals / Vision / Mission Objectives Related indicators
3. Enable M&E of the NAP-AMR
implementation based on the M&E
framework [N/A]
6. Engagement with relevant experts to identify
research topics on AMR [N/A]
National
Multisectoral
Action Plan for
Prevention and
Control of
Common
Noncommunicable
Diseases (2017-
22)
91
Goal: Promote healthy choices, reduce
preventable morbidity, avoidable
disability and premature mortality due
to NCDs in India [Partially – only
selected healthy choices covered]
Vision: All Indians enjoy the highest
attainable status of health, well-being
and quality of life at all ages, free of
preventable NCDs and premature
death [Partially – see footnote 38
above]
1. Priority accorded and resources allocated
to the prevention and control of NCDs in
the national agenda and policies [N/A]
2. National capacity to lead multisectoral
partnerships to accelerate and scale-up
national response to NCDs [N/A]
3. Capacity of individuals, families and
communities to make healthier choices by
creating healthy environments that
promote health and reduce the risk of
NCDs [No]
4. Health systems provide accessible and
affordable good quality care to all people
with disease or risk factors through
primary health care approach [No]
5. Sustainable surveillance, monitoring and
evaluation systems for programme
development and monitoring that
promotes evidence-based policy and
programme development [N/A]
1. Probability of dying between ages 30-70
from cardiovascular disease (CVDs), cancer,
diabetes, or chronic respiratory disease
(CRDs) [No]
2. Cancer incidence by type of cancer [Yes –
self-reported cancer, no information
about type]
3. Alcohol consumption (aged 18+ years) [Yes
– only for covered age groups]
4. Obesity (adolescents and aged 18+ years)
[Yes – only for covered age groups]
5. Blood glucose / diabetes [Yes]
6. Physical activity (adolescents and aged 18+
years) [No – sexual activity covered]
7. Blood pressure (aged 18+ years) [Yes –
only for covered age groups]
8. Mean population intake of salt per day in
grams (aged 18+ years) [No – presence of
iodized salt in household and ORS (oral
rehydration salt) covered]
9. Tobacco use (smoking and smokeless)
(adolescents and aged 18+ years) [Yes –
only for covered age groups]
10. Households using solid fuels as a primary
source of energy for cooking [Yes]
11. Adults consuming less than 5 servings of
fruit and vegetables per day [Yes – only for
covered age groups]
91
The NNMS, conducted by ICMR’s NCDIR (Bengaluru), is meant to take care of India’s data reporting requirements for this Plan.
91
Document Goals / Vision / Mission Objectives Related indicators
12. Adults receiving drug therapy and
counselling to prevent heart attacks and
strokes [No]
13. Availability and affordability of quality, safe
and efficacious essential NCD medicines
including generics and basic technologies in
both public and private facilities [No]
14. Access to palliative care assessed by
morphine-equivalent consumption of strong
opioid analgesics per death from cancer
[No]
15. Vaccination coverage against hepatitis B
virus monitored by number of third doses of
Hep-B vaccine administered to infants [Yes]
16. Women aged 30-49 screened for cervical
cancer at least once [No]
92
17. Women aged 30 and above screened for
breast cancer by clinical examination by
trained health professional at least once [No]
18. High risk persons (using tobacco, smoking
and smokeless and betel nut) screened for
oral cancer by examination of oral cavity
[No]
National Mental
Health Policy
(NMHP) of India
2014
93
Goals:
1. To reduce distress, disability,
exclusion morbidity and premature
mortality associated with mental
1. To provide universal access to mental
health care [No]
2. To increase access to and utliisation of
comprehensive mental health services
Not available
92
NFHS-4 had a question (number 727) in the women’s schedule, which asked them – ‘have you ever undergone: a) a cervix examination, b) breast examination, c) an
oral cavity examination?’. It does not mention cancer and contrary to assumption in some quarters, this is not about cancer screening in particular, but rather about
general screening. There are 2 other questions about cancer – number 253 which asks whether hysterectomy, if done, was due to cancer; and number 723 which asks
whether the woman has cancer (as well as other diseases like diabetes, asthma, goiter or any other thyroid disorder or heart disease)?
93
The National Mental Health Survey, 2015-16, was meant to provide data for this policy.
92
Document Goals / Vision / Mission Objectives Related indicators
health problems across life-span of
the person [No]
2. To enhance understanding of
mental health in the country [No]
3. To strengthen the leadership in the
mental health sector at the national,
state and district levels [No]
Vision: to promote mental health,
prevent mental illness, enable recovery
from mental illness, promote
destigmatization and desegregation,
and ensure socio-economic inclusion of
persons affected by mental illness by
providing accessible, affordable and
quality health and social care to all
persons through their life-span within a
rights-based frame work [No]
(including prevention services, treatment
and care and support services) [No]
3. To increase access to mental health
services for vulnerable groups including
homeless person(s), person(s) in remote
areas, difficult terrains, educationally /
socially / economically deprived sections
[No]
4. To reduce prevalence and impact of risk
factors [No]
5. To reduce risk and incidence of suicide
and attempted suicide [No]
6. To ensure respect for rights and
protection from harm of person(s) with
mental health problems [No]
7. To reduce stigma related to mental health
problems [No]
8. To enhance availability and equitable
distribution of skilled human resources for
mental health [No]
9. To progressively enhance financial
allocation and improve utliisation for
mental health promotion and care [No]
10. To identify and address the social,
biological and psychological determinants
of mental health problems and to provide
appropriate interventions [No]
Electronic Health
Record (EHR)
Standards for India
2016
1. Promote interoperability and where
necessary be specific about certain
content exchange and vocabulary
standards to establish a path
forward toward semantic
interoperability [No]
2. Evolution and timely maintenance of
adopted standards [N/A]
Not available Not available
93
Document Goals / Vision / Mission Objectives Related indicators
3. Technical innovation using adopted
standards [N/A]
4. Participation and adoption by all
vendors and stakeholders [N/A]
5. Implementation costs as low as
reasonably possible [N/A]
6. Consider best practices,
experiences, policies and
frameworks [Partially]
7. To the extent possible, adopt
standards that are modular and not
interdependent [N/A]
Source: Developed by Priyanka Tomar and Ali Mehdi. 94
4. Scope of NFHS vis-à-vis MoHFW’s schemes
Regular monitoring and data reporting of schemes happens through their MIS, some of which were
discussed in chapter 2. It is interesting to note that in one of the MoHFW’s quarterly HMIS national
report
94
(30/6/2019), there were 8 indicators from SRS,
95
5 from the erstwhile DLHS
96
and only 3
from NFHS.
97
There is a tendency in government documentation – both at the Central and state
levels – to use vital indicators from the SRS and only a few service / impact indicators from NFHS.
Surprisingly, despite being discontinued 7 years back, DLHS data continues to be used even now.
Table 4.1: Summary of the scope of NFHS-4 vis-à-vis MoHFW schemes
Schemes
Themes /
indicators
Completely Partially Scope (%)
Central sector schemes
Pradhan Mantri Swasthya Suraksha Yojana 2 0 0 0
National AIDS and STD Control Programme 2 1 0 50
Family Welfare Schemes 2 1 1 100
Establishment and strengthening of NCDC
branches and health initiatives…
2 0 1 50
Pharma co-vigilance Programme of India 1 0 0 0
Development of Nursing Services 1 0 1 100
Health sector disaster preparedness and
response…
1 0 0 0
National Organ Transplant Programme 2 0 0 0
Setting up of nationwide network of
laboratories for managing epidemics
1 0 0 0
94
https://nhm.gov.in/New_Updates_2018/Quarterly_MIS/June-2019/National_Overview.pdf (24/10/2019, 14:34 hours).
95
TFR, CBR, life expectancy at birth, CDR, NNMR, IMR, U5MR and MMR.
96
Contraceptive prevalence rate, children under 3 years breastfed within one hour of birth (%), children age 0-5 months
exclusively breastfed (%), children age 6-35 months exclusively breastfed for at least 6 months (%) and children age 6-
9 months receiving solid / semi-solid food and breast milk (%).
97
Unmet need for family planning (%), institutional deliveries (%) and fully immunised children (%).
95
Schemes
Themes /
indicators
Completely Partially Scope (%)
Development of infrastructure for promotion
of health research
1 0 0 0
Human resource and capacity development 1 0 0 0
Development of tools / support to prevent
outbreaks of epidemics
2 0 0 0
Centrally sponsored schemes
National Rural Health Mission 49 15 1 33
National Urban Health Mission 8 4 2 75
Human resources for health and medical
education
5 0 3 60
Ayushman Bharat – Pradhan Mantri Jan
Arogya Yojana
7 0 3 43
Tertiary Care Programs 14 2 0 14
Source: Developed by Priyanka Tomar and Divya Chaudhry.
Tables 4.1 and 4.2 convey a similar impression about the scope of NFHS vis-à-vis various schemes
of the MoHFW as did tables 3.1 and 3.2 earlier. We have only included the outputs, outcomes and
their respective indicators against schemes from the Union Budget 2019-2020 Output Outcome
Framework that are relevant from the perspective of a population health survey like the NFHS. It
is largely the MIS indicators of NHM – that too primarily of its RMNCH+A component – which are
available on MoHFW’s Statistics Division’s website. The MIS indicators of other schemes, we were
told, are with the respective program managers, and not available in the public domain. Below are
2 illustrative comments – the tables otherwise are detailed and self-explanatory.
1) For the PMSSY scheme, in the light of what we said above, the NFHS could ask people about
the availability as well as accessibility of AIIMS and AIIMS-like institutes in particular and about
availability of affordable / reliable tertiary care in general.
2) It can also provide data on the efficiency of disease surveillance programs by asking questions
about disease symptoms, their risk factors, etc. – if not on diseases themselves. For diseases
where rapid diagnostics are available and could be leveraged, it could also consider collecting
disease prevalence data. Similarly, for the pharmacovigilance programme, it could ask people
whether any adverse drug reactions (ADRs) happened, etc. 96
Table 4.2: Assessment of the scope of NFHS-4 vis-à-vis MoHFW schemes (outlay, outputs and outcomes and their indicators relevant from a
population survey perspective taken from Output Outcome Framework with the Union Budget of India 2019-20)
Sn.
Name of the scheme
(relevant MoHFW department)
Outlay
(INR crores)
Relevant outputs Relevant outcomes
CENTRAL SECTOR (CS) SCHEMES
1
Pradhan Mantri Swasthya Suraksha
Yojana (Department of Health and Family
Welfare / DoHFW)
4,000
1. Increased accessibility to AIIMS and
AIIMS-like institutes [No]
2. Availability of affordable / reliable tertiary
care [No]
Not applicable (N/A)
2
National AIDS and STD Control
Programme (DoHFW)
2,500 N/A
1. People living with HIV who know their
HIV status [Yes]
2. … and are on ART (antiretroviral
therapy) [No]
3 Family Welfare Schemes (DoHFW) 700 N/A
1. To achieve family planning (FP) 2020
goal [Partially]
98
2. Increase in awareness level [Yes]
4
Establishment and strengthening of
NCDC branches and health initiatives,
inter-sectoral coordination for
preparation and control of zoonotic
diseases and other neglected tropical
diseases, National Viral Hepatitis
Surveillance Programme, and Anti-
Microbial Resistance Containment
Programme (DoHFW)
49
1. Improved capacity of states and district
level manpower for prevention and control of
zoonosis diseases [No]
2. Surveillance systems for hepatitis and
AMR [Partially – hepatitis B vaccination for
children covered]
N/A
98
NFHS is the source of 13 out of 18 FP 2020 core indicators (indicators 1-9, 14, 16-18). Source: FP Core Indicator Summaries (2018). https://bit.ly/31AETYX (22/10/2019,
12:10 hours). It could also collect data on indicator 15 (women provided with information on FP during recent contact with a health service provider).
97
Sn.
Name of the scheme
(relevant MoHFW department)
Outlay
(INR crores)
Relevant outputs Relevant outcomes
5
Pharma co-vigilance Programme of India
(DoHFW)
12 N/A
To create a nationwide system to report
ADRs for patient-safety [No]
6
Development of Nursing Services
(DoHFW)
15 N/A
Availability, knowledge and skills of nurses
[Partially – partial performance covered]
7
Health sector disaster preparedness and
response and human resources
development for emergency medical
services (DoHFW)
130 N/A
Doctors, nurses and paramedics trained in
emergency life support [No]
8
National Organ Transplant Programme
(DoHFW)
41 Increase awareness on organ donation [No] Organ donation rate [No]
9
Setting up of nationwide network of
laboratories for managing epidemics and
national calamities (Department of Health
Research / DoHR)
80 N/A
Timely diagnosis of epidemics and
availability of trained viral research and
diagnostic professionals at medical
colleges, state and regional level
laboratories [No]
10
Development of infrastructure for
promotion of health research (DoHR)
73 N/A
Outcome indicator
Increase in transfer of new technologies for
improving the quality of health services to
rural population [No]
11
Human resource and capacity
development (DoHR)
87 N/A
Outcome indicator
Number of evidence-based guidelines
issued on Health Technology Assessment
(HTA) [No]
12
Development of tools / support to
prevent outbreaks of epidemics (DoHR)
7.35
Providing diagnostics for non-viral infectious
pathogens [No]
Research activity for preparedness and to
generate quality and uniform pan-India
data [No]
98
Sn.
Name of the scheme
(relevant MoHFW department)
Outlay
(INR crores)
Relevant outputs Relevant outcomes
CENTRALLY SPONSORED SCHEMES (CSS)
13 National Rural Health Mission (DoHFW) 27,039 See scheme components below See scheme components below
13(1)
Health systems strengthening under
NRHM
1. Expanded basket of primary care services
provided by Health & Wellness Centers
(HWCs) [No]
2. Implementation of NHM Free Diagnostics
Services Initiative at public health facilities
[No]
1. Improved utilization of primary care
services and screening & management of
NCDs [No]
2. Increased availability of drugs and
diagnostics at public health facilities [No]
3. Improved utilization of public health
facilities [Yes – only for delivery / birth
covered]
4. Improved access to emergency obstetric
care services [Yes – caesarean section
delivery]
5. Patients receiving free dialysis care [No]
Outcome indicators
1. Number of total 30+ population
screened for NCDs [No – see footnote 38
above]
2. Reduction in OOPE on health in public
health facilities (proxy – childbirth) [Yes]
13(2)
RCH Flexipool (including Routine
Immunization Programme, Pulse Polio
Immunization Programme, National
Iodine Deficiency Disorders Control
Programme, etc.)
Output indicators
1. Percentage of pregnant women who
received 4 ANCs (antenatal care) [Yes]
2. Percentage of SBA (skilled birth attendant)
deliveries to total ANCs registered [Yes]
3. Full immunization coverage [Yes – mother
and child]
Outcome indicators
1. MMR [No]
2. U5MR [Yes]
3. TFR [Yes]
99
Sn.
Name of the scheme
(relevant MoHFW department)
Outlay
(INR crores)
Relevant outputs Relevant outcomes
4. Use of modern methods of contraception
[Yes]
13(3)
National Iodine Deficiency Disorders
Control Programme
Output indicator
Availability of adequately iodized salt [Yes]
N/A
13(4) Disease Control Programme N/A N/A
13(4-A)-
A)
National Vector Borne Diseases Control
Programme
1. Malaria: Reduction in number of cases
[No]
2. Kala-azar: Reduction in PKDL (diagnosis of
post Kala-azar dermal leishmaniasis) cases
[No]
3. Japanese Encephalitis (JE) / Coverage of
JE in routine immunization at the national
level [No]
4. Lymphatic Filariasis (LF): Protect the
population by Mass Drug Administration in LF
endemic districts [No]
N/A
13(4-B)
National Viral Hepatitis Control
Programme
N/A
1. Free treatment of hepatitis C available
[No]
2. Free treatment of hepatitis B available
[No]
3. Enhanced coverage of birth dose
hepatitis B vaccine [Yes]
Outcome indicators
1. Number of new patients completed
treatment of HCV [No]
2. Number of patients who put on
treatment continuing on treatment [No]
13(4-C) National Leprosy Eradication Programme N/A
Grade II disability (G2D) due to leprosy
[No]
100
Sn.
Name of the scheme
(relevant MoHFW department)
Outlay
(INR crores)
Relevant outputs Relevant outcomes
13(4-D)
Revised National Tuberculosis Control
Programme
Output indicator
Percentage increase in TB case notification
(public & private) [Yes – TB prevalence
covered]
Increased detection of drug resistant TB
cases [No]
Outcome indicator
1. Percentage of patients whose outcomes
are successful [No]
2. Percentage increase in drug resistant TB
cases [No]
13(4-E)
Integrated Disease Surveillance
Programme (IDSP)
Improved capacity of districts to detect and
respond to disease outbreaks [Partially –
very limited coverage of IDSP diseases
under surveillance in NFHS]
N/A
13(5) Non Communicable Disease Programme See sub-scheme components below See sub-scheme components below
13(5-A)
National Programme for prevention and
control of Cancer, Diabetes,
Cardiovascular diseases and Stroke
1. Additional NCD clinics to be set up at
CHCs and district hospitals [No]
2. Screening for high blood pressure & high
blood sugar [No – see footnote 38 above]
1. Relative reduction in overall mortality
from cardiovascular diseases (CVDs),
cancer, diabetes, chronic respiratory
diseases (CRDs) [No]
2. Early detection of high blood pressure &
high blood sugar [Yes – see footnote 38
above]
13(5-B) National Mental Health Programme
Provision of mental health services under
District Mental Health Programme [No]
Improved coverage of mental health
services [No]
13(5-C) National Blindness Control Programme
Eye care services under NPCB&VI provided
at primary, secondary at district and below
level [No]
Output indicator
Cataract surgeries [No]
Improvement in surgical skills and quality
[No]
Outcome indicator
Reduction in prevalence of blindness [No]
13(5-D)
National Programme for Health Care of
Elderly
Provision of primary, secondary and tertiary
geriatric health care services at district
hospital and below [No]
N/A
101
Sn.
Name of the scheme
(relevant MoHFW department)
Outlay
(INR crores)
Relevant outputs Relevant outcomes
Output indicator
1. District hospitals with physiotherapy and
laboratory services [No]
2. CHCs with geriatric OPD and geriatric
physiotherapy services [No]
13(5-E)
Tobacco Control Programme & Drug De-
addiction Programme
1. Increase in availability of tobacco cessation
services [Yes]
2. Increase in facilities for treatment of drug
addiction [No]
1. Improved access for tobacco cessation
services [Yes]
2. Improved access to drug dependence
treatment services [No]
14
National Urban Health Mission – Flexible
Pool (DoHFW)
950
Output indicators
1. UPHCs and UCHCs providing
comprehensive primary health care services
with adequate staff [Partially – covered
under source of health care, modern
contraceptive methods and emergency
contraceptive pills]
2. Women getting at least 4 ANCs [Yes]
3. Children getting full immunization [Yes]
4. UHNDs (Urban Health & Nutrition Days)
outreach conducted by UPHCs [No]
1. Improved access to quality healthcare in
urban India [Partially – among reasons
for not using government health care
and reasons for not delivering the most
recent live birth in a health facility]
2. Increased utilization of public health
facilities [Yes – only for delivery / birth
covered]
Outcome indicators
1. MMR [No]
2. IMR [Yes]
15
Human resources for health and medical
education (DoHFW)
4,250 See scheme components below See scheme components below
15(1)
District hospitals – Upgradation of state
government medical colleges (PG seats)
N/A Availability of specialist doctors [No]
15(2)
Strengthening of government medical
colleges (UG seats) and central
government health institutions
N/A
Availability of doctors [Partially – RCH
services provided by doctors covered]
15(3)
Establishment of new medical colleges
(upgrading district hospitals)
N/A
Output indicator
Tertiary level services [No]
102
Sn.
Name of the scheme
(relevant MoHFW department)
Outlay
(INR crores)
Relevant outputs Relevant outcomes
15(4)
Upgradation / strengthening of nursing
services (ANM / GNM)
N/A
Nurses for healthcare [Partially – partial
performance covered]
15(5)
Setting up of state institutions of para-
medical sciences in states and setting up
of colleges of para-medical education
N/A
Availability of allied health professionals
[Partially – partial performance covered]
16
Ayushman Bharat – Pradhan Mantri Jan
Arogya Yojana (DoHFW)
6,400
1. Hospital admissions [No]
2. Beneficiary identification [No]
Output indicator
Public and private hospitals empaneled [No]
Reduction in health expenditure [Partially
– only OOPE in the case of delivery for
the most recent live birth in public and
private health facilities covered]
Outcome indicators
1. Proportion of households incurring
catastrophic health expenditure [No]
2. Percentage of out-of-pocket health
expenditure incurred by beneficiaries
[Partially – only OOPE in the case of
delivery for the most recent live birth in
public and private health facilities
covered]
3. Average out-of-pocket expenditure
incurred by beneficiaries [Partially – only
OOPE in the case of delivery for the
most recent live birth in public and
private health facilities covered]
17 Tertiary Care Programs (DoHFW) 550 See scheme components below See scheme components below
17(1) National Mental Health Programme
Improved coverage of mental health services
[No]
Improved availability of mental health
professionals [No]
17(2)
Assistance for capacity building of
trauma centres (1. trauma centres, 2.
prevention of burn injury)
N/A
Strengthened trauma care facilities and
burn units for enhanced quality care to
trauma and burn victims [No]
Outcome indicator
103
Sn.
Name of the scheme
(relevant MoHFW department)
Outlay
(INR crores)
Relevant outputs Relevant outcomes
Provision of quality services to the victims
of trauma and burn injuries by reducing
deaths and disabilities [No]
17(3)
National Programme for Health Care of
Elderly
Provision of tertiary geriatric care services
[No]
N/A
17(4)
National Programme for Control for
Blindness
Sensitization training sessions for trachoma
elimination in previously endemic states for
trachoma [No]
Trained eye surgeons, SPOs and
ophthalmologists [No]
17(5) Tele medicine Doctors for specialist consultation [No] N/A
17(6)
Tobacco Control & Drugs De-addiction
Programme
1. Increase in availability of tobacco cessation
services [Yes]
2. Increase in facilities for treatment of drug
addiction [No]
1. Improved access for tobacco cessation
services [Yes]
2. Improved access to drug dependence
treatment services [No]
17(7)
National Programme for prevention and
control of Cancer, Diabetes,
Cardiovascular diseases and Stroke
Output indicator
Drug treatment with OPD services [No]
Increase in availability of radio therapy
machines [No]
Source: Union Budget of India 2019-20. Developed by author. 104
5. Scope of NFHS vis-à-vis health-related SDGs
The SDG indicator frameworks are a bit complex, so let us start with a brief overview of the global
and national historical contexts.
Global context
On 25 September 2015, the United Nations General Assembly (UNGA) adopted the 2030 Agenda
for Sustainable Development (Resolution 70/1). ‘A plan of action for people, planet and prosperity’,
it is ‘integrated’, ‘indivisible’ and balances the ‘economic, social and environmental’ dimensions of
sustainable development. Its ‘integrated’ nature is considered of ‘crucial importance’ for realizing
its ‘purpose’ by means of ‘collaborative partnership’ between ‘all countries and all stakeholders’.
The Agenda envisages ‘a world free of poverty, hunger, disease and want, where all life can thrive’,
‘where physical, mental and social well-being are assured’, where ‘all human beings can fulfil their
potential in dignity and equality and in a healthy environment’. It sees poverty eradication ‘in all its
forms and dimensions’ as ‘the greatest global challenge’ as well as ‘an indispensable requirement
for sustainable development’.
A total of 17 SDGs with 169 targets, to be achieved between 2015 and 2030, were included in the
2015 UNGA (Resolution 70/1). On 6 July 2017, the UNGA adopted the Global Indicator Framework
(GIF) with 244 indicators – 232 excluding the ones repeated under different targets
99
– developed
by the Inter-Agency and Expert Group on SDG Indicators (IAEG-SDG) and accepted by the United
Nations Statistical Commission (UNSC) at its 48th session (7-10 March 2017). The 2017 UNGA
referred to GIF ‘as a voluntary and country led instrument that includes the initial set of indicators
to be refined annually and reviewed comprehensively’ by the UNSC at its 51st session (2020) and
56th session (2025), and to be ‘complemented by indicators at the regional and national levels’ to
be developed by the member states. The decision on National Indicator Framework (NIF) is to be
driven by national priorities, but aligned with SDG-GIF as much as possible.
100
‘Targets are defined
as aspirational and global’ and each government has to set ‘its own national targets guided by the
global level of ambition but taking into account national circumstances’. Recognizing gaps in data
availability, a ‘call for increased support for strengthening data collection’ was made (UNGA 2015),
and SDG target 17.18 (‘data, monitoring and accountability’) was included. SDG indicator 17.18.1
is about the ‘proportion of sustainable development indicators produced at the national level with
99
Repeated SDG indicators (9) – 1) 8.4.1/12.2.1, 2) 8.4.2/12.2.2, 3) 10.3.1/16.b.1, 4) 10.6.1/16.8.1, 5) 15.7.1/15.c.1, 6)
15.a.1/15.b.1, 7) 1.5.1/11.5.1/13.1.1, 8) 1.5.3/11.b.1/13.1.2, 9) 1.5.4/11.b.2/13.1.3. The official indicator list, with periodic
refinements agreed by the UN Statistical Commission, is available at – https://bit.ly/2oKA41P (16/10/2019, 10:51 hours).
Total SDG indicators remain 244 after annual refinements in 2018 (E/CN.3/2018/2) and 2019 (E/CN.3/2019/2).
100
https://bit.ly/35CJ0Hd (16/10/2019, 13:03 hours).
105
full disaggregation when relevant to the target, in accordance with the Fundamental Principles of
Official Statistics’. The UNGA 2015 called for ‘data which is high-quality, accessible, timely, reliable
and disaggregated by income, sex, age, race, ethnicity, migration status, disability and geographic
location and other characteristics relevant in national contexts’.
The IAEG-SDG has classified the 232 indicators into 3 tiers ‘based on their level of methodological
development and the availability of data at the global level’ (table 5.1).
101
Table 5.1: Tier classification criteria of SDG indicators
Tier Classification Criteria
Number of SDG indicators
(as on 26 Sep 2019)
1
Conceptually clear, internationally established methodology, standards
available, data regularly produced by at least 50% of countries and of the
population in every region where the indicator is relevant.
104
2
Conceptually clear, internationally established methodology, standards
available, but data not regularly produced by countries.
89
3
No internationally established methodology or standards available, but
being (or will be) developed or tested.
33
Multi-tier (different components of indicator classified into different tiers) 6
Source: United Nations Statistics Division. https://bit.ly/2jbOWzA (16/10/2019, 14:59 hours).
Given one’s interpretation of determinants of health, beyond the direct 27 health indicators under
SDG 3, one can add other SDG goals / targets / indicators to one’s list of health-related SDGs. For
instance, according to figure 5.1, SDG target 3.4 (NCDs) alone is linked with 8 SDG goals from the
lens of social determinants of health (SDH). SDGs are ‘integrated’, and such interconnections are
encouraged from conceptual as well as operational perspectives. In World Health Statistics 2019,
the WHO tracks 43 health-related SDG indicators. Of them, 36 have unique indicator numbers in
the official list of SDG indicators. Apart from SDG 3 indicators, WHO includes SDG 1.a.2 (domestic
government health expenditure), 2.2.1 (child stunting), 2.2.2 (child wasting), 2.2.3 (child obesity),
5.1.2 (intimate partner violence), 6.1.1 (safe drinking water), 6.2.1 (safe sanitation), 6.a.1 (water
sector ODA), 7.1.2 (clean energy coverage), 11.6.2 (fine particulate matter in urban areas), 16.1.1
(homicide) and 17.9.2 (cause of death data completeness). With the exception of 2 indicators under
SDG 3 – SDG 3.5.1 (Tier 3) and SDG 3.b.3 (Tier 2) – all other SDG 3 indicators in the GIF have been
classified as Tier 1 indicators (classification revised until 26 September 2019).
102
101
https://unstats.un.org/sdgs/iaeg-sdgs/tier-classification/ (16/10/2019, 14:57 hours).
102
https://unstats.un.org/sdgs/iaeg-sdgs/tier-classification/ (24/10/2019, 22:32 hours).
106
Figure 5.1: Interlinkages between SDG 3.4 (NCDs) and other SDGs
Source: Nugent et al 2018: 3.
Indian context
In India, MoSPI and the NITI Aayog have been involved at the central level for monitoring progress
vis-à-vis the SDGs.
In consultation with Union government ministries / departments, states, UN, civil society and other
stakeholders, MoSPI developed a National Indicator Framework (NIF) in the light of the global SDG
indicators, and released a baseline report in March 2019 (version 1.0). A year later, in March 2020,
it brought out an SDG progress report based on a revised list of NIF (version 2.0). NIF version 1.0
had a total of 306 indicators, while version 2.0 had 297. Of these, SDG 3 on health had the highest
number of indicators in both versions – 41 and 42 respectively. NFHS was identified as the source
of 30 SDG-NIF indicators in version 1.0 (9.8% of all NIF indicators) and 27 in version 2.0 (9.1%).
103
For SDG 3, NFHS was identified as the source of 14 (34.1%) and 16 (38.1%) indicators respectively.
103
Version 1.0 indicators with NFHS as the source – 1.3.1, 1.3.5, 1.4.3, 2.1.1, 2.2.1, 2.2.2, 2.2.3, 2.2.4, 2.2.5, 3.1.2, 3.1.3,
3.1.4, 3.2.3, 3.5.3, 3.7.1, 3.7.2, 3.7.3, 3.8.1, 3.8.4, 3.8.5, 3.8.7, 3.9.2, 3.a.1, 5.2.6, 5.3.2, 5.6.1, 5.6.2, 5.6.3, 6.2.1 and 7.1.1.
Version 2.0 indicators with NFHS as the source – 1.3.1, 2.1.1, 2.2.1, 2.2.2, 2.2.3, 2.2.4, 2.2.5, 3.1.2, 3.1.3, 3.1.4, 3.2.3,
3.5.1, 3.5.3, 3.7.1, 3.7.2, 3.7.3, 3.7.4, 3.8.4, 3.8.5, 3.8.6, 3.8.7, 3.9.2, 3.a.1, 5.2.6, 5.3.2, 5.6.2 and 5.6.3.
107
On its part, the NITI Aayog has also developed 2 versions of the SDG India Index (SII). SII version
1.0 was released in December 2018 with 62 National Priority Indicators (NPIs) from MoSPI’s NIF
version 1.0 – out of which, there were 5 SDG 3 indicators, with NFHS identified as the data source
for 10 indicators overall (16.1% of all SII indicators), including 3 SDG 3 indicators (60% of SDG 3
SII indicators). SII version 2.0 was released in December 2019 with 100 indicators
104
– out of which,
there were 8 SDG 3 indicators, with NFHS identified as the source for 6 indicators overall (6% of
all SII indicators), including 2 SDG 3 indicators (25% of SDG 3 SII indicators).
105
NFHS’ significance has relatively reduced vis-à-vis MoSPI’s NIF as well as NITI Aayog’s SII SDG
indicators over their 2 versions, with the exception of SDG 3 NIF indicators in MoSPI’s version 2.0.
However, it needs to be pointed out that there are certain indicators for which the NFHS provides
the data, but the SRS or another source has been identified. And, if those were also included, the
NFHS tally would be much higher.
At the same time, it is worth referring to what the SII version 1.0 report stated regarding the issue
of data availability: ‘the preparation of the index has highlighted data gaps related to the SDGs’, and
‘the need for India to develop its statistical systems at the national and State levels’, and ‘increase
the capacity and capability of data collection’ (NITI Aayog 2018: 4). SII version 2.0 report restated
that the ‘incomplete coverage of targets remains a partially resolved issue’ (NITI Aayog 2019: 7).
Health surveys in the country should try to provide data on all health-related SDGs to facilitate an
independent assessment of India’s progress, at least, at the national and state / UT levels. Likewise,
states / UTs should design their health surveys so that they could potentially offer data on all health-
related SDGs at the state, district and sub-district levels.
A related challenge is that of data interoperability, considering the ‘integrated’ nature of the SDGs.
The Economic Survey of India 2018-19 states that ‘data collection in India is highly decentralised.
For each indicator of social welfare, responsibility to gather data lies with the corresponding union
ministry and its state counterparts. Consequently, data gathered by one ministry is maintained
separately from that gathered by another’. Since ‘these datasets are unconnected, each ministry
only has a small piece of the jigsaw puzzle that is the individual / firm. However, if these different
pieces could be put together, we would find that the whole is greater than the sum of parts’ (Vol.
1: 85). Within ministries, there are jigsaw puzzles, and the idea of an integrated and interoperable
HIS appears to be a far cry in the given situation, not to talk of interoperability across GoI or at the
national level. Thankfully, there is a realization, and steps are being taken to address the problem.
104
While 68 were taken from MoSPI’s NIF, 20 were ‘modified or refined for the sake of data availability across all States/
UTs. 12 indicators which are not part of NIF were identified in consultation with line ministries’ (NITI Aayog 2019: 7).
105
Version 1.0 indicators with NFHS as the source – 1.3.1, 1.3.5, 2.2.1, 2.2.4, 3.2.1, 3.2.3, 3.8.8 (NFHS does not provide
data for this indicator [‘total physicians, nurses and midwives per 10,000 population’] – MoSPI identified DGHS / MoHFW
and not NFHS as the source for this indicator), 5.2.3 (MoSPI identified NCRB, not NFHS, as the source for this indicator),
5.6.1 and 7.1.2. Version 2.0 indicators with NFHS as the source – 1.3.1, 1.3.5, 2.2.4, 3.2.1, 3.7.1 and 5.2.6.
108
Recommendations
® The ‘integrated’ nature of Agenda 2030, and the ‘collaborative partnership’ required to realize
its ‘purpose’, has led to international discussions on data interoperability to leverage traditional
and new sources of data for monitoring progress on SDGs.
106
The MoHFW released the EHR
standards (version 1 in 2013 and version 2 in 2016) aimed at ‘achieving syntactic and semantic
interoperability of health records’ of individuals. MoSPI, NITI Aayog, MoHFW and others should
coordinate to develop a plan of action for data interoperability at the population health level,
and suggest syntactic and semantic changes for various health data sources. Simply mapping
data sources / schemes / ministries would not be enough to realize the ‘integrated’ nature of
the SDGs. The Intersecretariat Working Group on Household Surveys (ISWGHS), established
in 2015 by the UNSC ‘to foster coordination and harmonization of household survey activities’,
can also possibly be consulted. MoSPI has already been participating in UNSC’s SDG activities.
® Interoperability will help in understanding the social determinants of health, syndemics
107
as well
as the social implications of health (how pursuit of health impacts other sectors and concerns),
in the spirit of the 2030 Agenda. A One-Health approach for the surveillance of antimicrobial
resistance (AMR) – involving humans, animals, food and the environment – is already widely
accepted internationally, and there has been some renewed focus on syndemics. A syndemic
and comprehensive One-Health approach (beyond the narrow focus on AMR) to public health
surveillance should be developed by DoHR (MoHFW), MoSPI, MEITY, NITI Aayog and others.
® ‘Collaborative partnership’ between all stakeholders implies that data for monitoring progress
on SDGs should be sourced from as well as shared with various stakeholders, and not just the
governmental. Given that 71.7% spells of ailment in rural India and 78.8% in urban India were
treated in the private sector (NSS 71st round),
108
we need a plan to access and integrate this
106
Data interoperability in the context of SDGs has emerged as a key theme in international discussions since the 47th
session of UNSC in March 2016. The Cape Town Global Action Plan for Sustainable Development Data, launched at
the first United Nations World Data Forum (UNWDF) in January 2017, and adopted at the 48th UNSC Session in March
2017, calls on countries to modernize their statistical standards, to ‘define and implement standardized structures for the
exchange and integration of data and metadata on the social, economic and environmental pillars of sustainable
development’ at the global, regional, national and sub-national levels, and ‘promote interoperability of these systems to
facilitate such integration’ (strategic area 2, objective 2.2). The Plan also called for increasing ‘the integration of data
from different sources: surveys, administrative data and new sources’ (strategic area 3, objective 3.1). Interoperability
of SDG data continued to be discussed during the 49th (2018) and 50th (2019) sessions of UNSC as well as UNWDF 2
(2018), where ‘Data interoperability: A practitioner’s guide to joining up data in the development sector’ was launched.
107
‘Specifically, a syndemics approach examines why certain diseases cluster (ie, multiple diseases affecting individuals
and groups); the pathways through which they interact biologically in individuals and within populations, and thereby
multiply their overall disease burden, and the ways in which social environments, especially conditions of social
inequality and injustice, contribute to disease clustering and interaction as well as to vulnerability’ (Singer et al 2017: 941).
108
NSS. 2015. ‘Key indicators of social consumption in India: Health’. NSS 71st Round (January-June 2014). MoSPI.
109
data from the private sector – and possibly private personal data, with paramount priority given
to individuals’ consent and confidentiality in particular – for monitoring national health policies,
programs as well as health-related SDGs. Government-funded health insurance programs are
one mechanism through which data sharing by the private sector and individuals can be made
mandatory – again, with utmost importance assigned to the confidentiality of the data sources.
® The 2030 Agenda talks about poverty in all its forms as the greatest global challenge. However,
it is the notion of ‘income poverty’ that is most prevalent. Even in terms of health, SDG indicator
3.8.2 (proportion of population with large household expenditures on health as a share of total
household expenditure or income) is motivationally concerned with impoverishment, but of an
income / expenditure type, arising from access to health care. We should consider developing
the notion of and a composite index for ‘health poverty’, primarily focused on health outcomes,
but also access to efficacious, safe and quality health care. NITI Aayog’s Health Index can be
an inspiration for a ‘Health Poverty Index’ (HPI). The Multidimensional Poverty Index (MPI) –
developed by the Oxford Poverty & Human Development Initiative (OPHI) and the UNDP, with
10 indicators for 3 dimensions of poverty (health, education, living standards) – is great, but is
multisectoral. We need one specifically for health as well. The World Bank has one for ‘learning
poverty’ – why not have one for health poverty too?
Bibliography
MoSPI 2019. “Sustainable Development Goals: National Indicator Framework. Baseline report
2015-16”. MoSPI, GoI.
MoSPI. 2020. “Sustainable Development Goals: National Indicator Framework. Progress report,
2020”. MoSPI, GoI.
NITI Aayog. 2018. “SDG India Index: Baseline report, 2018”. NITI Aayog, GoI.
NITI Aayog. 2019. “SDG India: Index & Dashboard 2019-20”. NITI Aayog, GoI.
Nugent, Rachel et al. 2018. ‘Investing in non-communicable disease prevention and management
to advance the Sustainable Development Goals’. The Lancet Taskforce on NCDs and Economics
391 (10134): 2029-2035.
Singer, Merrill, Nicola Bulled, Bayla Ostrach and Emily Mendenhall. 2017. ‘Syndemics and the
biosocial conception of health’. The Lancet 389 (10072): 941-950. 110
6. Data for Health Technology Assessment (HTA)
With India’s increasing commitment to universal health coverage, especially following the launch
of Ayushman Bharat, the necessity for the establishment and strengthening of HTA frameworks
and capacities has increased in the country. Our journey in this direction has just begun, and it is
the right time to incorporate HTA as well in our discussion of institutional data requirements in the
country. We referred to NHP 2017’s commitment to the development of an institutional framework
and capacity for HTA in chapter 1. As a follow-up, DoHR in MoHFW set up the Medical Technology
Assessment Board (MTAB) in January 2017, later renamed as the Health Technology Assessment
in India (HTAIn),
109
with the following objectives
110
–
1) Maximize health and minimize out-of-pocket (OOP) expenditures and inequities in health care
services;
2) Assess new / existing health technologies vis-à-vis safety; cost and clinical efficiency; budget
impact; ethical, social and political feasibility to ensure rational allocation of limited resources
for access to quality and equitable health care;
111
3) To collect and analyze data in a systematic and reproducible manner, ensuring its accessibility
and usefulness to inform health policy;
4) Support central and state level health care decision-making;
5) Disseminate findings and related policy decisions to educate and empower the public to make
better informed decisions for health.
The HTAIn is not only responsible for collating, but also generating evidence wherever needed.
112
The Health Technology Assessment Board Act, 2019 is in public domain for comments.
113
A DHR-
HTA National Database will also be freely available to the public in the near future. As of now, data
sources available for HTA-relevant information and assessment are captured in table 6.1 below.
It has been argued that there is ‘marked absence’ of certain types of ‘data necessary for informing
HTA, particularly data relating to cost, service use, and quality of life’ in the Indian context (Downey
2019: 1). As we can see, NFHS is the source of very limited information from an HTA perspective.
109
HTAIn Secretariat. 2019. ‘A compendium of health technology assessment in India (2017-18)’. DoHR (hereafter ‘HTA
compendium’).
110
https://dhr.gov.in/about-mtab (4/10/2019, 15:36 hours).
111
This is particularly helpful as India progressively moves towards universal health coverage (UHC). सर्वेः सन्तु निरामयाेः
(let all be healthy) is inscribed in HTAIn’s logo, expressing its commitment to UHC. https://bit.ly/2LIP09o (4/10/2019,
15:36 hours). HTAIn has been working with Ayushman Bharat to review the packages offered by it (‘HTA compendium’).
112
http://htain.icmr.org.in/ (4/10/2019, 15:52 hours).
113
https://bit.ly/2MbdZRL (4/10/2019, 15:44 hours).
111
Table 6.1: Key national data sources for HTA in India (as on 17 May 2019)
HTA-related information Data source
Commissioning
body
Equity-relevant
information
Epidemiology
(communicable disease)
IDSP MHFW Geographic location
Epidemiology, service use,
health expenditure
HMIS MHFW Gender, geographic location
Epidemiology, service use,
OOP spending (for
institutional delivery only)
NFHS-4 MHFW
Location, gender, ethnicity, age,
marital status, contraception
use, HIV status, health
insurance, water/ sanitation
access, literacy, female parity
Epidemiology SRS, Census 2011 MHA
Location, gender, religion,
education, occupation,
caste/tribe, language, socio-
economic status
Health and service
use/utilization for RMNCH
indicators
DLHS-3 MHFW
Accessibility of services to
women and children in rural
villages
Epidemiology, service use,
OOP spending
NSS MS
Location, socioeconomic
status, gender, rural/urban,
age
Safety, efficacy, clinical
comparator(s)
ICMR Clinical Trials
Registry
ICMR No
Epidemiology - Cancer
ICMR Cancer Registry
Program
ICMR
Location, gender, rural/ urban,
age
Health expenditure
National heath Accounts
(2014/2015)
MHFW
Public and private sector
expenditure
Billing/Price
Database of Indian
Health Benefit Packages
Health Benefit Packages
WHO India Country
Office
Database listing
service packages
and rates across 22
GFHIS
Billing/Price
Central Government
Insurance Scheme Rates
information
CGHS No
Billing/Price
RSBY package
reimbursement rates
RSBY No
Equity
Socio-economic and
Caste Census (2011)
MRD
Socio economic status, caste,
religion, living conditions, source of
income
Source: Downey, Laura et al. 2018. ‘Identification of publicly available data sources to inform the conduct of
Health Technology Assessment in India’. F1000Research 7:245.
SECTION 2 –
PERSPECTIVES FROM THE FIELD
112
7. Overview of selected states
As part of our field research, we conducted interactions with key stakeholders and experts in New
Delhi and 6 states. The rationale for state selection is discussed in the study methodology section
of the introduction. In the next chapter, we will share the perspective of respondents (kindly refer
to Annexure A for their list) largely on the NFHS by broad themes, but also on other surveys / health
data sources. As one can see, their perspectives are, quite naturally, divergent on several themes.
Our effort has been to share them objectively from our side for readers’ consideration, irrespective
of our own views. This is the reason why we did not try to put them out in a narrative / paragraph
format, rather as points, which also helped us in presenting them in a focused and concise manner.
Let us start out with a brief overview of selected states. Table 7.1 below highlights the importance
of independent data on health through representative population surveys. To begin with, selected
states represent various population sizes – from large (UP) to medium (Maharashtra and Bihar) to
the smaller ones (Assam and Kerala). However, it is interesting to note that, though Assam’s share
of national population as well as deaths was similar (2.6% each), Rajasthan, Bihar and Maharashtra
had lower shares of deaths vis-à-vis their national population shares, with UP and Kerala – states
usually at two different ends of the human development spectrum in the country – had higher death
shares. India had the world’s largest share of deaths between 1990 and 2013 (with the exception
of 2004), with China taking a major lead since 2014, thanks to the latter’s ageing population. If one
looks at the pattern of under-70 year deaths, there is a huge variation between the two – with India
being at the top by a huge margin and China recording massive declines between 1993 and 2007
(GBD). So, although India has recently become the world’s second rather than the top contributor
to deaths, India’s share of premature deaths is much higher than China’s, which is the worrisome
part from a health perspective. Reductions in premature mortality have been part of international
commitments, including SDG 3. And this is where UP is of utmost concern – out of the 5.9 million
premature deaths (under 70) in India in 2017, more than a fifth (1.2 million) were in UP alone (GBD).
If we look at the causes of death, although a substantially higher percentage of deaths within Kerala
and Maharashtra were due to NCDs, number of NCD deaths was substantially higher in UP, given
its much larger share of total deaths. The proportion of deaths due to the traditional communicable,
maternal, neonatal and nutritional diseases was anyways substantially higher in Bihar, UP as well
as Assam and Rajasthan – the health-backward states in our selection. No wonder life expectancy
was also lower in these states, with Bihar’s performance being surprisingly better than its peers.
We have taken IMR from 2 sources – SRS for 2016 and NFHS-4 for 2015-16. Almost all respondents
in Kerala referred to differentials in their IMR from these sources. Since the NFHS-4 IMR matched
with IMR from their HMIS (2015-16, 2016-17) and CRS (2016) data – 5.6, 6 and 5.59 respectively
– they felt that the quality of the NFHS-4 data is, therefore, suspect – the quality of HMIS and CRS
113
data seen as essentially suspect and that of the SRS beyond doubt. We also see huge differentials
between the IMR of other states – most spectacularly, UP’s – from the 2 sources, Rajasthan being
the only exception. Coming back to Bihar, we see that, while it did best among its peers as far as
the SRS data was concerned, it was behind Rajasthan and Assam as far as IMR from NFHS-4 was
concerned. Figure 7.1 shows that there could be substantial variations in IMR from 3 data sources
(CRS, SRS and AHS – the latter 2 being surveys) from the same agency (Registrar General of India,
Ministry of Home Affairs). It needs to be noted that data variations from different sources does not
automatically imply superiority of one or inferiority of the other, as data integrity measure indicator
of the NITI Aayog’s Health Index hints at. While differences in concepts and methodology explain
differences in data from various sources, the imperative of relying on a particular data source for
the purposes of policy and program design and evaluation implies that a through investigation is
needed on the source of differentials on a case-by-case basis which can help select a data source
to rely on for critical matters of priority-setting and resource allocation, for instance. Nevertheless,
it is interesting to note in figure 7.1 that differentials in data are lesser in better-off states like Kerala
and Maharashtra compared to the health-backward states. So, one could possibly argue that the
quality of the administrative health data (CRS) is relatively better in the former set of states – which
also hints at a correlation between their IMR and the robustness of their administrative health data.
Figure 7.1: IMR for India and selected states from CRS 2010, SRS 2010 and AHS 2010-11
Source: ORGI 2013 and AHS 2010-11 (variance is mapped on the secondary / right axis).
The less robust the administrative health data, the more the reliance on independent survey data
should be. Nevertheless, we observed a far greater sense of keenness vis-à-vis the NFHS in Kerala
-100
-90
-80
-70
-60
-50
-40
-30
-20
-10
00
10
20
30
40
50
60
70
80
Kerala Maharashtra India Bihar RajasthanUttar PradeshAssam
CRS 2010 SRS 2010 AHS 2010-11 Variance of CRS 2010 from SRS 2010 (%)
114
than in the health-backward states. In Rajasthan, a very high-ranking health official was absolutely
against the conduct of health surveys and humiliatingly argued that the era of surveys is over and
they should be done away with, and the Government of India should rather support administrative
data – this at a time when the nationally infamous Kota child deaths were happening. This not only
reflects an antipathy toward data in general, independent data in particular, but one of the reasons
for persistent backwardness vis-à-vis a primordial health outcome like IMR.
Coming over to treatment source, we see that the reliance on government / public hospitals is the
highest in Kerala and lowest in UP. Nevertheless, out-of-pocket health care expenditure is high in
Kerala, but still lower than Bihar and UP. Given that the latter two had a much high proportion of
their population below the poverty line than Kerala – 33.7%, 29.4% and 7.1% respectively (2011-
12, Tendulkar methodology) – such expenditure may not be as catastrophic as it would be in the
latter two states. People in Kerala prefer going to specialists even for small health care concerns
vis-à-vis states like UP and Bihar (field research), which explains its high out-of-pocket expenditure.
As far as level of registration of births is concerned, there is a huge differential between the NFHS-
4 and CRS figures in the case of Rajasthan, which may be a partial cause of the Rajasthan official’s
anger against the NFHS. As far as level of registration of deaths is concerned, the figure is again
high for Rajasthan, especially vis-à-vis its peers, but since there is no independent data available,
we cannot be sure about it. It is very high in health-advanced states vis-à-vis the other 3 -backward
states (Bihar, UP and Assam). Difference in the level of registration of births and deaths in health-
backward states is also quite significant, which is not the case in health-advanced states.
From the perspective of this study, as we said at the beginning of this chapter, all these data points
highlight the importance of population-wide estimates from independent surveys, especially in the
health-backward states. Low population coverage of government facilities and high out-of-pocket
expenditures (largely made in private facilities) mean that states are not able to capture population-
wide data as part of their administrative health data systems. With private sector data still largely
not shared with governments, the only way to have population-wide estimates – with the exception
of extrapolations, which are not usually preferred – is through population sample surveys – not to
mention poor quality and highly fragmented capture of administrative data, which further enhance
the need for high-quality population-wide survey data. As we said at the beginning of the report,
this is one of the major reasons why NFHS has become eminently important in the Indian context.
Despite administrative data being the currency which policymakers across various levels deal with
on a routine basis, survey data is largely invoked for policy- / program-making and accountability
purposes on a random rather than a systematic basis. Among surveys, preference is widely given
to the government’s own survey (SRS) vis-à-vis the NFHS, coordinated and conducted by external
agencies. Inasmuch as this is the case, the governmental system continues to be caught up in the
socialist mindset of yesteryears, with the nongovernmental seen as suspicious. However, it should
be acknowledged that, despite this, the NFHS has received a lot of respect within, and particularly
outside, the government – at least, centrally, if not in the states, particularly in the health-backward.
115
Table 7.1: Overview of selected states
State Source India Rajasthan UP Bihar Assam Maharashtra Kerala
Projected population (% of national, 2016) MoHFW 1.3 billion 5.8 16.7 8.8 2.6 9.2 2.7
Total deaths (% of national, 2017)
GBD
9.9 million 5.5 18.1 7.5 2.6 8.8 3.1
Cause of death (COD, 2017) - NCDs 63.5 59.7 54.4 52.6 59.8 70.6 81.9
CoD - Communicable, maternal, neonatal and
nutritional diseases
26.7 31.8 35.9 38.5 32.2 19.8 8.8
CoD - Injuries 9.9 8.5 9.7 9.0 8.1 9.6 9.4
Life expectancy at birth (2011-15)
SRS
68.3 67.9 64.5 68.4 64.7 72 75.2
IMR (2016) 34 41 43 38 44 19 10
IMR (2015-16) NFHS-4 40.7 41.3 63.5 48.1 47.6 23.7 5.6
Ailments treated on medical advice by source:
government / public hospital (%, 2017-18)
NSS 75th round 30.1 39.8 14.1 18.5 43 25.2 47.5
Out-of-pocket health expenditure (% of total
health expenditure, 2015-16)
NHA 60.6 56.4 76.5 79.9 55.1 58.9 71.3
Children (0-4 years) whose births were
reported registered (%, 2015-16)
NFHS-4 79.7 66.6 60.2 60.7 94.2 95.1 97.7
Level of registration of births (%, 2016)
CRS
86 100 60.7 60.7 100 94 97.1
Level of registration of deaths (%, 2016) 78.1
93.3 40.2 28.3 59.8 93.7 94.3
116
8. Perspectives on the NFHS and other surveys
114
Objective (PRCs)
115
1) One of the initial objectives of the NFHS was to strengthen the survey capabilities of Population
Research Centres (PRCs) of the MoHFW.
116
Earlier, PRCs were involved with NFHS. However,
with their subsequent exclusion from the NFHS landscape, the quality of NFHS has deteriorated
– as has the relevance of the PRCs.
2) There is now a weak link between the PRCs and the government as well. There are no regular
interactions with policymakers. At least, one quarterly meeting should be conducted. ‘It is not
clear to what extent, and in what form is our research being used in policymaking. PRCs have
published a lot of research, undertaken several government projects. But there is no feedback
on our work. We are not aware if any actions have been taken based on our recommendations’.
3) It was thought that since PRCs would not be able to conduct the NFHS alone due to shortages
in their manpower, skills, etc., private consulting agencies (CAs) were identified for each state.
It was decided that data collection will be done by the PRCs and CAs together. Capacity-building
of PRC staff was also done. However, after NFHS-1, the role of PRCs was significantly reduced.
4) Earlier, the role of the Pune PRC, in particular, was quite substantial. During the first round of
the NFHS, a workshop was organized by it, in which all the questionnaires were finalized from
their DHS templates. It has also given inputs for NFHS training of the trainers of the field staff.
5) PRC Lucknow was involved in NFHS-4. Due to several issues, it had to withdraw from NFHS-
5. According to rules, payments are made to third parties / government institutes on the basis
of the number of schedules covered by them – agreement happens on a per unit rate. If there
are any savings, they are used for infrastructural development. ‘We have always been allowed
to retain them. That is how institutions like the PRCs develop further. There were certain issues
in the fourth round also despite the written agreement – the last tranche of payment was not
released to us. So, in this round, I raised a question – if we will have savings, what will happen
to them? Private institutions were allowed to keep these savings, but the same was not allowed
for the PRCs! In fact, the IIPS discouraged the PRCs to take up the NFHS. It was an extremely
unfortunate thing! The message I got from somewhere else was that the IIPS feels that private
114
Field interactions in New Delhi were conducted by the author, Priyanka Tomar (1), Divya Chaudhry (2) and Nilanjana
Gupta (3). The author conducted field interactions in Rajasthan and Kerala; 1 and 2 in Maharashtra and Uttar Pradesh;
and 3 in Bihar and Assam. Transcriptions were done by 1, 2 and 3; coded and prepared for this chapter by the author.
115
For details about the PRCs, see – https://bit.ly/3dYjQGB and https://bit.ly/3d4VKsM (13/6/2020, 17:03 hours).
116
http://rchiips.org/NFHS/nfhs1.shtml (13/6/2020, 17:05 hours).
117
parties should get involved in the NFHS and PRCs / government institutes should not be. So,
eventually, we decided not to go for the bid. The agreement we had for NFHS-4 has not been
fulfilled. If it was framed in a manner saying that PRCs were handed over the studies on a cost
basis, we would have had no issues on that. But once you are going into an agreement – firstly,
the agreement should be honored; secondly, government institutions should be more favored
vis-à-vis private parties. It should not happen the other way round. We do so many studies and
much of government funding is on the lines of cost basis. We do not have a problem with that’.
6) Since 2012, PRCs have been extensively involved with the monitoring and evaluation of NHM
Programme Implementation Plans (PIPs) and quality monitoring of HMIS data. They regularly
review and validate HMIS data. ‘When we started, the quality of data was very poor – only 20-
30 percent was reliable. But now, we can say that nearly 70 percent of the data is reliable’.
7) Given serious shortages of funding and manpower, PRC in Guwahati is almost defunct. During
NFHS-1 in Assam, it played a prominent role as the NFHS field survey agency (FSA) of Assam.
8) Currently, the PRC Kerala monitors all PIPs for Assam and few other northeastern states / UTs.
PRC Kerala is also looking after Odisha.
9) PRC Kerala is involved in conducting vulnerability analysis and primary health surveys in urban
areas, which would form the basis for deciding the location of urban health centres in the state.
10) PRCs should, once again, be involved in the NFHS. Several PRCs were also involved in DLHS.
‘Given the fact that we have more than 25 years of field-related research experience, experts
at PRCs can play a crucial role in the conduct of NFHS. We understand the nuances of such
surveys, and are equipped to deal with field-related challenges. Involving PRCs in the exercise
would definitely improve the data quality of the NFHS and other surveys’.
Thematic scope
1) The issue of multiple health surveys versus one health survey has been extensively discussed,
especially in 2012. Consolidating all health surveys into one could have addressed the problem
of differences in sample design / frame, which could have helped in linking several indicators.
2) NFHS should be as comprehensive as it could be. It is not possible to plan such a large-scale
survey for different themes as it is extremely resource-intensive. For households too, it is not
possible to spare time again and again for different surveys. NSS schedules are also elaborate,
and have expanded overtime. NFHS should follow the same trajectory.
3) NFHS needs to evolve in the light of the changing demographic and epidemiological profiles.
4) The scope of NFHS may be revised to cater to the data requirements of NHP 2017.
118
5) NFHS-5 covers a total of 29 out of 44 SDG health-related indicators. The number of indicators
covered will be different depending on whether the calculation is based on the standard global
SDG indicators or the India-specific SDG indicators.
6) It is not reasonable to have one health survey that provides all required data / indicators.
7) If NFHS questionnaires are expanded further, more data collection will happen, but the quality
of data will suffer. After 1 or 1½ hour of interview, it is very difficult to maintain the respondent’s
attention and interest. Besides, from an ethical point of view, it is not justified to ask so many
questions, especially when respondents are not being paid / offered any reward / incentive for
providing their valuable time.
8) In view of differences in objectives, design and periodicity of different surveys (such as SRS,
NSS, NMHS, etc.), mapping of indicators across the surveys may not yield the desired results.
9) NFHS should continue to focus on its primary strength (RCH) and drop all other questions.
There should be other surveys to cover other aspects of health on a regular basis.
10) In data collection, the traditional approach is to collect everything possible. However, it defeats
the very purpose for which data is being collected. This approach has also been applied in the
design of the NFHS in recent years. It should continue to focus on its primary strength (RCH).
11) IIPS has its expertise in demography. They tend to push their own agenda in the NFHS.
12) While certain international sponsors wanted to include adolescent health in NFHS-3 and -4,
IIPS was reluctant to do so. It influences the Technical Advisory Committee (TAC), which is
the nodal body responsible for the design and implementation of NFHS. It knows how to go
around and get things done as it wishes.
13) ICMR is too biomedical and is not social scientific. If NFHS was to be conducted by them, its
social scientific dimension will be affected.
14) Surveys like the NFHS also need to provide morbidity statistics.
15) Although the scope of NFHS is being broadened to include NCDs and other emerging health-
related challenges, it is necessary to conduct cost-benefit analyses before augmenting the
scope of large-scale surveys such as the NFHS.
16) We need better mechanisms to capture the prevalence of NCDs / their risk factors in surveys.
17) NFHS-5 will provide estimates on 5 types of disabilities.
18) NCDs and disabilities should not be included in the NFHS.
19) Indirect estimation of DALYs is possible using the NFHS data.
20) Indicators to monitor performance of national health programs should be included in NFHS.
21) Only a few crucial health outcome indicators should be prioritized and measured regularly.
119
22) It is better to restrict the number of questions and focus on key aspects. Rather than increasing
the size of questionnaires, emphasis should be on getting complete and accurate information.
23) NFHS does not collect information on causes of death (CoD) in detail. In large-scale household
surveys, it is not feasible to canvas modules based on CoD.
24) NFHS indicators are not mapped to WHO’s ICD.
25) Under NFHS-5, data on biomarkers is being collected from children aged 0-5 years, women
aged 15-49 years and men aged 15-54 years.
26) It has been decided to conduct HIV testing in every alternate round of NFHS. In NFHS-4, HIV
testing was done. HIV testing will now be done in NFHS-6.
27) As incidence of mental health issues (depression, anxiety, etc.) is rising rapidly, it is important
to conduct a separate survey on mental health. Further rounds of NMHS should be conducted.
Further, in order to destigmatize mental health, it is important to mainstream its evidence.
28) It is difficult for NFHS to include questions on socially sensitive issues such as mental health.
29) As communicable diseases like kala-azar are endemic to a few states / districts, questions
related to them should be avoided in national health surveys. Alternatively, conducting local
surveys can significantly contribute to the understanding of determinants and estimating the
incidence of such diseases. However, questions related to certain communicable diseases
which are of public health importance at the national level (malaria, dengue, etc.) should be
covered in the national surveys.
30) NFHS should also include indicators like out-of-pocket health care expenditures.
31) As migration is one of the three components of demography, NFHS should include questions
to evaluate the proportion of population which has migrated (within or across states) and the
kind of health concerns being faced by it. There is substantial migration from the backward
districts of Maharashtra to cities like Mumbai and Pune, for instance. There should be focus
on migrant health as well. The Maharashtra health department is trying to address the health
challenges being faced by sugarcane farmers, who are mostly intra-state migrants, for
instance. Due to non-adherence of treatment, incidence of drug-resistant tuberculosis is rising
among this section of the population.
Geographical scope
1) Health is a state subject. There are different state health requirements, and sometimes national
surveys cannot cater to these state-specific requirements.
2) A data strategy is needed for Centre, states and districts as well as for different kinds of health
conditions / diseases.
120
3) It is possible for states to design and conduct their own health surveys. However, unlike NFHS,
inter-state comparisons would be difficult through such surveys.
4) State-specific surveys may be conducted to complement the NFHS. NFHS state samples may
be used to address the problem of inter-state comparability and national aggregation.
5) States must be consulted in the design and implementation of NFHS. This is important from
the perspective of highlighting the burden of certain state-specific health challenges in a large-
scale health survey such as the NFHS.
6) It is good to have state- / district-level health surveys in addition to one national health survey.
It is impossible to include everything in a national health survey. Maharashtra and Tamil Nadu
conduct independent health surveys, for instance, as part of which they collect data up to the
block level. Such surveys could be designed based on the epidemiological profile of the states.
7) Customized surveys at the state level would be a better strategy. For instance, in Maharashtra,
RCH issues are not as worrisome as micronutrient deficiency issues like stunting. Hence, from
a policy perspective, we should channelize more resources for dealing with stunting and other
nutritional concerns vis-à-vis RCH. National surveys will be useful to compare the performance
of the states / UTs.
8) It is important to understand the data requirements of every state – for instance, illicit drug use
is very high among women in northeast states which also has severe implications for the health
of children. But, due to lack of data, we have very limited understanding on this issue. Similarly,
there are several other important factors which are not being captured in the NFHS that have
a bearing on maternal and child health in Assam.
9) From the perspective of estimating the burden of NCDs, state-level data is sufficient. However,
within each state, there can be a lot of heterogeneity. Based on whether the population is tribal
/ non-tribal and topography of the state, there can be certain pockets where the prevalence of
modifiable risk factors – such as tobacco and alcohol consumption, physical activity, etc. – is
higher vis-à-vis others.
10) NFHS should also provide data on mortality indicators at the district level as there is no reliable
source for it at the moment.
11) Due to small sample size, NFHS cannot provide district-level TFR, IMR, U5MR, etc. estimates.
12) NFHS consists of state and district-level modules. Out of 114 indicators, district-level estimates
are provided for 93 indicators.
13) We need SRS data at the district level as well.
121
14) Districts should be the data units. They have their data needs. Presently, local data collection
happens for higher requirements. Local data requirements are neither assessed nor taken into
consideration. Even the HMIS is designed for central government’s data requirements.
15) Another challenge is that small-big town differences are not taken into consideration in NFHS
as city estimates are difficult to calculate – one cannot calculate estimates for cities with less
than 10 lakh population. Smaller cities should not be ignored as these are the ones which are
providing services in rural areas.
16) Although we are focusing on districts now, we should focus on agro-climatic zones, as districts
are administrative, not natural, units. Interventions can be similar in similar agro-climatic zones.
For states like UP, data by agro-climatic zones would be a huge enabler.
17) There is a need for local level data. In Kerala, a lot of health-related activities are happening at
the panchayat level. Administrative data is not very reliable. Therefore, local population-based
surveys / studies would be helpful.
18) It is important to devise a strategy for better use of health data. Before data collection, it is vital
to consider local data requirements and educate local experts about the use value of this data.
19) Population health surveys should also try to cover the impact of local determinants on health.
For instance, if shopkeepers in a locality sell non-iodized salt, people there would be at risk of
goitre. Similar problems can arise when soil in a region does not have adequate salt content.
20) It is difficult to customize in large-scale sample surveys. Customization of questions related to
HIV was done during NFHS-2. Different questions could be included for different NFHS zones.
Collecting data on same parameters across the country offers a comparable view, but misses
out on region-specific issues. For instance, the entire dynamics change across regions in UP
– the factors that could be responsible for a particular outcome in Lucknow, for instance, would
be very different for Eastern UP. Nevertheless, while desirable, customization of questionnaires
would make the entire process very tedious and data quality may be at stake.
Periodicity
1) A major issue vis-à-vis health surveys in India is their frequency. There should be a mechanism
to conduct health surveys annually. Countries in Africa are already designing / implementing
mini-DHS – of which NFHS has been a part – to capture critical health data at shorter intervals.
2) Beyond health, frequency of other surveys / estimates should also be improved. For instance,
unit-level data from the 2011 Census is not yet available in the public domain; the latest poverty
estimates are available for 2011-12, so on and so forth.
3) A periodicity of 3-4 years should be fixed for the NFHS. Availability of data from other sources
such as the Census should be expedited.
122
4) Due to its periodicity in particular, NFHS data does not contribute much to state-level planning
in the health sector.
5) For policy purposes, there is a need for real-time data, which the NFHS does not provide. Lack
of availability of quality survey data at regular intervals negatively affects the quality of regular
policymaking and program monitoring and evaluation in the country.
6) Small / region-specific / special surveys should be conducted between 2 NFHS rounds.
7) One national health survey must be conducted independently at regular intervals. In addition,
local health surveys may be conducted on the basis of local requirements.
8) Annual estimates of key outcome indicators like IMR and MMR are not needed as interventions
to tackle them take time to materialize. Annual estimates put undue pressure on states to show
improvement in their performance. During the design of NFHS-4, involved experts had several
rounds of discussion on this issue. A consensus was reached to capture such critical outcome
indicators at intervals of 5 or 10 years. However, annual estimates of input or access indicators
are indeed worthwhile.
Sample design
1) The concept of village is different in North vis-à-vis South India, especially Kerala. The sample
design and house-listing of the NFHS needs to be accordingly revised.
2) Household mapping was not done properly for NFHS-4 and -5. Kerala’s IMR data from NFHS-
4, therefore, should be taken with a pinch of salt.
3) Backward areas were selected in Rajasthan for the NFHS, leading to skewed data for the state.
4) There are practical challenges with NFHS sample design – for example, accessibility in tribal
/ remote areas.
5) NFHS sample size does not permit obtaining reliable data on indicators like MMR. We need to
find ways to bridge such data gaps in the NFHS.
6) NFHS sample design is primarily oriented towards RCH issues – an alternate sample design
is needed to cover other health issues in the NFHS.
7) Selecting a sample for demographic estimates and selecting a sample for estimating the NCD
burden are two different things. For demographic estimates, there are no issues with the NFHS
methodology, but for NCD-related indicators, there are problems with its sample design. There
are statistical techniques to tackle such challenges – for instance, assigning weights. However,
it is preferable to have different sampling strategies for such varying health themes.
8) It is unnecessary to include men in all rounds of the NFHS.
123
9) Adolescents are not being covered under NFHS, which is another major concern. NCD burden
is quite high among adolescent age-groups as well.
10) Due to its large sample size, NFHS is vulnerable to data quality issues.
11) NFHS sampling is such that after a point, you start getting similar responses.
Questionnaires
1) Although the NFHS predominantly follows the DHS questionnaire template, it has undergone
substantial revisions based on Indian requirements.
2) NFHS questionnaires are finalized by IIPS in discussion with the central ministries. State health
departments are not consulted.
3) NFHS questionnaires are extremely lengthy, which commonly leads to respondent fatigue and
reduced probability of obtaining responses, especially to questions towards the end.
4) After an hour, it is impossible to hold the attention of the respondent. Besides, from an ethical
point of view, it is not justified to ask so many questions to a respondent, especially when they
are not paid or given any kind of reward / incentive for providing their valuable time.
5) In an extensive survey like NFHS, questions on important / thrust areas should be asked first.
6) Though context-specific questions should be included in the NFHS to capture context-specific
data, length of NFHS questionnaires cannot be further increased. Covering everything under a
single survey is not suitable either from a survey design, fieldwork or data quality perspective.
7) It is difficult to customize in large-scale sample surveys. Customization of HIV questions was
done during NFHS-2. Additional questions can be added for different NFHS zones. However,
collecting data on same parameters across the country helps in offering a comparable picture,
but misses out on specific issues relevant for various regions. For instance, dynamics change
across regions in such a large state as UP – factors which could be responsible for a particular
outcome in one region might be very different for another. Nevertheless, although desirable,
customization of questionnaires would make the process very tedious and data quality might
be affected.
8) In-depth view is not emerging from health surveys like the NFHS.
9) There is limited local autonomy and no scope for innovations in the questionnaires. You cannot
enter challenges to health that you observe during the fieldwork or respondents’ views in the
questionnaires.
10) Qualitative observations of the field staff / respondents should be enabled in the questionnaires
and included in the survey reports. Larger surveys in India are deprived of qualitative research
124
component. Qualitative assessments should be made a part of all health surveys. In diseases
such as tuberculosis, such assessments play a vital role in ensuring treatment adherence.
11) However, if we want to capture qualitative data, then only key, specific themes / higher disease
burden regions should be covered – it will not be feasible to cover the entire country / states.
The qualitative dimension should also be very focused / specific.
12) Another challenge is that field investigators may not be able to get the right kind of information
for qualitative components. Government officials are better suited for conducting focus group
discussions. Supplementary survey reports can cover such discussions. IIPS can collaborate
with states for undertaking such an exercise.
Biomarkers
1) Although estimating the prevalence of risk factors is important, collecting data on biomarkers
as part of sample surveys is not desirable and reliable as results can be affected by immediate
factors – for instance, blood glucose by sugary food consumption. Asking people about their
immediate consumption might not necessarily be helpful in low literacy / awareness contexts.
Rather than trying to estimate point prevalence of risk factors – as cross-sectional surveys like
the NFHS do – it is desirable to have estimates of period prevalence.
117
Longitudinal surveys
could still help. Even doctors ask for several readings taken at different points of time to assess
the incidence of diabetes or hypertension, for instance.
2) People in urban areas are more reluctant for biomarker assessments; in rural areas, they are
much more willing to share their personal information as well as undergo biomarker tests.
3) Adequate treatment facilities and referral mechanisms should be available to respondents who
test positive in biomarker tests as part of surveys.
Field staff
1) Extensive training should be provided to the field staff. The survey capabilities of many are not
up to the mark.
2) Training provided to NFHS field staff is of differential quality – so is the NFHS data quality.
3) Surveys are generally conducted by students who are not motivated enough due to temporary
nature of the job. At the same time, many students quit the surveys after completion of training
to go for better avenues. This is one of the biggest challenges for field survey agencies (FSAs).
117
Point prevalence refers to prevalence measured at a particular point in time. Period prevalence refers to prevalence
measured over an interval of time. It is the proportion of persons with a particular disease or attribute at any time during
the interval. https://www.cdc.gov/csels/dsepd/ss1978/lesson3/section2.html (12/6/2020, 23:10 hours).
125
4) Skills of field enumerators (FEs) are crucial in data collection and quality. The manner in which
a question is being asked makes a lot of difference in the kind of response that the enumerator
gets during interviews. ‘It is important to build a rapport with the interviewees; only skilled FEs
know how to do that. During one of the NFHS rounds, I realized that, even in a city like Mumbai,
women are ready to share more if the interview is done properly. I got a lot of useful information
about aspects of women’s health – for instance, mental health during menopause and due to
domestic violence. However, there were no options in the NFHS questionnaire to capture these
aspects of women’s health. If interviewees are assured that their data is anonymized and will
not be shared, they are more willing to discuss issues in detail’.
5) The sheer volume of NFHS questionnaires makes field work a daunting task for data collectors.
Using IT has substantially reduced their burden, but there are challenges which they still have
to deal with. For instance, they have to be trained in IT tools, and it becomes difficult to check
errors in cases where the data has been wrongly entered. Field Checked Tables (FCTs) allow
entering data on smartphones within a particular limit. There are more than 10,000 fields in a
questionnaire. FCTs are embedded in questionnaires. Through this method, however, only key
questions can be verified. Further, even though a majority of present-day surveys make use
of geo-tagging, this feature often results in providing inaccurate location coordinates.
6) FEs should be compensated properly. It is extremely difficult to find well-qualified enumerators
who are willing and motivated enough to work at such low salaries as the NFHS offers.
7) The accommodation and security of FEs, especially females, is a major concern that needs to
be taken care of in a better way by the FSAs.
118
8) Low FE incentives and limited FSA accountability adversely affect the quality of NFHS data.
9) Government rather than private agencies should be involved in the conduct of NFHS as they
not only have greater expertise, experience and integrity, but better logistical and manpower
arrangements for conducting such large-scale surveys.
10) Like NSO, local field investigators should be involved. Government agencies can do that, given
their vast local networks.
Field work
1) When NFHS-1 was conducted, completing 2½ questionnaires a day, on average, was the norm
for FEs. Now, questionnaires are much lengthier and FEs are under enormous pressure to fill
more number of questionnaires a day. FSAs involved are completely profit-making companies
118
There have been some serious allegations vis-à-vis violation of the rights of the NFHS field staff. https://bit.ly/2ztZSod
and https://bit.ly/2UFZDOc (13/6/2020, 17:11 hours) is one set of writings that has raised a number of issues.
126
and do not care much about data quality – they want to finish the survey in the least possible
time and with minimal resources.
2) NFHS-1 FEs in Bharatpur (Rajasthan) did not even go to the field, filling out responses at home.
3) Unemployed students are hired to conduct the NFHS field work in Rajasthan – they sit at home
and fill out the responses.
4) It is extremely difficult to collect data in metropolitan cities – people are not willing to participate
in surveys. Response rate is quite low in these cities. Some incentive mechanisms need to be
developed. They are more reluctant for biomarker tests / measurements. People in rural areas,
on the other hand, are much more willing to respond as well as undergo tests / measurements.
5) Given its geographical and cultural complexities, data collection in Assam is very challenging.
For instance, people come from different cultures and speak different dialects, so canvassing
the survey in one language leads to poor understanding of questions and, therefore, poor data
quality. Furthermore, political turmoil and fragility makes it difficult to reach out to respondents.
NFHS FSAs are not conducting surveys in Assam as per stipulated norms.
6) The biggest operational challenge vis-à-vis the conduct of the NFHS is that respondents do
not agree to respond to many of the questions. Its high overall response rate is misleading.
7) There is respondent-fatigue in NFHS; questionnaires are too long; people give socially desirable
responses; there is little respondent privacy – sometimes, the whole village is standing when
people are responding.
8) There are various provisions to ensure respondent privacy in the NFHS. If there is no privacy,
FEs are told to skip sensitive questions.
9) ANMs’ help is sought for conducting NFHS in very critical situations.
10) There have been difficulties in conducting NFHS-5 in some places due to fears related to the
CAA (Citizenship Amendment Act) and NRC (National Register of Citizens). These fears would
continue to pose problems for any survey-related field work as well as the quality of responses.
Data quality
Data quality-related issues have been highlighted under several themes above. Let us highlight a
few more here.
1) Kerala’s IMR as per NFHS-4 became a huge issue in the state. No one was willing to believe
it. The state government as well as academics believed that IMR in Kerala has not gone below
10 – as reported by the SRS in 2016 (NFHS-4 reported it as 5.6). This made people suspicious
of NFHS-4 data more generally. They felt that the SRS is more reliable.
127
2) With 5 lakh births and 2,700 infant deaths in Kerala,
119
reliable estimates of IMR are difficult.
3) IIPS is very particular about data quality – they try their best to ensure it.
4) Data fudging is not possible under NFHS – use of CAPI eliminates the need for separate data
entry, and there are data consistency checks as well as GPS to ensure data quality.
5) Quality varies even if the same agency is managing the NFHS fieldwork in different states. FEs
would be different anyways, for instance.
6) Timely dissemination of survey data is essential; otherwise, it pertains to a period which is no
more relevant from a program perspective.
7) Survey data can also be wrong; why only distrust HMIS data? There should be a methodology
to tackle the variance between HMIS and survey data.
Stakeholder involvement – National
1) Before each round of NFHS, there is detailed consultation with different programme divisions
of MoHFW and other ministries on their data requirements. Accordingly, indicators are
finalised and questions are added or aligned to SDG health indicators.
2) State level health authorities have never been consulted for the design and conduct of NFHS.
3) States like Rajasthan do not take interest in the design of NFHS. Some states represent states
as a whole. There is proactive participation, for instance, from Kerala and Tamil Nadu.
4) States must be consulted in the design and implementation of NFHS. This is important from
the perspective of highlighting the state-specific disease burdens and health challenges in a
large-scale survey such as the NFHS.
5) Consultation workshops and extensive meetings with relevant state-level government officials
and other stakeholders should be done before NFHS is canvassed in the states.
6) Since achieving consensus with states is almost impossible, the National Statistical Institute in
Mexico does not consult states in the design of surveys, and this is completely justified. States
should be consulted only from the perspective of implementation of health surveys.
7) The National Centre for Disease Informatics and Research (NCDIR) / ICMR provided questions
on NCDs for NFHS-4 – they were subsequently modified.
8) ICMR is proactively involved in the design of NFHS-6, and will be for its conduct as well.
119
To be precise, there were a total of 496,292 births and 2,774 infant deaths in Kerala in 2016 (Annual Vital Statistics
Report – 2016. Vital Statistics Division, Department of Economics and Statistics, Government of Kerala, Trivandrum).
https://bit.ly/3e0OU8G (13/6/2020, 21:34 hours).
128
Stakeholder involvement – International
1) When the first round of the NFHS was conducted, IIPS and MoHFW had a much bigger role
to play as there were not many international donor agencies – only USAID was involved then.
2) Give money and your questions will be included in the NFHS – it very donor-driven.
3) With respect to the design of NFHS, the CDC model has been replicated in India, which has
not been modified as per Indian requirements. Funding agencies prioritize their own agenda.
4) NFHS is conducted independently and is not driven by the DHS model.
5) ICF is not involved in the process of selection of indicators for NFHS.
6) For NFHS-5, technical and financial support is being provided by ICF and USAID respectively.
7) Some international donors wanted to include the component of adolescent health in NFHS-3
and -4. IIPS was reluctant to do it. It knows how to go around and get things done as it wishes.
8) USAID will not fund the NFHS next time. Along with it, ICF will also go. IIPS largely manages
NFHS-related logistics and training – it is Fred Arnold and his team at ICF who provide the
more valuable technical guidance – for instance, how to frame questions, which requires a
high level of technical expertise and experience. If they are not included, the NFHS survey
design and data quality would be seriously affected as we do not have that sort of expertise in
the country. Nearly 90% of pass-outs from the prestigious Indian Statistical Institutes (ISIs) go
abroad.
Data dissemination
1) Since dissemination of the final NFHS data takes a lot of time, it would be useful if some interim
data is made available to states before final data is released. Efforts should also be made to
disseminate the final data as early as possible.
2) All NFHS data should be put out in the general public domain in a user-friendly manner to
promote transparency and generate public accountability (senior Rajasthan district official).
3) Data curation needs to be done from the perspective of various stakeholders. A lot of countries
are doing that. India is also trying, but things are still at a preliminary stage.
Data analysis
1) Data analysis is very preliminary within the government system at both central and state levels.
We need separate teams for data analysis and data feedback to government departments. We
cannot expect government officials to themselves search for / use relevant data on their own.
129
The entire data cycle – design, data collection, dissemination, analysis and use – needs to be
seen as a continuum and carefully developed at both the central and state government levels.
Otherwise, wasting resources on just data collection through multiple sources does not make
sense. We need to have the entire package from the beginning till the end. Simply collecting
raw materials for food does not help – we also need to cook and serve the food in a palatable
manner. Right now, we have random raw materials from a variety of sources. Focus needs to
shift to properly cooking and serving the food.
2) In addition to simply collecting and disseminating data, we have to improve data analysis so
as to strengthen health intelligence. Currently, a substantial proportion of human resources in
the health sector are engaged in routine administrative work and have little time for analysis
of data which is being collected. There is shortage of human resources even in National AIDS
Control Organization (NACO) and the ORGI / MHA, where data is extensively analyzed.
3) Training institutes are being developed for data analysis and quality checks, with funding from
the Centre.
Data use
1) There is enough evidence of NFHS data use for policy and program purposes in India, as input
in the erstwhile five year plans to the recently launched National Nutrition Mission. The MoHFW
has used NFHS data as scientific evidence for various policy decisions – including for adoption
of a target-free approach in 1996, setting goals for the National Population Policy 2000, framing
of different national health policies, etc. Notable policy and program changes have also been
based on NFHS data in areas such as domestic violence, child marriage, menstrual protection,
sanitation and caesarean-section deliveries. In addition, various states have brought out state-
specific population and health policies and programs based on state-level findings of various
rounds of NFHS.
2) NFHS has significantly contributed to the formulation of MCH policies in the country.
3) NFHS has contributed to generating valuable baseline data and showcasing impact of health
policies and programs on health outcomes. The fact that NFHS provides data by background
characteristics also makes it an important source of health data.
4) Governments refer to SRS and NFHS data for the formulation of health policies and schemes.
5) NFHS and SRS are also used for developing government proposals, planning and evaluation.
6) NFHS and SRS are being referred to, but we cannot totally rely on national surveys since the
situation in Kerala is quite different.
7) NFHS is never used by the Directorate of Economic and Statistics (DES) in Kerala; it uses data
collected at the local level.
130
8) As survey data cannot show the impact of health programs on health outcomes, it is impossible
to evaluate health programs on the basis of health surveys.
9) Surveys should be able to show policy implications. More often than not, surveys digress from
this objective and only focus on generating factsheets and conducting additional rounds.
10) NFHS is being used rigorously for research purposes.
11) Data use is problematic – too much useless data is being collected.
12) Instead of collecting more data, the focus should be on making the best use of available data.
13) For planning purposes, there is a need for real-time data, which the NFHS does not provide.
14) The usability of NFHS data for planning and monitoring purposes is limited.
15) NFHS and other national health surveys have limited relevance in state-level health planning.
Although its data is used for the verification of data generated from state-level sources, state
planning decisions are mostly based on data from HMIS and other government sources. Due
to periodicity, NFHS data does not contribute much to state-level planning in the health sector.
16) Use of NFHS data is limited in Assam. Sometimes, NFHS factsheets are used while deciding
on state level schemes and programs. Lack of availability of NFHS data on a regular, periodic
basis is one of the major reasons for its limited use.
17) In Assam, administrative data sources like HMIS and disease registries are being extensively
used. There is a high dependence on HMIS data, especially for monitoring the implementation
of health schemes / programs, and analyzing emerging data trends at district and block levels.
18) Despite the DLHS being discontinued, it is still being widely used by policymakers in Assam.
19) Using voluminous NFHS reports is a challenge for government officials, especially if they have
to refer to different state reports for comparisons. Presenting data in a user-friendly style has
to be prioritized to enable ease of data use. ‘I use a spatial visualization tool based on NFHS
data developed by a private agency, Riddhi. It is very convenient’ (senior Kerala NHM official).
20) In order to make data more self-explanatory, efforts should be made to make it available in a
more useable format.
21) Policymaking is largely politically- rather than evidence-driven.
22) Governments do not take data seriously.
23) Central and state governments do not give sufficient importance to data. Most officials do not
appreciate the importance of data.
24) Fear of accountability hinders government officials from using independent data and promotes
the collection of low quality data within the system. If, somehow, we could delink accountability
131
and data collection / use, it would not only lead to more reliable administrative health data, but
also greater use of survey data from the NFHS and other sources (senior Kerala NHM official).
25) It is important to devise a strategy for better use of health data. Before data collection, it is vital
to consider local data requirements and educate local experts about the use value of this data.
26) Sensitizing government officials, capacity-building and -strengthening for data use is essential.
27) Capacity to use data for decisions needs to be developed (senior Rajasthan official).
28) Data use training was given to health department officials in Rajasthan. ‘Trainers got the money;
trainees got time-off from regular work and refreshments; everything remained as usual’.
29) Concerted efforts need to be made to enhance the usability of survey and administrative data
in health-related planning.
30) All bureaucrats say survey data like the NFHS has issues. Surveys help in checks and balances
on the activities of relevant departments.
31) NFHS and SRS data is used by Rajasthan state government officials for internal accountability
and monitoring purposes. State and district officials take NFHS data seriously.
32) NFHS data is very useful (senior Rajasthan district official).
33) Officials in the Directorate of Health Services in Pune use NFHS to verify HMIS data.
34) While one can use NFHS data as a tool to validate the HMIS data, it is technically inappropriate
to have such an expectation from population health surveys.
35) Surveys have population-based sampling; their data will never match with the HMIS data, which
is largely government facility-based. It is wrong to use survey data to validate the HMIS data.
36) After NFHS estimates, it was realized that HIV prevalence is not as high as it was reported by
administrative data – the same patient was being counted as HIV-positive in every hospital he
was seeking treatment in, leading to overestimation of the HIV burden. In this way, NFHS has
helped policymakers in Maharashtra to understand the true burden of HIV.
37) The IDSP unit in Pune also occasionally refers to indicators from NFHS, including those related
to child health, vaccination coverage, tuberculosis and household characteristics like drinking
water source and indoor air pollution.
38) A unique and strong dimension of population health surveys like the NFHS is that they provide
respondent level data – ‘in a moment, you can get the entire kundali (horoscope) of a person’.
And a lot of disaggregated information is made available by background characteristics. Even
census gives data at state, district, block, village and ward level, but it is impossible to get the
complete set of information about an individual through census data.
132
Other health surveys
1) The SRS data is of no relevance to RGI / MHA; it is only conducted to independently assess
the impact of health policies. This is precisely the reason why there is no coordination between
the MHA and MoHFW vis-a-vis the SRS.
2) There was a proposal to bring SRS under the administrative scope of MoHFW. However, since
SRS is managed by Indian Statistical Service (ISS) officials, they do not want SRS to slip away
from their authority.
3) The RGI / MHA is very sensitive about SRS and does not make raw data available to the public.
4) Although SRS collects data by background characteristics like religion and caste, it does not
publicly disseminate its data by these characteristics. It should do so.
5) The data quality of the SRS is very good – it is the most reliable health data source.
6) The SRS was initiated with a view to generate reliable and continuous data on vital indicators
– birth rate, death rate, IMR and TFR. And for these indicators, its data is more robust than the
NFHS since it comes from government sources and focuses on estimation of only 4 indicators.
7) Compared to the NFHS, SRS is much more widely used because SRS data is made available
at shorter intervals.
8) Government of India has mandated the use of SRS for core outcome indicators / vital statistics.
9) MMR data in general, including that of the SRS, is not quite reliable. There are issues with the
reporting of maternal deaths – it may legally implicate the husband’s family, for instance. And
collection of data on such a sensitive theme by a government agency – that too the MHA / RGI
– makes it more complicated.
10) Sometimes, SRS FEs take data from PHCs rather than collecting it from the field on their own.
11) SRS is based on a sample of 1 percent of the Census population, which is why the SRS can,
at best, provide estimates.
12) SRS should also provide data at the district level.
13) Since proportion of medically certified deaths in India is very low, the RGI developed a sample
frame (based on previous Census, followed for a period of 10 years) to conduct verbal autopsy
(VA) surveys. The primary objective of VA surveys is to build up a statistical database on the
most probable causes of death for rural and urban areas using lay diagnosis reporting method
(VA). However, since there is usually a considerable time-lapse between the occurrence of
death and the SRS-CoD surveys, the problem of recall bias is quite pervasive in these surveys.
14) In collaboration with RGI, the Million Death Study (MDS) was undertaken by Centre for Global
Health Research (CGHR), Canada, between 2004 and 2013. Any deaths that occurred in the
133
nationally representative households in this duration were assigned a probable cause on the
basis of VA. For adequacy of sample size, CoD data for 3 years was combined.
15) Since 2015, Dr Anand Krishnan, AIIMS Delhi, has been given charge of the SRS-CoD surveys.
Apparently, things are not progressing according to the expectations.
16) HMIS reports are still using DLHS, mostly for facility-level data which is not being captured by
NFHS. ‘I don’t think there is any difference in data quality of NFHS and DLHS. We are not very
comfortable using the HMIS data. ANMs are worried about achieving their targets, that’s where
a lot of data gets corrupted. In some cases, incentives work in another direction – for instance,
there are cases where a borderline malnourished child is declared severely malnourished just
to meet the targets’.
17) The National NCD Monitoring Survey (2017) was developed on the basis of WHO’s STEP-wise
approach to surveillance (STEPS). Modules have been given to certain states for customizing
and implementing the survey at the state level.
18) A Kerala Information Residents Association Network (KIRAN) survey was conducted to collect
data vis-à-vis NCDs – on lifestyles (dietary habits, physical activity, etc.) as well as disease and
treatment patterns – by the Kerala Directorate of Health Services and Achutha Menon Centre
for Health Science Studies (AMCHSS), Trivandrum. All 14 districts were covered and with the
use of electronic tablets, data was available in real-time to these agencies.
19) The Kerala DES’ NSS Division has been proactively engaged in conducting NSS with matching
samples to provide sub-state estimates, which is not possible through the national sample. As
part of 71st round (January-June 2014), a report titled ‘Health in Kerala’ was published by the
DES on the basis of the state sample data on health-related consumption.
20) No health survey is being conducted by the UP government at the moment. Some proposals
are being developed within Department of Health and discussions are ongoing for conducting
one. As part of NHM activities, 2 surveys – the Annual Family Survey (AFS) and Annual Survey
on NCDs (under the Health and Wellness Centers initiative of GoI) – are being conducted. The
principal objective of the latter survey is to undertake screening of household members above
30 years of age for NCDs and related risk factors. A Community Based Assessment Checklist
(CBAC) questionnaire has been developed for NCD screening, including questions to assess
family history and associated risk factors. Scores are assigned and high-risk cases are referred
to nearby government health facilities. However, based on these surveys, no reports are being
published. Both the surveys are being conducted by the ANMs and ASHAs.
21) Household surveys are being conducted by CARE India as part of the Bihar Technical Support
Unit (BTSU) activities, covering all districts and blocks in the state. ‘In these surveys, our main
priority is to focus on households with 0-23 month old children. Sample size for each population
134
group is 15,000 to 16,000 households in all rounds. Apart from basic demographic indicators,
maternal and child health, immunization and family planning are covered in every round’.
22) While challenges in Bihar’s health data ecosystem are similar to those observed in other states,
there are certain key differences. Firstly, several development partners are working in different
regions of the state to strengthen delivery of health care services. Though a number of surveys
and data collection activities are carried out by the agencies, no attempts have been made to
integrate their data. Data usability in terms of planning and health-related decision-making has,
thus, remained limited. Secondly, as a significant share of population is located in remote and
isolated regions of the state, it is difficult for surveyors to include them in the surveys. Thirdly,
due to low literacy levels, survey respondents are often not aware of ongoing health programs
and schemes, which often renders certain evaluation and survey exercises futile.
Miscellaneous
1) Government of India should allow WHO and UNICEF to conduct health surveys in the country.
2) A comprehensive sampling frame should be developed by nodal statistical agencies to ensure
some degree of uniformity across surveys. Due to variations in sampling design / frame across
surveys, interoperability in health survey data is limited.
3) Data extrapolation and data integration exercises should be undertaken to facilitate evidence-
based policymaking.
4) Monitoring and evaluation of various health programs is not organised, and a lot of work needs
to be done on this front. Independent evaluations of health programs constitute a major lacuna
in the health sector. Health survey data can be used by independent agencies for this purpose.
5) There seems to be a lot of focus / fascination with IT. However, IT cannot compensate for weak
statistical capacity and recognition of importance of data for evidence-based decision-making
within the system. It was paradoxical that, despite so many MIS portals, a focus on data is seen
as something which is at odds with a focus on service delivery, indicating that the importance
of using data for evidence-based decision-making and service delivery is not realized. Several
senior statistical officers also point out that data use is weak. The MIS data is used to develop
reports / bureaucratic reporting purposes. IT can help in this regard too through user-friendly
data visualizations / dashboards, but ultimately people in the system – from top to the bottom
– have to realize the importance of data and evidence-based decision-making. There has to
be a focus on developing capacities at various levels for data use.
6) ‘A data culture is missing in India’ (senior multilateral agency official in New Delhi) – that needs
to be developed.
SECTION 3 –
HEALTH SURVEYS IN THE UNITED
STATES, CANADA AND UNITED KINGDOM
135
9. United States
The history of sample health surveys in the US goes back to the first National Health Survey (NHS),
carried out during 1935-36 by nearly 6,000 unemployed welfare recipients, who collected data on
chronic diseases and disability from approximately 2.8 million people in 737,00 urban households
across 19 states (Weisz 2011). Subsequently, a series of health survey methodology experiments
were carried out and the institutional framework for health statistics was laid down. In 1949, the
US Congress established the National Committee on Vital and Health Statistics (NCVHS) to serve
as – 1) the statutory review and advisory body to the Secretary, Department of Health and Human
Services (DHHS) on national health information policy; 2) a forum for stakeholder interactions on
health data to inform DHHS, state as well as private sector health data decision-making.
120
In 1956,
the National Health Survey Act was signed into law to enable a series of continuous health surveys.
The NHS founders did not envision a single, but ‘a program of surveys, using different approaches’
and with different objectives as data techniques and needs evolve. In 1957, what is now known as
the National Health Interview Survey (NHIS) – the principal source of health data on the US civilian,
non-institutional population – was launched with interviews in 36,000 households (Haywood 1981:
195).
121
In 1960, the National Center for Health Statistics (NCHS) – federal government’s principal
health statistical agency
122
– was established, becoming a part of the Centers for Disease Control
and Prevention (CDC), DHHS in 1987. NCHS aims ‘to provide statistical information that will guide
actions and policies to improve the health of the American people’ through its elaborate structure
of offices and divisions (table 9.1), with a budget of USD 160.4 million in the financial year 2016.
123
Table 9.1: Organizational structure and functions of the National Center for Health Statistics (NCHS)
Office / Division Functional areas
Core
Classifications and Public Health
Data Standards Staff
Data standards – classification systems and terminologies (e.g.,
ICD, ICF, SNOMED), message formats (e.g., HL-7, ANSI X12,
NCPDP), identifiers (provider, plan, individual), implementation
guides, core data sets (vital statistics, hospital discharge data),
120
https://ncvhs.hhs.gov/about/ (28/4/2020, 18:16 hours).
121
https://www.cdc.gov/nchs/nhis/about_nhis.htm (28/4/2020, 18:59 hours).
122
The US federal statistical system has 13 principal statistical agencies, with the Chief Statistician, Office of Information
and Regulatory Affairs (OIRA), Office of Management and Budget (OMB), as their coordinator. The NCHS is the principal
federal statistical agency, responsible for collection and dissemination of vital and health statistics. https://bit.ly/2y6NBp5
(28/4/2020, 20:22 hours).
123
https://www.cdc.gov/nchs/about/budget.htm (28/4/2020, 20:15 hours).
136
privacy and security; demographic standards and collection of
socioeconomic status data in DHSS surveys, comparisons and
integration of disparate data systems, data exchange between
clinical and population-based data systems
Division of Research and
Methodology
Collaborating Center for Statistical Research and Survey Design,
Collaborating Center for Questionnaire Design and Evaluation
Research, Research Data Center
Division of Analysis and
Epidemiology
Health promotion statistics, measures research and evaluation, data
linkage methodology and analysis, population health reporting
Division of Vital Statistics
Data acquisition classification and evaluation, mortality statistics,
reproductive statistics, IT
Division of Health Interview
Statistics
Data production and systems, survey planning and special surveys,
data analysis and quality assurance
Division of Health and Nutrition
Examination Surveys
Planning, operations, informatics, analysis
Division of Health Care Statistics
Ambulatory and hospital care statistics, long-term care statistics,
technical services
Managerial
Office of Planning, Budget and
Legislation
Planning, budget and legislation
Office of Management and
Operations
Operations and services, logistics, workforce and career
development
Office of Information Services Information design, dissemination, publishing
Office of Information Technology IT solutions and services
Source: https://www.cdc.gov/nchs/about/organization.htm (28/4/2020, 19:52 hours).
It is interesting to note that NCHS is not only responsible for DHSS’ population health and provider
surveys, but also vital statistics. Coordinating with state agencies, its National Vital Statistics System
(NVSS) offers monthly / quarterly / yearly provisional statistics of births and deaths – and, as part
of its public health surveillance activities, daily / weekly COVID-19 death data by demographic and
geographic characteristics. The NVSS mortality data, stored in its centrally computerized National
Death Index (NDI), helps in ascertaining death, causes of death, drug overdose, characteristics of
the deceased, life expectancy, maternal mortality, etc. Several NCHS surveys are linked with NDI
to help study factors associated with mortality in detail. A dedicated National Mortality Followback
137
Survey (NMFS) was initiated in 1961 to gather additional information on the deceased’s life history
from next of kin / a related person. The sixth, and the last, NMFS (1993) collected data on disability,
socioeconomic differentials in mortality, associations between risk factors and the cause of death,
access and utilization of health care facilities in the last year of life and data to assess the reliability
of information in the death certificate.
124
Let us now briefly discuss some of the key data collection systems of the NCHS.
National Vital Statistics System (NVSS)
One of the oldest data collecting organizations, the NVSS is also the most successful example of
inter-governmental health data sharing in the US. The recording of vital events dates back to 1632
when a registration law was passed in Virginia. With a few changes, the law was also enacted by
Massachusetts in 1639, with a modern vital registration system launched there in 1842 (Gutman
1958). During the late 19th century, decennial censuses included questions on vital events, which,
however, proved to be deficient in providing mortality statistics. Accordingly, when the US Bureau
of Census was established as a permanent agency in 1902, it started collecting data on vital events
from statistical offices in states and cities with a proper registration system annually. By 1933, all
states and cities registered vital events with acceptable coverage and shared information with the
Bureau to generate national estimates (NCHS 2013).
The legal authority to register vital events (marriages, divorces, births, induced abortions, deaths,
fetal deaths) in the US lies with 57 registration areas – 50 states, 5 territories of Puerto Rico, Guam,
American Samoa, US Virgin Islands and Commonwealth of the Northern Mariana Islands, and the
cities of Washington, DC and New York – involving nearly 6,000 local registers around the country.
While some states have centralized vital record offices, most of them have local offices. In 1933,
the National Association for Public Health Statistics and Information System (NAPHSIS) was formed
to represent state vital records and public health statistical offices in the US, and serve as a forum
for discussion and research to solve problems related to the collection and documentation of vital
statistics. NAPHSIS plays a key role in ensuring quality, confidentiality and usage of vital statistics
(Schwartz 2009). It, inter alia, co-organizes vital record courses for newcomers with the NVSS.
125
Periodic revision of NVSS’ ‘U.S. Standard Certificates and Reports’ takes place every 10-15 years
in collaboration with state vital statistical offices, NAPHSIS and experts. ‘Making these changes is
in keeping with a long history of rigorous evaluation of the quality and usefulness of data generated
by the vital statistics system and efforts to improve these data’ (DHSS Secretary).
126
124
https://www.cdc.gov/nchs/nvss/nmfs.htm (29/4/2020, 14:27 hours).
125
https://www.naphsis.org/vital-records-and-their-administrat (29/4/2020, 14:06 hours).
126
https://bit.ly/2W8n56L. For details on the evaluation process and recommendations for the latest 2003 revision, kindly
refer to https://bit.ly/3bMqHSp (29/4/2020, 14:13 hours).
138
Table 9.2: An overview of major health surveys in the US
Survey /
coordinating /
conducting
agencies
Features Inception Frequency Methods Sample Representative Major themes
National
Health
Interview
Survey
(NHIS)
/ NCHS
/ US Census
Bureau
Primary source of
data on health of
the US population
– widely used by
DHHS to track the
nation’s health and
progress towards
national health
objectives, health
policies, programs
1957 Continuous
Personal
interviews
~ 87,500
persons in
35,000
households a
year
National, states
(by combining
data years)
Health status
(diseases and
conditions) - health
care access and
utilization (incl. doctor
visits) - functioning,
disability, chronic
impairments - health
behaviors - health
insurance
National
Health and
Nutrition
Examination
Survey
(NHANES)
/ NCHS and
others
127
Program of studies
conducted as a
series of surveys
on various groups
/ health topics to
assess health and
nutritional status of
adults and
children in the US
1959 Continuous
Personal
interviews –
health tests /
examinations by
highly trained
medical staff in
mobile centers
and during doctor
visits – the higher
the age, the more
extensive the
examination
~ 5,000
persons a
year –
oversampling
of persons
aged 60+
years, Black,
Hispanic and
Asian
National
Selected diseases and
conditions - sexual
behavior, reproductive
history - environmental
and metabolic risk
factors - nutrition -
dietary supplement
use - children's growth
- healthy behaviors -
hearing and balance -
cognitive functioning -
prescription drug use
127
US federal agencies that collaborate with NCHS for NHANES are – within the DHSS (Food and Drug Administration, National Institutes of Health, Health Resources
and Services Administration, National Institute of Occupational Safety and Health, Agency for Toxic Substances and Disease Registry), Department of Agriculture,
Environmental Protection Agency, Social Security Administration. https://bit.ly/3aObUp8 (29/4/2020, 18:52 hours).
139
Survey /
coordinating /
conducting
agencies
Features Inception Frequency Methods Sample Representative Major themes
National
Survey of
Family
Growth
(NSFG)
/ NCHS
/ several
DHSS
agencies
128
Major source of
information on
reproductive-age
US women since
1973 and men
since 2002
1973
Continuous
(2006-)
Personal
interviews by
female
interviewers -
sensitive
questions
answered
privately (self-
administration)
~ 5,000 men
and women
aged 15-49
years a year -
Blacks,
Hispanics
and teens
oversampled
National
Family planning -
teenage sexual activity
and pregnancy -
infertility - adoption -
breastfeeding -
marriage, divorce,
cohabitation -
father's involvement -
HIV risk behavior
Behavioral
Risk Factor
Surveillance
System
(BRFSS)
/ several
CDC centers
and federal
agencies
/ All 50 states
The BRFSS is
considered as the
gold standard in
behavioral
surveillance –
collaborative
project between
all US states,
participating
territories and the
CDC – collects
data on health-
related risk
behaviors, chronic
health conditions
and use of
preventive
services
1984
Monthly
(states)
Annual
(CDC)
Mail - telephone
400,000+
adults (18+
years) a year
National, state,
local
Health status - healthy
days / health-related
quality of life - health
care access -exercise
- sleep - chronic health
conditions - oral health
- tobacco and alcohol
use – falls -
immunization - seat
belt use - drinking and
driving - breast,
cervical, prostate and
colorectal cancer
screening - HIV/AIDS
knowledge - emerging
health issues
128
https://www.cdc.gov/nchs/nsfg/about_nsfg.htm (29/4/2020, 19:04 hours).
140
Survey /
coordinating /
conducting
agencies
Features Inception Frequency Methods Sample Representative Major themes
National
Immunization
Surveys
(NIS)
/ National
Center for
Immunization
and
Respiratory
Diseases
(NCIRD),
CDC
/ NORC,
University of
Chicago
A group of phone
surveys used to
monitor
vaccination
coverage among
children and teens
1994 Annual
Telephone - mail
Parents /
caregivers as
respondent
~ 25,000
children (2
years) and
20,000 teens
(13-17 years)
(2018)
National, state,
other areas
Vaccinations as
recommended by the
CDC’s Advisory
Committee on
Immunization
Practices (ACIP)
State and
Local Area
Integrated
Telephone
Survey
(SLAITS)
/ NCHS and
sponsor
(public or
private)
Supplements
national data by
providing in-depth
state and local
area data to meet
various program
and policy needs -
- quick data for
evaluating
programs at
subnational levels,
etc. - data specific
to certain
populations
1997 Periodic Telephone
Variable
(uses NIS
sampling
frame)
National, state,
local
Variable
141
Survey /
coordinating /
conducting
agencies
Features Inception Frequency Methods Sample Representative Major themes
National
Survey of
Children's
Health
(NSCH)
/ Maternal
and Child
Health
Bureau,
DHHS
/ US Census
Bureau
Provides rich data
on multiple,
intersecting
aspects of
children’s (0-17
years) lives -
including physical
and mental health,
access to quality
health care, their
family,
neighborhood,
school and social
contexts
2003 Annual
Mail - web -
paper - telephone
Parents /
caregivers as
respondent
30,530
children
(2018) -
children with
special health
care needs
and 0-5 years
of age
oversampled
National, state
Physical and
emotional health -
factors related to well-
being, including
medical home, family
interactions, parental
health, school
experiences and safe
neighborhoods
Household
Pulse Survey
(HPS) / US
Census
Bureau
A 20-minute online
survey to evaluate
how the COVID-19
pandemic is
affecting
households across
the country from a
socioeconomic
perspective
2020
(23 April)
Weekly
Random selection
from the Census
Bureau’s Master
Address File
(MAF) - sample
households
contacted via
email and/or SMS
- longitudinal -
each household
interviewed thrice
- data collection
for 90 days -
release on a
weekly basis
1,048,950
households
(first 2 weeks
respondents)
State, 15 largest
Metropolitan
Statistical Areas
(MSAs)
Employment status -
spending patterns -
food security –
physical and mental
health - access to
health care - housing -
educational disruption
Source: NCHS and other sources. https://www.cdc.gov/nchs/index.htm (25/5/2020, 20:42 hours). Developed by author.
142
Table 9.2 above provides an overview of some of the key health surveys in the US. It is interesting
to note the mix of health survey strategies – for instance, while the 3 major NCHS surveys (NHIS,
NHANES and NSFG) are continuous and nationally representative, others play a complementary
role by providing annual / periodic and state / locally representative data as well. The BRFSS is a
collaborative effort between the federal and state governments inasmuch as it is conducted by all
50 US states even as the CDC coordinates with them to ensure standardization and puts together
national data. The 3 NCHS national surveys also provide a good mix of extensive (NHIS), focused
(NSFG) and intensive (NHANES) data. Let us discuss these 3 surveys here in some detail as their
combination appears to be quite relevant for the Indian context. Before we do that, let us make a
quick reference to the 20-minute online Household Pulse Survey (HPS), conducted by US Census
Bureau, to measure the social and economic impacts of the COVID-19 pandemic on households
– another example for the Government of India to consider. It is also worth highlighting here that
IPUMS
129
Health Surveys harmonizes data from 2 key health surveys – NHIS (1963-present) and
Medical Expenditure Panel Survey (MEPS)
130
(1996-present).
National Health Interview Survey (NHIS)
131
The NHIS is the principal source of information on the health of the civilian noninstitutionalized US
population, and is one of the major data collection programs of the NCHS. The National Health
Survey Act of 1956 provided for a continuing survey and special studies to secure accurate and
current statistical information on the amount, distribution, and effects of illness and disability, and
services rendered for or because of such conditions. The survey referred to in the Act, now called
the National Health Interview Survey, was initiated in July 1957. Since 1960, the survey has been
conducted by NCHS, which was formed when the National Health Survey and the National Vital
Statistics Division were combined.
NHIS data is used widely by DHHS to monitor trends in illness and disability and progress towards
achieving national health objectives. It is also used by the public health research community for
129
IPUMS provides, free of charge, census and survey data from around the world harmonizes and integrated across
time and space, making ‘it easy to study change, conduct comparative research, merge information across data types,
and analyze individuals within family and community contexts’. https://ipums.org/ (29/4/2020, 23:12 hours).
130
Starting in 1996, MEPS, sponsored by DHSS’ Agency for Healthcare Research and Quality (AHRQ), ‘is a set of large-
scale surveys of families and individuals, their medical providers, and employers’, and ‘is the most complete source of
data on the cost and use of health care and health insurance coverage’ in the US. https://www.meps.ahrq.gov/mepsweb/
(29/11/2020, 23:03 hours). Something for the Ayushman Bharat to consider.
131
While the authors have developed survey tables providing an overview, all descriptions of health surveys, henceforth,
have been reproduced from the original country sources (referenced in footnotes) with minor changes only. This has
been done to ensure the accuracy and originality of descriptions, paraphrasing which did not seem to be of any value,
especially in the context of budgetary and time constraints of the study. A critical review of international health surveys
from existing literature was also not undertaken due to these constraints. https://bit.ly/2Slj5yS (30/4/2020, 12:46 hours).
143
epidemiological / policy analysis of such timely issues as characterizing those with various health
problems, determining barriers to accessing and using appropriate health care and evaluating
federal health programs.
The NHIS also has a central role in the ongoing integration of household surveys in DHHS. The
designs of two major DHHS national household surveys have been or are linked to the NHIS. The
National Survey of Family Growth – which we will discuss shortly – used the NHIS sampling frame
in its first five cycles, while the Medical Expenditure Panel Survey currently uses half of the NHIS
sampling frame. Other linkages include linking NHIS data to death certificates in the NDI.
While the NHIS has been conducted continuously since 1957, the content of the survey has been
updated about every 10-15 years. In 1996, a substantially revised NHIS questionnaire began field
testing. This revised questionnaire, described in detail below, was implemented in 1997 and has
improved the ability of the NHIS to provide important health information.
Purpose and scope
The main objective of NHIS is to monitor the health of US population through the collection and
analysis of data on a broad range of health topics. A major strength of this survey lies in the ability
to display these health characteristics by many demographic and socioeconomic characteristics.
It covers the civilian noninstitutionalized population residing in the US at the time of the interview.
Because of technical and logistical problems, several segments of the population are not included
in the sample or in the estimates from the survey. Examples of persons excluded are patients in
long-term care facilities, persons on active duty with the armed forces (although their dependents
are included), persons incarcerated in the prison system and US nationals in foreign countries.
Sample design
NHIS is a cross-sectional household interview survey. Sampling and interviewing are continuous
throughout each year. The sampling plan follows an area probability design that permits the
representative sampling of households and noninstitutional group quarters (college dormitories,
for instance). The sampling plan is redesigned after every decennial census. The current sampling
plan was implemented in 2016. Clusters of addresses were defined within each state; the sizes of
the clusters correspond generally to the size of an interviewer’s workload over the course of the
sample design period. Each cluster is located entirely within a county, a small group of contiguous
counties, or a metropolitan statistical area. The current sampling plan is a sample of these clusters
of addresses.
The current NHIS sample design is not oversampling any race / ethnicity groups at the household
level. For the sample adult selection stage, persons aged 65 or older who are Black, Hispanic or
144
Asian had a higher chance to be selected than other adults in the family. This was an oversampling
feature of the previous sample design that continued in the current sample design until 2018.
As with previous two sample designs, the NHIS sample is drawn from each state and the District
of Columbia. While NHIS sample is too small to provide state-level data with acceptable precision
for each state, selected estimates for most states may be obtained by combining data years.
In the previous three sample designs, most sample addresses came from lists which were created
by field-listing operations. In the current sample design, field-listing is being done on a limited basis.
The main source of sample addresses is a commercial address list that is updated periodically.
There was a separate sampling mechanism for college dormitories in 2016-2017. This mechanism
was discontinued in 2018 due to low response rates. Changes were made in survey questionnaire
for 2018 to ask if there are any people who usually live at the sampled address, but are currently
living away at school in on-campus housing. If so, the household respondents are asked to include
them in the household roster for their ‘home’ address (for e.g., parents’ home).
The total NHIS sample is subdivided into 4 separate panels, or sub-designs, such that each panel
is a representative sample of the US population. This design feature has a number of advantages,
including flexibility for total sample size. With 4 sample panels and no sample cuts or augmentations,
expected NHIS sample size (completed interviews) is nearly 35,000 households, 87,500 persons.
The annual response rate of NHIS is approximately 70% of the eligible households in the sample.
Questionnaires
The NHIS questionnaire that was used from 1982-1996 consisted of 2 parts – 1) a set of basic
health and demographic items (the ‘core questionnaire’), and 2) one or more sets of questions on
current health topics. The core questionnaire remained the same over that time period, while the
current health topics changed depending on data needs. The core questionnaire, while collecting
data on health conditions and utilization, did not collect any information on insurance, access to
health care or health behaviors. In addition, much of the interview time in the core questionnaire
was devoted to collecting detailed information on events, such as doctor visits and hospitalizations
rather than on information that would better characterize the individual. The 1997 revision of the
NHIS questionnaire tried to address these and other shortcomings.
The revised NHIS questionnaire, implemented since 1997, has core questions and supplements.
The core questions remain largely unchanged from year to year and allow for trends analysis and
for data from more than one year to be pooled to increase sample size for analytical purposes.
The core contains 4 major components – household, family, sample adult and sample child.
The household component collects limited demographic information on all of the individuals living
in a particular house; family component verifies and collects additional demographic information
145
on each member and collects data on topics including health status and limitations, injuries, health
care access / utilization, health insurance, income and assets. The family core component allows
the NHIS to serve as a sampling frame for additional integrated surveys, as needed.
From each family in the NHIS, one sample adult and one sample child (if any children are present)
are randomly selected and information on each is collected with the sample adult core and sample
child core questionnaires. Because some health issues are different for children and adults, these
two questionnaires differ in some items, but both collect basic information on health status, health
care services and health behaviors.
The supplements are used to respond to new public health data needs. As with the previous NHIS
supplements, the questionnaires are sometimes fielded only once or repeated, as needed. These
questionnaires may be used to provide additional details on a subject already covered in the core
or on a different topic not covered in other parts of the NHIS. The first supplement from the current
questionnaire design was fielded in 1998, and focused on data needed to track the Healthy People
2000 and 2010 objectives. Other topics covered in the supplements include cancer screening,
complementary and alternative medicine, children’s mental health and health care utilization.
Questionnaire redesign
The content and structure of the NHIS are updated in the redesign to better meet the needs of
data users. Aims of the redesign are to improve the measurement of covered health topics, reduce
respondent burden by shortening the length of the questionnaire, harmonize overlapping content
with other federal health surveys, establish a long-term structure of ongoing and periodic topics,
and incorporate advances in survey methodology and measurement. Public comments received
are instrumental in determining the survey content for the redesigned NHIS.
132
Data collection procedures
Data are collected through a personal household interview, conducted by interviewers employed
and trained by the US Census Bureau, according to procedures specified by the NCHS.
For the family core component, all adult members of the household 17 years of age and over, who
are at home at the time of the interview, are invited to participate and to respond for themselves.
Beginning in 1997, data were collected for active duty military personnel, provided there is one
civilian in family. However, these persons were not weighted for analytical purposes. For children
or adults not at home during interviews, information can be provided by a responsible adult family
member, 18 years of age and over, residing in the household. For the sample adult questionnaire,
132
The 2020 as well as previous questionnaires, instrument flowchart and the field representative manual are available
at – https://www.cdc.gov/nchs/nhis/data-questionnaires-documentation.htm (30/4/2020, 13:41 hours).
146
one civilian adult per family is randomly selected. Generally, this adult must self-report responses
to questions in this section. Information for the sample child questionnaire is obtained from a
knowledgeable adult in the household, usually a parent.
The US Census Bureau, under a contractual agreement, is the data collection agent for the NHIS.
Nationally, NHIS uses about 600 interviewers, trained and directed by health survey supervisors
in each of the 6 Bureau regional offices. The supervisors are career civil service employees, while
the interviewers are part-time employees, selected through an examination and testing process.
Interviewers receive thorough training in basic interviewing procedures and in the concepts and
procedures unique to the NHIS.
The revised NHIS questionnaire, fielded since 1997, uses the CAPI mode – the interviewers enter
responses into the laptop during the interviews, ensuring timeliness of data and improved quality.
Early Release (ER) Program
The ER Program of NHIS provides estimates, analytical reports and preliminary microdata files on
an expedited schedule. NHIS data users have access to timely estimates, reports and microdata
files without having to wait for the final annual NHIS microdata files by selected characteristics.
The first 2019 NHIS ER estimates were available in early 2020.
133
National Health and Nutrition Examination Survey (NHANES)
The NHANES is a program of studies designed to assess the health and nutritional status of adults
and children in the US. It is unique in that it combines interviews and physical examinations.
The NHANES program began in the early 1960s and has been conducted as a series of surveys
focusing on different population groups or health topics. In 1999, the survey became a continuous
program with a changing focus on a variety of health and nutrition measurements to meet emerging
needs. The survey examines a nationally representative sample of about 5,000 persons each year,
located in counties across the US, 15 of which are visited a year. Interviews includes demographic,
socioeconomic, dietary and health-related questions. Examinations consist of medical, dental and
physiological measurements and laboratory tests conducted by highly trained medical personnel.
Findings from the survey are used to determine prevalence of major diseases and risk factors as
well as assess nutritional status and its association with health promotion and disease prevention.
Its findings are also the basis for national standards for such measurements as height, weight and
blood pressure; its data is used in epidemiological studies and health sciences research that help
develop sound public health policy, direct and design health programs and services, etc.
133
https://bit.ly/3d7Pm4G and https://bit.ly/3caCSaZ (30/4/2020, 13:45 hours).
147
Purpose and scope
The purpose of NHANES is to –
➢ Estimate the number / percentage of persons in the US population and designated sub-groups
with selected diseases and risk factors
➢ Monitor trends in the prevalence, awareness, treatment and control of selected diseases
➢ Monitor trends in risk behaviors and environmental exposures
➢ Analyze risk factors for selected diseases
➢ Study the relationship between diet, nutrition and health
➢ Explore emerging public health issues and new technologies
➢ Establish a national probability sample of genetic material for genetic testing for liver health
,
As in past health examination surveys, data is collected on prevalence of chronic conditions in the
population. Estimates for previously undiagnosed conditions, and those known to and reported by
respondents, are produced in the survey. Such information is a strength of the NHANES program.
Risk factors are examined – smoking, alcohol consumption, sexual practices, drug use, physical
fitness and activity, weight and dietary intake are studied. Data on certain aspects of reproductive
health such as use of oral contraceptives and breastfeeding practices are also collected. Diseases
/ medical conditions / health indicators studied include – cardiovascular diseases, diabetes, kidney
and respiratory diseases, osteoporosis, oral health, eye diseases, hearing loss, infectious diseases,
sexually transmitted diseases (STDs), reproductive history and sexual behavior, obesity, nutrition,
anemia, physical fitness / functioning and environmental exposures.
Sample design
NHANES uses a complex, multistage probability design to sample the civilian noninstitutionalized
population residing in the 50 states and D.C. Sample selection followed these stages, in order –
1) Selection of PSUs, which are counties or small groups of contiguous counties
2) Selection of segments within PSUs that constitute a block / group of blocks containing a cluster
of households
3) Selection of specific households within segments
4) Selection of individuals within a household
In 2015-2016, 15,327 persons were selected for NHANES from 30 different survey locations. Of
those selected, 9,971 completed the interview and 9,544 were examined.
148
To facilitate oversampling of the Asian population, survey materials were translated into Mandarin
Chinese (traditional / simplified), Korean and Vietnamese. Recorded and written translations were
also posted on the NHANES participants’ webpage, which included a short video that explained
what was involved for the participant when participating in the survey. This video was designed to
promote interaction and show participants some of the benefits of participating in the survey. The
video was also available in Amharic, French, Haitian Creole, Hindi and Spanish languages.
The staff participated in cultural competency training to help them recognize and respect cultural
differences. Local interpreters were hired when necessary and provided with translated glossaries
of terms, hand cards and exam scripts to minimize interpretation errors. In addition, a professional
medical interpreter phone service was available to assist with any needs not otherwise provided.
Beginning in 1999, NHANES oversampled low-income persons, adolescents aged 12-19, persons
aged 60 and over, Black or African American persons and persons of Mexican origin The sample
for data years 1999-2006 was not designed to give a nationally representative sample for the total
Hispanic population residing in the US. Starting with 2007-10 data collection, all Hispanic persons
were oversampled, not just persons of Mexican origin; adolescents were no longer oversampled
In 2011-14, the sampling design was changed and following groups were oversampled – Hispanic
and non-Hispanic Black and Asian persons; non-Hispanic white and other persons at / below 130%
poverty; and non-Hispanic White and other persons aged 80 and above. In 2015-16, the sampling
design was revised again, changing the cut-point for low-income oversampling from at / below
130% poverty to at / below 185% poverty.
For NHANES 1999-2000 to NHANES 2011-12, number of persons selected ranged from 12,160
to 13,431. The percentage who were interviewed ranged from 73% to 84%, while the percentage
who were examined ranged from 70% to 80%. For NHANES 2013-14, a total of 14,332 persons
were eligible, of which 71% were interviewed and 68% completed health examination component.
For NHANES 2015-16, a total of 15,327 persons were eligible, of which 61% were interviewed
and 59% completed the health examination component.
Survey process
In each location, local health and government officials are notified of upcoming NHANES survey.
Households in the study area receive a letter from the NCHS Director to introduce the survey.
Local media may feature stories about the survey.
Selected persons are invited to take part in the survey by being interviewed in their households.
Household interview data is collected via CAPI and includes demographic, socioeconomic, dietary
and health-related questions. Upon interview completion, sample persons are asked to participate
in a physical examination conducted in a specially equipped / designed Mobile Examination Center
(MEC). The MEC is composed of 4 trailers, and houses all the state-of-the-art equipment for the
149
physical exams and tests conducted. The trailers are divided into rooms to assure the privacy of
each study participant during the examination and interview. This examination includes a physical
examination conducted by a physician, measurements like height, weight, laboratory tests, bone
density scans and other health measurements (including laboratory analysis of blood, urine and
other tissue samples) and interviews conducted by highly trained health professionals. In general,
the older the individual, the more extensive the examination. Medical examinations and lab tests
follow very specific protocols and are standardized as much as possible to ensure comparability
across sites and providers. Study teams are largely bilingual (English / Spanish), and consist of a
physician, medical and health technicians as well as dietary and health interviewers. NHANES is
designed to facilitate and encourage participation. Transportation is provided to / from the mobile
center if necessary. Participants receive compensation and a report of their medical findings.
An advanced computer system using high-end servers, desktop PCs and wide-area networking
collect and process all NHANES data, nearly eliminating need for paper forms and manual coding
operations. This system allows interviewers to use notebook computers with electronic pens. The
staff at the mobile center can automatically transmit data into databases through such devices as
digital scales and stadiometers. Touch-sensitive computer screens let respondents enter their
own responses to certain sensitive questions in complete privacy. Survey information is available
to NCHS staff within 24 hours of collection, enhancing the capability of collecting quality data and
increasing the speed with which results are released to the public.
Questionnaires
NHANES has 2 screener modules, a family and a sample person questionnaire. Screener module
1 is administered on the doorstep to determine if anyone in the household is eligible to be in the
sample. Screener Module 2 establishes the relationship of everyone in the household to everyone
else in the household. The family questionnaire has the following sections – consumer behavior,
demographic background, family questionnaire handcards, food security, housing characteristics,
income, salt sample selection and smoking. The sample person questionnaire has these sections
– acculturation, audiometry, blood pressure, cardiovascular disease, demographic, dermatology,
diabetes, diet behavior / nutrition, dietary supplements / prescription medication, early childhood,
functioning, health insurance, hepatitis, hospital utilization and access to care, immunization, infant
formula questionnaire, kidney condition, medical conditions, occupation, oral health, osteoporosis,
physical activity and physical fitness, respondent selection section, sample person questionnaire
handcards, sleep disorders, smoking and tobacco use, standing balance and weight history.
150
Data uses
NHANES information is made available through an extensive series of publications and articles in
scientific and technical journals. For data users and researchers throughout the world, survey data
are available on the internet and on easy-to-use CDs. Research organizations, universities, health
care providers and educators benefit from its information. Primary data users are federal agencies
that collaborated in the design and development of the survey. The NIH, FDA and CDC are among
the agencies that rely upon NHANES to provide data essential for implementation and evaluation
of program activities. The US Department of Agriculture and NCHS cooperate in planning and
reporting dietary and nutrition information from the survey. Partnership with the US Environmental
Protection Agency allows continued study of many important environmental influences on health.
Past surveys have provided data to create the growth charts used nationally by pediatricians to
evaluate children’s growth. The charts have been adapted and adopted worldwide as a reference
standard and are updated using the latest NHANES data.
Because NHANES is an ongoing program, information collected contributes to annual estimates
in topic areas included in the survey. For small population groups and less prevalent conditions /
diseases, data must be accumulated over several years to provide adequate estimates.
Data release and access policy
This policy addresses when, to whom and in what form the Division of Health Examination Surveys
(DHANES) should disseminate NHANES data and outlines dissemination procedures. The policy
is consistent with CDC and NCHS policies, including the guiding principles of making high quality
data available –
➢ As widely as practicable
➢ As soon as possible after data collection
➢ In as much detail as possible
➢ While maintaining survey participant confidentiality
Various mechanisms of data release and access are used to follow the principles, including public
data release as well as limited data access arrangements.
Since NHANES 1999-2000, public use data releases have been and continue to be made on a bi-
annual basis. Due to voluminous nature of NHANES and the large amount of post data-collection
processing, release of all data from 2 years of data collection does not occur at a point in time. An
initial data release occurs approximately 9 months after completion of each 2-year data collection
cycle and intermittent releases follow as remaining data is processed, until all releasable data are
available for public use.
151
Whenever new data items are developed using physical samples from NHANES surveys, such as
stored sera, DNA or imaging studies, these items are made publicly accessible under either public
use or RDC access, depending on nature of the derived data item and disclosure risk. If requested
data are not currently collected or available in NHANES, a proposal to obtain (and fund) the new
data items can be submitted via email to NHANES Biospecimen Program. The NHANES Project
Officer and a technical panel evaluate all proposals for scientific merit. The NCHS Human Subject
Contact and Ethics Review Board (ERB) then review the proposal for any potential human subjects
concerns and the NCHS Confidentiality Officer for disclosure risk. Any data developed under this
mechanism are made accessible under either public use, or RDC access to appropriate recipients
as noted above.
134
National Survey of Family Growth (NSFG)
The NSFG gathers nationally representative data – not for individual states – on family life, marriage
and divorce, pregnancy, infertility, use of contraception, general and reproductive health. The first
NSFG surveys were conducted as periodic cycles by NCHS in 1973, 1976, 1982, 1988 and 1995,
based on personal interviews conducted in the homes of a national sample of women 15-44 years
of age in the civilian, noninstitutionalized population of the US. Later changes to the NSFG include
addition of an independent sample of men in 2002, shift from being a periodic survey to continuous
interviewing in 2006, and shift in age range to 15-49 years in 2015. Under continuous interviewing,
fieldwork takes place continually (48 weeks / four 12-week quarters each year) in a smaller number
of areas in the US than is the case for periodic interviewing. In each interviewing year, over 5,000
interviews have been completed. Significant oversampling of Blacks, Hispanics, teens aged 15-19
years and a slightly higher sampling rate for females is observed. The samples for different time
intervals could be combined. Like periodic interviewing, all interviews are conducted in person by
female interviewers (CAPI), with some of the more sensitive questions answered privately using
audio-computer assisted self-interviewing (Audio-CASI / ACASI) through self-administration. In this
procedure, respondents answer questions on the laptop either by reading them or listening to pre-
recorded questions read over headphones and enter their answers directly into the computer.
The NSFG responds to the Congressional mandate for NCHS to collect and publish reliable national
statistics on “family formation, growth, and dissolution” (Sec. 306 (a and b), paragraph 1(H) of the
Public Health Service Act) as well as vital statistics on births and deaths, and a number of aspects
of health status and health care. The NSFG collects and publishes the most reliable – in most cases
the only – national data on such major topics as adoption, unplanned births, contraceptive use and
effectiveness, infertility and use of infertility services, pelvic infection and sexually transmitted
134
https://bit.ly/35kHVEk; https://bit.ly/2zL4K8j; https://bit.ly/3c2UBSs; https://bit.ly/2ZFbgbv; https://bit.ly/3aNq4qF and
https://bit.ly/2TI1rWy. For details on NHANES Biospecimen Program – https://bit.ly/3d2SxtS (30/4/2020, 22:03 hours).
152
disease, sterilization, expected future births, the sexually active population, and the use of and need
for family planning services. The NSFG aims at providing time series data on these variables through
continuous interviewing, while improving sample sizes at an affordable cost.
Purpose and scope
The primary purpose of the survey, particularly since the inclusion of a sample of men, has been
to produce reliable national estimates of –
➢ factors affecting pregnancy, including sexual activity, contraceptive use and infertility
➢ the medical care associated with contraception, infertility and childbirth
➢ factors affecting marriage, divorce, cohabitation and family building
➢ adoption and caring for non-biological children
➢ father involvement with their children
➢ use of sexual and reproductive health services
➢ attitudes about sex, childbearing and marriage
No clinical or examination data are collected from survey participants. Only one survey participant
is selected from each sample household. Socioeconomic variables include income and poverty,
education, employment status (full-time or part-time), source of payment for delivery and selected
health services, and receipt of public assistance.
Sample design
The NSFG sample is designed to meet a number of key objectives, including –
➢ minimizing the overall design effects for women and men
➢ controlling the costs of both screening and interviewing
➢ obtaining overall sample size of at least 5,000 interviews per year
➢ providing for oversamples of Blacks, Hispanics and teens aged 15-19 years
The NSFG survey population consists of all noninstitutionalized women and men aged 15-49 years
as of first contact for the survey, whose usual place of residence is the 50 United States or District
of Columbia. The NSFG is based on a stratified multi-stage area probability sample using probability
proportionate to size (PPS) selection with 5 stages of sample selection –
➢ selection of PSUs
153
➢ selection of secondary sampling units (SSUs)
➢ listing and selection of housing units within SSUs
➢ selecting one of the eligible persons within each sampled household
➢ two-phase sampling for non-response
135
Each year, about 15,000 households are contacted in order to get approximately 5,000 interviews.
Each year of data is an independent national sample, but the desired sample size and precision
for several key estimates and statistics are attained after about 4 years of interviewing. In addition,
despite each year of fieldwork being designed to yield nationally representative data, sample
weights are only constructed for 2 years of data, which is the minimum timespan for NSFG public
use file releases that permit statistically reliable estimates to be made.
Survey process
Fieldwork for 2015-2017 NSFG was conducted from September 2015 through September 2017,
based on a survey protocol and informed consent procedures approved by the NCHS Research
Ethics Review Board. After a sample respondent per household was selected based on screening
interviews in NSFG sample households, in-person interviews were conducted with 5,554 women
and 4,540 men 15-49 years of age for a total sample size of 10,094. Signed parental permission
and minor assent were obtained for all minor respondents aged 15-17 years. Adult respondents
could provide consent without signature. For 2015-2017, the interviews for female respondents
averaged 73.0 minutes in length, and the interviews for male respondents averaged 49.9 minutes,
both within the limits of 80 minutes for females and 60 minutes for males approved by the Office
of Management and Budget. Respondents in Phase 1 of data collection were offered a $40 token
of appreciation in cash. Those adult respondents screened in Phase 1 and selected into Phase 2
for a main interview were offered an additional $40 (total of $80) as a prepaid token of appreciation
for completion of the survey. Households selected for Phase 2 that were not yet screened in Phase
1 were also sent a $5 prepaid token of appreciation for completion of the screener.
Respondent burden for the NSFG is kept to a minimum through the use of sampling procedures
that permit generation of statistically valid national estimates for roughly 149 million people 15-49
years of age with about 20,000 interviews over 4 years of interviewing; keeping the length of the
questionnaires under the approved 80 minutes for women and 60 minutes for men, and by using
faster and more efficient laptops and the latest edition of BLAISE CAPI software. CAPI reduces
burden for the respondent because it collects data using a laptop computer, along with a highly
skilled interviewer. The computer customizes the questionnaire and question wording for the
135
Further details on sample selection are available at – https://bit.ly/3e8f2OI (22/5/2020, 18:40 hours).
154
respondent, based on answers given during the administration of the instrument. A portion of the
NSFG interview (around 15-20 minutes) is conducted using ACASI. However, only material that
is sensitive and fairly simple to ask and answer is collected in ACASI. Respondents often report
that they enjoy the ACASI part of the interview because they can control the pace of the interview
themselves and be more active participants in it. Despite its appeal, however, it is not practicable
to use ACASI to conduct the entire survey as much of the questionnaire material is too complex
to be self-administered.
Questionnaires
The questions in the NSFG questionnaires may be divided into 2 categories – (1) questions that
have generally been asked in some form in the NSFG since the 1970s – including demographic
characteristics like education and marital status and behaviors like contraceptive use, marriage,
divorce and unmarried cohabitation; and (2) more sensitive questions that are asked in ACASI
and have mostly been asked since 2002 (Incarceration, drug use, on-voluntary sexual experience,
behavior, identity and attraction, same-sex sexual activity, sexually transmitted diseases, income).
There are 4 NSFG questionnaires – the household screener questionnaire, female questionnaire,
male questionnaire and verification questionnaires – as well as an interviewer observation form.
136
Data uses
NSFG provides data for major areas of Healthy People 2020, and is the primary source of data for
family planning objectives. In addition, NSFG is an important contributor of data for objectives in
the areas of HIV, STDs, maternal, infant and child health. NSFG data has been used to brief the
DHHS Secretary, Surgeon General as well as others. One of the NSFG-based objectives (receipt
of reproductive health services in the past 12 months) was selected as one of 26 leading health
indicators for the nation. NSFG data are used by many DHHS agencies; for instance –
➢ The Office of Population Affairs uses NSFG data to estimate the characteristics of women who
use Title X-funded clinics for family planning and related health services and for research on
factors affecting contraceptive use, unintended pregnancy, teenage sexual activity and use of
medical services for family planning and reproductive health. The data on men’s reproductive
behavior are also used to improve family planning and related health services targeting men.
➢ Population Dynamics Branch, NICHD, NIH, uses the data from men and women as a resource
for intramural and extramural research on marriage, cohabitation, fertility and infertility,
contraceptive use, sexually transmitted infections and breastfeeding in the United States.
136
NSFG questionnaires are available at – https://bit.ly/2LZtYTe (22/5/2020, 17:41 hours).
155
➢ The Children’s Bureau has a special research interest in data collected on children in foster
care, fertility and family formation behaviors of adults who experienced foster care as children.
➢ The Administration for Children and Families in the Office of Planning, Research and Evaluation
relies on NSFG data on fatherhood, marriage and teen pregnancy risk behaviors for planning
programs to improve the economic and social well-being of children and families.
➢ Division of HIV/AIDS Prevention, CDC, undertakes research based on NSFG data on behaviors
that affect the risk of transmission of HIV – including condom use, numbers of sexual partners,
etc.
➢ Division of Cancer Prevention and Control, CDC, uses NSFG data on screening for cervical
cancer, human papillomavirus (HPV) and breast cancer, which can be analyzed in relation to
the NSFG’s extensive data on pregnancy histories, sexual behavior and reproductive health.
It has also supported recent questionnaire additions to evaluate adherence to revised cancer
screening guidelines.
➢ Division of Reproductive Health, CDC, uses NSFG data for surveillance of reproductive health
outcomes and research on teen pregnancy prevention, sexual activity and contraceptive use.
DRH also uses NSFG data for their work on establishing recommendations for family planning
services, including contraceptive services.
➢ Within CDC’s National Center for Chronic Disease Prevention and Health Promotion, the NSFG
has long been supported by the Divisions of Cancer Prevention and Control and Reproductive
Health. Since 2016, Division of Nutrition Monitoring, Physical Activity and Obesity has begun
co-sponsoring NSFG to support overall data collection on fertility and infant feeding practices,
including breastfeeding, as well as nutrition-related counseling that mothers of young children
receive from health care providers and other sources.
➢ The Division of Birth Defects and Developmental Disabilities, CDC, uses estimates of number
and characteristics of women at risk of an alcohol-exposed pregnancy that could lead to Fetal
Alcohol Syndrome.
137
Lessons for India
The GoI / state / UT governments should consider the following lessons from the US context –
® Develop a national / state health information policy / Act.
® Establish –
137
https://bit.ly/3bS69aA, https://bit.ly/2XitqgE, https://bit.ly/3gdyw6w; https://bit.ly/36xCGBH; https://bit.ly/3c8dF0S and
https://bit.ly/36khpeH (22/5/2020, 17:41 hours).
156
o National Vital Statistics System (NVSS)-type agency with Centre-state coordination;
o National / state committees on vital and health statistics (like the NCVHS) that function
as statutory, advisory and monitoring bodies for vital and health statistics;
o National / state centers for vital and health statistics (like the NCHS) that function as
principal health statistical agencies (much like the DES at a broader level).
® We should consider a mixed health survey strategy – for e.g., 3 major surveys (NHIS, NHANES
and NSFG) are continuous and nationally representative, while others play a complementary
role by providing annual / periodic and state / locally representative data. These 3 surveys also
offer a good mix of extensive (NHIS), focused (NSFG) and intensive (NHANES) data. In India,
the SRS is already continuous – we can easily learn from its experience, while acknowledging
/ addressing its limitations / challenges.
® Continuous surveys could have a smaller sample size and be representative at national and
state / UT level. For smaller population groups / districts, there should be the option of pooling
data over a few years to get representative estimates.
® At least, some of the health surveys should be linked up with vital statistics for follow-up data
collection. This is particularly important for death and cause of death statistics as well as RCH.
India should consider mortality follow-up surveys. The SRS-CoD survey is already doing that.
However, we need to have a much bigger sample size for these surveys – and for that, linkage
with / follow-up on the CRVS mortality data is needed, as in the US.
® All vital and health statistics should be interoperable. An IPUMS-type agency should be set up
for this purpose, under the administrative control and supervision of the NSO.
® A survey on the health and socioeconomic impact of COVID-19 on households (like Household
Pulse Survey) should be developed that could provide data by background characteristics like
the NFHS to assess the differential impact of the pandemic on different population groups.
Bibliography
Gutman, Robert. 1958. ‘Birth and death registration in Massachusetts: II. The inauguration of a
modern system, 1800-1849’. The Milbank Memorial Fund Quarterly 36(4): 373-402.
Haywood, Alice. 1981. ‘The National Health Survey–in the beginning’. Public Health Reports 96(3):
195-199.
Schwartz, Steven. 2009. ‘The U.S. Vital Statistics System: The role of state and local health
departments’. https://bit.ly/2VHxlUo (28/4/2020, 15:46 hours).
Weisz, George. 2011. ‘Epidemiology and health care reform: The National Health Survey of 1935-
1936’. American Journal of Public Health 101(3): 438-47.
157
10. Canada
Statistics has been a federal responsibility in Canada since its creation in 1867. The Statistics Act,
1918 created the Dominion Bureau of Statistics, which came to be known since 1971 as Statistics
Canada (StatCan), the country’s central statistical office – functioning under the Canadian Ministry
of Innovation, Science and Economic Development. Both the Ministry and StatCan are headed by
persons of Indian origin – Navdeep Bains and Anil Arora (Chief Statistician of Canada). With over
6,000 employees across the country, StatCan conducts more than 350 surveys around almost all
aspects of public life (31 themes) as well as population and agriculture censuses quinquennially.
According to the Statistics Act (current to 17 May 2020), StatCan’s duties are to –
<
1) Collect, compile, analyse, abstract and publish statistical information relating to the commercial,
industrial, financial, social, economic and general activities and condition of the people;
2) Collaborate with departments of government in the collection, compilation and publication of
statistical information, including statistics derived from the activities of those departments;
3) Take the census of population of Canada and the census of agriculture of Canada as provided
in this Act;
4) Promote avoidance of duplication in the information collected by departments of government;
and
5) Generally, to promote and develop integrated social and economic statistics pertaining to the
whole of Canada and to each of the provinces thereof and to coordinate plans for the
integration of those statistics.
In functional terms, StatCan has 2 main objectives –
1) To provide statistical information and analysis to help:
a. improve public and private decision-making;
b. develop and evaluate public policies and programs;
2) To promote sound statistical standards and practices by:
a. using common concepts and classifications to provide better quality data;
b. working with provinces and territories for higher efficiency in data collection, including
reducing duplication as well as respondent burden through greater use of data sharing
agreements (for e.g. employee payroll, tax and customs records);
c. improving statistical methods and systems through joint research studies and projects.
158
The agency collaborates with various departments and levels of the governments as well as public
and private agencies, academics and others to develop surveys based on Canada’s priorities and
requirements. StatCan maintains close relationships with key federal departments / agencies to
foster awareness of each other’s needs / priorities. The Federal-Provincial-Territorial Consultative
Council on Statistical Policy and its subcommittees comprise a network of 13 provincial / territorial
official representatives, who collaborate with StatCan to determine data requirements, consult on
current statistical activities and coordinate dissemination of StatCan’s products to provincial and
territorial governments. There are 3 special initiatives, in the areas of health, education and justice
statistics. StatCan’s priorities in health are developed with the assistance of the Board of Directors
of Canadian Institute for Health Information (CIHI). The Board comprises senior federal, provincial
and private sector representatives, including the Chief Statistician.
138
A set of advisory groups offer impartial and independent advice to StatCan vis-à-vis overall quality
of the national statistical system (Canadian Statistics Advisory Council), continuous review of the
Agency’s statistical outputs and helping set priorities and foster program relevance (a network of
professional advisory committees in major subject areas), data access, privacy, data governance
to maintain and support data needs (Advisory Council on Ethics and Modernization of Microdata
Access), developing common approaches for collection of vital statistics, sharing information and
facilitating problem-solving through sharing experiences, research findings and expertise among
various jurisdictions (Vital Statistics Council for Canada). All this is seen as critical to the fulfilment
of StatCan’s mission – ‘serving Canada with high-quality statistical information that matters’.
139
StatCan’s Health Statistics Program (HSP)
In Canada, StatCan and CIHI jointly produce national health data and indicators related to health.
The distinction in their roles is – CIHI is mandated to lead the development and maintenance of
comprehensive and integrated health information which enables sound policy and effective health
system management, while StatCan (HSP) provides information about the health of the Canadian
population, the determinants of health and the use of Canada’s health care resources. HSP could
be compared to the NCHS in the US – only that the former functions under a much broad-based
agency (StatCan) than the latter does (CDC).
HSP’s aim is to provide reliable, timely and relevant information about the health of Canadians. Its
Health Statistics Division (HSD), in collaboration with the Health Analysis Division (HAD), provides
statistical analysis and information about the health of the population, the determinants of health,
the scope / use of Canada’s health care resources. This information is used to assist and support
health planners and decision-makers at all levels of the government, to sustain demographic and
138
https://www.statcan.gc.ca/eng/about/relevant (3/6/2020, 15:42 hours).
139
https://www.statcan.gc.ca/eng/about/about?MM=as (3/6/2020, 15:48 hours).
159
epidemiological research and to report to the Canadian public on their collective health and health
care system performance.
HSD works in partnership with provincial and territorial vital statistics registrars, cancer registries
as well as data providers and users at the federal level – Health Canada and Public Health Agency
of Canada (PHAC) – with the provincial ministries of health and health regions. It also works with
many other partners, including CIHI, to provide statistical information and undertake research and
analysis on the health status of Canadians and the health care system.
HAD’s mandate is to provide high-quality, relevant and comprehensive information on the health
status of the population and on the health care system to a broad audience – health professionals,
researchers, policymakers, educators and students. A significant part of HAD’s research program
is undertaken in collaboration with various partners – federal ministries, health organizations and
universities. HAD brings out Health Reports, a peer-reviewed and indexed journal of population
health and health services research, publishing original and timely analyses of surveys as well as
national / provincial administrative databases. HAD analysts also regularly publish in other journals.
HAD’s program of work involves data development, development of methods and modelling, and
research in 5 broad-based health themes – maternal, child and youth health; ageing; vulnerable
populations; health and the environment; behaviors, chronic disease and cancer – using a number
of analytical lenses, including determinants, outcomes, mortality / morbidity, heath care utilization,
international comparisons and longitudinal as well as person-oriented focuses. Research is based
on StatCan’s comprehensive suite of data on the health of Canadians and the functioning of the
health system, including census, vital statistics, administrative health data, population, post-censal
and special surveys. Its record linkage activities
140
deserve special mention. Various projects have
combined census, vital statistics, cancer, population health surveys as well as other data to avoid
duplication / burden in data collection and to help better understand the health of Canadians and
the functioning of the health system. Survey respondents are notified of planned linkages before
/ during the survey. If any respondent objects to the linking of their data, their objection is recorded
and no data linkage happens. Income information obtained from income tax records, for instance,
is provided to federal, provincial and territorial agencies only after the respondents’ consent.
141
Microdata linkages
Linking of separate records from different sources can be a very useful and cost-efficient technique
in the design, production, analysis and evaluation of statistical data, leading to important savings
in cost, time and respondent burden, and, in some cases, it may be the only feasible way to obtain
140
‘Record linkage is the process of bringing together 2 or more records relating to the same entity (e.g., person, family,
event, community, business, hospital or geographical area)’. https://bit.ly/304k1fm (3/6/2020, 16:12 hours).
141
https://bit.ly/36UHAJ4 and https://bit.ly/2TNML8n (3/6/2020, 16:15 hours).
160
important statistical information. When possible, instead of conducting additional surveys, StatCan
uses information that individuals, businesses and institutions have already provided to the Agency
or other government departments for methodological purposes, data enhancement and subject-
matter studies. When health survey and administrative data are combined through record linkage,
relationships between social determinants and health outcomes could be analysed in more depth.
Microdata linkages are conducted in accordance with StatCan’s Directive on Microdata Linkage,
in place since 1986. StatCan performs microdata linkages – a) to support the design, maintenance,
evaluation, research and redesign of ongoing data collection / methodological studies in StatCan;
b) provide statistical information in aggregate or anonymous format in support of research studies.
StatCan has pre-approved specific types of microdata linkages where privacy risks and situations
of potential conflict of interest are low and where procedures to mitigate risk to confidentiality and
privacy are in place. All other microdata linkages must undergo a prescribed review and approval
process, which involves submission of documented proposals to senior management. When such
linkages include personal information, a summary of the approved microdata linkage is posted on
StatCan’s website.
142
Before we move on to provide an overview of major active health surveys in Canada (table 10.1)
and discuss some of them, let us highlight the issue of declining responses rates in health surveys
in the country. Figure 10.1 illustrates this vis-à-vis a leading Canadian health survey. It is important
to note that telephonic interviewing (CATI) has become less favourable than personal interviewing
(CAPI), with the differentials in response rates between the two modes increasing over the years.
Figure 10.1: Response rates of Canadian Community Health Survey, 2001-2013
Source: https://www.statcan.gc.ca/eng/about/er/hspfr (3/6/2020, 17:54 hours).
142
https://bit.ly/3eJv2Xz; https://bit.ly/3duU90g; https://bit.ly/2U7Kipp and https://bit.ly/30aQBw6 (3/6/2020, 16:25 hours).
60
65
70
75
80
85
90
2001 2003 2005 2007 2009 2011 2013
CATI CAPI Total
161
Table 10.1: An overview of major active health surveys in Canada
Survey Objective Inception Frequency Methods Sample Representative Major themes
Canadian
Community
Health
Survey
(CCHS) -
Annual
component
A cross-sectional,
flexible survey for
data at sub-
provincial levels
(health region /
combined health
regions)
143
on
health status,
health care
utilization and
determinants –
single data source
for research on
small populations
and rare
characteristics
2000
Annual
(2007-),
Biennial
(2001-05)
Instrument with 3
components:
common (core
and theme),
optional (unique
provincial and
territorial data
needs), rapid
response (data
on an emerging
or specific issue)
- computer-
assisted personal
and telephone
interviews (CAPI
and CATI)
12+ year
olds living in
all the 10
provinces
and 3
territories -
65,000
respondents
annually
(2007-),
130,000
biennially
(2001-05)
Health region
level on a
biennial basis -
data years can
be combined to
study small
populations /
rare
characteristics
Diseases and health
conditions - health
care services - lifestyle
and social conditions -
mental health and
well-being - record
linkages
Canadian
Health
Measures
Survey
(CHMS)
A cross-sectional
survey to assess
extent of chronic
and infectious
diseases, lifestyle
characteristics and
environmental
exposures as well
as help explore
emerging public
health issues
2007 Biennial
2 questionnaires:
household (HH)
interview and
physical
examination in a
mobile clinic (as
in NHANES, US)
3-79 year
olds in 10
provinces -
6,361 HHs,
7,944
persons,
5,786
physical
examination
National
Diseases and health
conditions -
environmental factors
- lifestyle and social
conditions
143
‘Health region’ refers to administrative areas defined by the provincial ministries of health. For details, kindly refer to https://bit.ly/2ZLRjzO (28/5/2020, 20:46 hours).
162
Survey Objective Inception Frequency Methods Sample Representative Major themes
Canadian
Health
Survey on
Children and
Youth
(CHSCY)
A cross-sectional
survey to paint a
portrait of health
and well-being of
children and youth
and the factors
influencing their
physical and
mental health
2017 Occasional
Questionnaire:
online (self) /
phone (StatCan
interviewer)
1-17 year
olds in 10
provinces, 3
territories -
92,170 raw
units
National,
provincial,
territorial
Chronic conditions -
nutrition - injuries -
physical activity - time
spent in school and
extracurricular
activities - use of
electronic devices -
social environment
(family, friends, and
communities)
Childhood
National
Immunization
Coverage
Survey
(CNICS)
A cross-sectional
survey to assess
child immunization
in accordance with
the immunization
schedules for
publicly-funded
vaccines, parental
knowledge, beliefs
attitudes about
vaccines
2011 Biennial
CATI with the
person most
knowledgeable
(PMK) about the
child’s
immunizations,
child’s vaccine
history from
child’s health
care provider(s)
Children
aged 2, 7,
14 and 17
years in 10
provinces, 3
territories -
14,960 units
National,
provincial,
territorial
Child immunization -
knowledge, attitudes
and beliefs - education
- income - ethnicity -
immigration status
Survey on
Maternal
Health (SMH)
A random, cross-
sectional survey to
monitor maternal
health (pregnancy
and postpartum
experiences) and
improve the health
and wellness of
women
2018 One-time
Electronic
questionnaire
(EQ) / CATI
Biological
mothers
who gave
birth (1 Jan
to 30 Jun
2018) in 10
provinces -
13,000 initial
sample
Provincial
Maternal health and
well-being - pregnancy
and postpartum
experiences
163
Survey Objective Inception Frequency Methods Sample Representative Major themes
Longitudinal
and
International
Study of
Adults (LISA)
To improve the
understanding of
what is happening
in the lives of
Canadians - how
people’s lives
change over time -
causality between
major life
experiences and
their impact on the
educational,
employment and
financial outcomes
- to help assess
what services are
suitable for them
2011 Biennial
In-person CAPI
interview
(StatCan) -
record linkages
from other
surveys or
administrative
data sources
15+ years -
ca. 10,000
households
in 10
provinces
(1 eligible
person per
household)
Provincial
Education and training
- labour market
activities and job
characteristics - work
schedules - income,
pensions and finances
- life satisfaction - life
after retirement and
planning for it - family
and relationship status
homelessness -
caregiving - health
(child, self-reported,
mental, work-related,
disability)
Canadian
Health
Survey on
Seniors
(CHSS)
A supplement to
the CCHS (annual
component), this
cross-sectional
survey aims to
help better
understand what
contributes to
healthy aging and
help policymakers
to make informed
decisions about
health care, social
services and
support programs
for the seniors
2019 Occasional
Computer-
assisted
interviewing (CAI)
65+ seniors
- 10
provinces
(oversample
in 8
provinces to
achieve the
sample
target) -
ca. 25,000
respondents
a year
National,
provincial
Diseases and health
conditions - lifestyle
and social conditions -
mental health and
well-being - health
care services
164
Survey Objective Inception Frequency Methods Sample Representative Major themes
Canadian
Survey on
Disability
(CSD)
A cross-sectional
survey to provide
information about
Canadian youth
and adults whose
everyday activities
are limited due to
a long-term
condition / health-
related problem
1986 Quinquennial
Self-reported /
interviewer-led
methods:
questionnaire
completed
directly by the
respondent on-
line (rEQ) /
conducted by the
interviewer over
telephone (iEQ)
(2017-)
15+
disabled -
10
provinces, 3
territories -
50,000
persons
National
Type and severity of
disability - use of aids
and assistive devices -
daily help received or
required - therapy and
social service supports
use - educational
attainment - labour
force participation -
accommodations at
school / work -
experience of being
housebound - internet
use - methods used to
access government
services - sources of
income
Impacts of
COVID-19 on
Canadians
Collects data on
the current
economic and
social situation as
well as people’s
physical and
mental health to
assess needs of
communities and
implement suitable
support measures
during and after
the pandemic
2020
(24 April)
Weekly
Electronic
questionnaire -
participant self-
completion
All from 10
provinces, 3
territories -
no sampling
- crowd-
sourcing
initiative -
46,000
respondents
(24 Apr - 11
May 2020)
for mental
health
component
Not known
Health - mental health
and well-being -
disability - economic
accounts - income and
expenditure accounts
Source: Statistics Canada and other sources. https://bit.ly/3dhpdAx (28/5/2020, 20:33 hours). Developed by author.
165
Let us now discuss some of the major surveys in some detail. Let us reiterate that we have adopted
the description of these surveys from official sources and referenced them, with minimal changes
/ rephrasing from our side, given the resource constraints. However, we have tried to put together
information from a variety of sources to provide a fuller overview of the surveys described below.
Canadian Community Health Survey (CCHS) – Annual Component
In 1991, the National Task Force on Health Information highlighted a series of challenges with the
country’s health information system. StatCan, CIHI and Health Canada came together to create a
Health Information Roadmap to address these challenges. CCHS was one of the key outcomes.
The CCHS is a cross-sectional survey that collects information related to health status, health care
utilization and health determinants for Canadian population. It is offered in both official languages.
It has a large sample of respondents, designed to provide reliable estimates at the health region
level every 2 years. The CCHS produces an annual microdata file and a file combining 2 years of
data. Collection years can also be combined by users to study populations or rare characteristics.
The CCHS has the following objectives –
1) Support health surveillance programs by providing health data at the national, provincial and
intra-provincial levels;
2) Provide a single data source for health research on small populations and rare characteristics;
3) Timely release of information easily accessible to a diverse community of users;
4) Create a flexible survey instrument that includes a rapid response option to address emerging
issues related to the health of the population.
The survey began collecting data in 2001 and was repeated every two years until 2005. Starting
in 2007, data for the Canadian Community Health Survey (CCHS) were collected annually instead
of every 2 years. While a sample of approximately 130,000 respondents were interviewed during
reference periods of 2001, 2003 and 2005, the sample size was changed to 65,000 respondents
each year starting in 2007 (data collection period: January to December).
In 2012, CCHS began work on a major redesign project that was completed and implemented for
the 2015 cycle. The objectives of the redesign were to review the sampling methodology, adopt
a new sample frame, modernize the content and review the target population. Consultations were
held with federal, provincial and territorial share partners, health region authorities and academics.
Target population
The CCHS covers the population 12 years of age and over living in 10 provinces and 3 territories,
with certain exclusions.
166
Instrument design
Each component of the CCHS questionnaire is developed in collaboration with specialists from
StatCan, other federal and provincial departments and / or academic fields. CCHS questions are
designed for computer-assisted interviewing (CAI) – as questions were developed, the associated
logical flow into and out of the questions was programmed. This includes specifying the type of
answer required, the minimum and maximum values, on-line edits associated with the question
and what to do in case of item non-response.
The CCHS content is comprised of 3 components – 1) common content (core content and theme
content), 2) optional content, and 3) rapid response content. The core content is collected from
all survey respondents and remains relatively unchanged over several years. The theme content,
also collected from the entire sample, varies from year to year. The optional content fulfils the
unique data needs of each province and territory, and may vary from year to year. The rapid
response component is offered to organizations interested in national estimates on an emerging
or specific issue related to population health. Provincial estimates may be derived from the rapid
response component; however, they may be of limited quality. A rapid response component may
be added to the survey in each 3-month collection period. Data is released about 6 months after
the collection period through an announcement in The Daily.
144
It also needs to be noted that, until the 2015 redesign, CCHS had cycles. In addition to the main
cycles, CCHS had special surveys on specific themes – cycle 1.2 on mental health (2002), cycle
2.2 on nutrition (2004) and cycle 4.2 on healthy aging (2008).
145
A decade later, a pilot CCHS –
Nutrition (CCHS-N) was conducted in 2014, followed by a survey with a desired sample of 24,000
respondents in 37,694 selected dwellings, aged 1+ years, living in 10 provinces.
146
New modules and revisions to existing CCHS content are tested using different methods.
Qualitative tests using individual cognitive interviews or, more rarely, focus groups are used to
ensure that questions and concepts are appropriately worded.
The computer application for data collection is extensively tested in-house each time changes are
made. The objective of these tests is to identify any errors in the program flow and text before the
start of the main survey.
144
The Daily (TD) is StatCan’s official release bulletin, its first line of communication with the media and the public. TD
issues news releases on current social and economic conditions and announces new products, offers a comprehensive
overview of new information available from StatCan. It is released at 8:30 am Eastern time each working day, has been
published since 1932 and posted on the internet since 1995. https://bit.ly/3gIX4V0 (3/6/2020, 21:08 hours).
145
https://www150.statcan.gc.ca/n1/pub/91-549-x/2009001/par7-eng.htm (4/6/2020, 11:19 hours).
146
https://www23.statcan.gc.ca/imdb/p2SV.pl?Function=getSurvey&Id=201486 (4/6/2020, 11:27 hours).
167
Sampling
CCHS is a sample survey with a cross-sectional design.
To provide reliable estimates at the health region (HR) level, a sample of 130,000 respondents is
required on a 2-year basis – 120,000 respondents to cover the population aged 18 and over and
10,000 respondents to cover the population aged 12 to 17 years.
Since 2015, a multi-stage sample allocation strategy has been used to give relatively fair sample
distribution to HRs and provinces. For each age group (18 and over, 12 to 17), the sample is first
allocated among the provinces using a power allocation of 0.75 according to the size of their
respective population. Each province’s sample is then allocated among its HRs using a power
allocation of 0.35 according to the size of the population in each HR.
From 2015 onwards, the CCHS sample is selected using 2 different frames – an area frame and
the Canada child benefit (CCB) frame. Using the area frame, a sample of dwellings is selected to
target the population aged 18 and over. During collection, all members of the dwelling are listed
and a person aged 18 years or over is automatically selected using various selection probabilities
based on age and household composition. The CCB frame is used to sample persons aged 12 to
17 years. One child is then pre-selected to complete the survey.
The area frame is mainly designed to serve the Labour Force Survey (LFS). Thus, the sampling
plan of the LFS must be considered in selecting the CCHS dwelling sample. The LFS plan is a
complex 2-stage stratified design in which each stratum is formed of clusters. The LFS first selects
clusters using a sampling method with a probability proportional to size (PPS), and then the final
sample is chosen using a systematic sampling of dwellings in the cluster. For CCHS, LFS clusters
are grouped in each HR. Then, a sample of clusters and systematic dwellings are selected in each
HR. The process maximizes overlaps between clusters selected by both surveys and ensures that
the same dwelling is selected only once.
For the CCB frame, an HR is assigned to each child in the target population based on the address.
The CCB frame is then stratified by HR. A simple random sample (SRS) of children aged 12 to 17
is selected within each HR.
The size of the sample is enlarged during the selection process to account for non-responses and
units outside the coverage.
Data sources
Data are collected directly from survey respondents.
Data are collected using CAPI and CATI software. Cases from the area frame are collected using
a combination of both modes, while CCB cases are collected exclusively by telephone interview.
168
In both cases (area frame and CCB), proxy reporting is allowed, although certain questions may
be skipped.
Respondents are initially offered to complete the interview in either English or French. To remove
language as a barrier to conducting interviews, each StatCan Regional Office recruits interviewers
with a wide range of language competencies. If necessary, cases are transferred to an interviewer
with the language competency needed to complete an interview.
The average time to complete the survey was 50 minutes.
The information collected during 2018 CCHS was linked to tax records of respondents and that
of all household members. Household information (address, postal code, and telephone number),
respondent’s information (social insurance number, surname, name, date of birth / age, sex) and
information on other members of the household (surname, name, age, sex and relationship to
respondent) are key variables for the linkage.
Respondents are notified of the planned linkage before and during the survey. Any respondent
who objects to the linkage of their data have their objections recorded and no linkage to their tax
data takes place. Income information obtained from income tax records will also be provided to
federal, provincial and territorial share partners only with respondents’ consent.
Error detection
Most editing of data is performed at the time of interview by CAI application. It is not possible for
interviewers to enter out-of-range values and flow errors are controlled through programmed skip
patterns. For example, CAI ensures that questions that do not apply to a respondent are not asked.
In response to some types of inconsistent / unusual reporting, warning messages are invoked but
no corrective action is taken at the time of the interview. Wherever appropriate, edits are instead
developed to be performed after data collection at the Head Office. Inconsistencies are usually
corrected by setting one or both of the variables in question to ‘not stated’.
Imputation
Household income data in 2018 CCHS is imputed. Missing values due to either respondent refusal
or respondent’s lack of knowledge of household income was replaced using a nearest neighbour
imputation method based on a modeled household income.
Quality evaluation
Throughout the data collection process, control and monitoring measures are put in place and
corrective action is taken to minimize non-sampling errors. These measures include response rate
169
evaluation, reported / non-reported data evaluation, on-site observation of interviews, improved
collection tools for interviewers and others. Subsequently, 3 data validation steps are undertaken.
A validation program is run in order to compare estimates for the health indicators taken from the
common content with previous years. This validation is performed at various geographical levels
as well as by age and sex. Significant differences are examined further to find any anomalies in
data. The work of analysts who use CCHS data allows for an in-depth look at many variables of
the survey and represents a very effective way to find errors.
Key themes in CCHS 2020
➢ Activity limitation
➢ Biking
➢ Body mass index
➢ Breastfeeding initiation
➢ Chronic disease
➢ Citizenship and immigration status
➢ Colorectal cancer screening
➢ Contact with health professionals
➢ Difficulties accessing health information or advice
➢ Disability
➢ Drinking status
➢ Exclusive breastfeeding
➢ Food insecurity of households and persons
➢ Fruit and vegetable dietary practices
➢ Functional health status
➢ Health-adjusted life expectancy
➢ Health care received
➢ Household total income
➢ Influenza immunization
➢ Injury
➢ Leisure-time physical activity level
➢ Life satisfaction
➢ Life stress
➢ Mammogram
➢ Mood disorder diagnosis
➢ Neurological condition
➢ Pain or discomfort
➢ Pap smear
➢ Perceived health and mental health
➢ Quality rating of health care services received
➢ Regular family physician
➢ Sense of belonging to local community
➢ Smoking
147
147
https://bit.ly/2z5Go9o and https://bit.ly/2U7kauY (3/6/2020, 20:42 hours).
170
Canadian Health Measures Survey (CHMS)
Launched in 2007, CMHS collects key information related to the health of Canadians by means of
direct physical measurements like blood pressure, height, weight and physical fitness. In addition,
the survey collects blood, urine, saliva and hair samples to test for chronic and infectious diseases,
nutrition and environment markers – storing blood, urine and DNA samples at the CHMS Biobank
for future health research projects. Through household interviews, it gathers information vis-à-vis
nutrition, smoking and alcohol use, medical history, health status, sexual behaviour, lifestyle and
physical activity, environmental, housing, demographic and socioeconomic characteristics.
This information is supposed to create a national baseline data on the extent of such major health
concerns as obesity, hypertension, cardiovascular disease, exposure to infectious diseases and
environmental contaminants. Additionally, the survey provides clues about illness and the extent
to which many diseases might be undiagnosed among Canadians. It helps to determine linkages
between disease risk factors and health status as well as highlights emerging public health issues.
The CHMS data are representative of the population, whether they are healthy or not, and provide
a picture of the actual health of Canadians. Following are some of the measures that it covers –
Physical measures
➢ Anthropometry (standing height, weight, waist circumference, neck circumference)
➢ Cardiovascular health and fitness
➢ Musculoskeletal health and fitness
➢ Physical activity (accelerometry)
➢ Vision (visual acuity, visual field, retinal photography, intraocular pressure)
Blood measures
➢ Nutritional status (for e.g., Vitamin B12, Vitamin D, ferritin)
➢ Diabetes (for e.g., glucose, glycated hemoglobin A1c)
➢ Cardiovascular health (for e.g., apolipoprotein A1 and B, lipid profile)
➢ Musculoskeletal health
➢ Environmental exposure
➢ Infection marker (toxoplasmosis)
Urine measures
➢ Environmental exposure
➢ Nutritional status (for e.g., iodine, sodium, potassium)
➢ Infection marker (chlamydia trachomatis)
171
Saliva measures
➢ DNA extraction for future health research projects
Hair measures
➢ 25 metals and trace elements (e.g., lead, cadmium, mercury)
<
The CHMS team works closely with the Health Canada and PHAC Research Ethics Board and the
Office of the Privacy Commissioner of Canada in order to address privacy issues and to implement
proper laboratory procedures.
Target population
The target population for CHMS consists of persons 3 to 79 years of age living in the 10 provinces,
with approximately 4% of the target population excluded.
Instrument design
Two questionnaires were used for cycle 5 (January 2016 to December 2017) of the CMHS.
148
Household questionnaire
The household questionnaire content was developed with input from stakeholders (Health Canada
and PHAC) as well as from the external experts who participated as members of various advisory
committees. Prior to finalizing questions, one-on-one qualitative test interviews were conducted
to look at specific questionnaire content, particularly the content new to cycle 5. As a result of this
testing, improvements were made to questionnaire wording, instructions and the flow of questions.
Clinic questionnaire
Development of the clinic questionnaire proceeded in much the same way as that of the household
questionnaire. Its content was developed by means of a comprehensive consultation process, and
multiple iterations of collection application were generated. Each iteration was assessed on flow
within the mobile examination center (MEC) for both the respondent and staff. Quantity and quality
of data collected was also assessed. The clinic questionnaire includes a set of self-reported health
questions similar to the type of questions asked within the household questionnaire. The questions
included at MEC are related to medication use, fish / shellfish consumption and vision. In addition,
148
CHMS cycle 5 questionnaires are available at – https://bit.ly/2BnX64r (4/6/2020, 12:21 hours).
172
the questionnaire includes introductory text / instructions, screening and administrative questions
related to the physical measures tests conducted at the MEC.
Sampling
This is a sample survey with a cross-sectional design.
The CHMS uses a stratified 3-stage sample made up of 1 or 2 selected respondents from each
dwelling selected in a sampled collection site.
The sampling unit at the first stage is a collection site. A collection site is a geographical unit
limited to a radius of about 50 km in urban areas and up to 75 km for rural areas. The sampling
unit at the second stage is the dwelling and at the third stage, the sampling unit is the person.
The CHMS consists of a full sample and several sub-samples.
For the full sample, at the first stage, a sample of 16 collection sites was required. The sites are
allocated by region: Atlantic (2), Quebec (4), Ontario (6), Prairies (2) and British Columbia (2).
Within each region, sites are sorted according to the size of their population and whether or not
they belonged to a census metropolitan area. Within the Prairies and Atlantic regions, they were
first sorted by province. Sites are then randomly selected using a systematic sampling method
with probability proportional to the size of each site’s population.
The sample size determination and allocation for the second and third stage are done together.
The target sample size for cycle 5 was 5,700 respondents for the clinic component of the survey,
which works out to approximately 356 respondents per collection site. To determine the number
of dwellings to sample in each collection site to reach this target, previous response rates were
used from both the CHMS and the CCHS. The CHMS and CCHS are both used to calculate –
➢ The expected probability that a dwelling would be eligible for the CHMS (the eligibility rate)
➢ The expected probability that a roster of all occupants of the household would be completed
(the roster rate)
➢ The expected probability that a selected person would respond to the household questionnaire
(the questionnaire rate)
Finally, rates from the previous CHMS sites are used to calculate the expected probability that a
household questionnaire respondent would also be a respondent to the clinic (the clinic rate).
The sample is allocated amongst the 6 age group strata (3-5, 6-11, 12-19, 20-39, 40-59 and 60-
79), with a small portion of the sample going to an “other” stratum. A maximum number of 35
dwellings per site is selected in this stratum, with fewer being selected for sites that had fewer
dwellings in the stratum. This stratum size helped to prevent extreme dwelling sampling weights.
173
The allocation of the dwelling sample to each of the age group strata is done to allow for the best
chance of meeting the age and sex clinic respondent targets for cycle 5 without going too far over.
Where possible, the sample is allocated in a way that emphasized the strata where more sample
was required to meet the targets.
Once the sample of dwellings is in the field, when the household interviewer makes contact with
a sampled dwelling, the goal is to create a roster for the household. A roster is a list of all persons
residing in the household and includes pertinent information such as age, sex and whether the
individual works full-time for the Canadian Forces. With this information, the computer application
randomly selects one or two persons to take part in the remaining part of the survey, including
the questionnaire and the clinic visit. The number of persons selected depends on the composition
of the household –
➢ If there is at least one child between the ages of 3 and 11, two people are selected: one child
between the ages of 3 and 11 and one other person between the ages of 12 and 79
➢ If there are no children between the ages of 3 and 11, only one person in the 12 to 79 age
group is selected
➢ If there are no one eligible for the survey, no one is selected. This includes households where
all in-scope persons are under the age of 3, over the age of 79 and / or are full-time members
of the Canadian Forces
<
When the roster is completed, the computer application assigns a sampling factor to each eligible
member of the household and this information is used to determine the probability of selection.
The sampling factor assigned to each individual is based on their age group and sex and the
factors vary between groups in order to do a better job of reaching the clinic targets for each age
group by sex. In households where two people are selected, the selection of the child (aged 3-11)
is done independently of the person aged 12 to 79.
Data sources
Data are collected directly from survey respondents.
Collection includes a combination of a personal interview using a CAI method and, for the physical
measures, a visit to an MEC specifically designed for the survey.
CHMS collects data in 16 sites across the country. The collection sites are located in 7 provinces.
Collection is scheduled so that each region is distributed in the 2-year collection period, between
seasons and in a way which tries to minimize the movement of staff and equipment between sites.
The MEC stays in a site for 5-7 weeks, collecting measures from nearly 350 respondents per site.
174
First step: personal interview at the household
The first contact with respondents is a letter sent through mail. The letter informs people living at
the sampled address that an interviewer will visit their home to collect some information about the
household. At home, the application randomly selects one or two respondents and the interviewer
conducts a separate health interview with each of them. The interview takes 45 to 60 minutes per
respondent. The interviewer then assists the respondent in setting an appointment for the physical
measures at the CHMS MEC.
Second step: visit to the CHMS MEC
MECs are used to conduct physical measures portion of the survey. Similar MECs have been used
successfully for years for NHANES in the US (as discussed in the previous chapter).
The MEC consists of 3 trailers (side by side), linked by enclosed pedestrian walkways. One trailer
serves as a reception and administration area, the second has a laboratory and physical measure
room, while the third contains additional physical measure rooms.
For each respondent, a complete visit to MEC lasts for about 2 hours. This is an approximate time
since each respondent is assessed for their suitability for each measure and tested accordingly.
For children under 14 years of age, a parent or legal guardian has to be present at the MEC and
has to provide written consent for the child to participate in the tests.
At the end of their visit to the MEC, respondents are provided with a waterproof activity monitor.
This small device is worn for a week at all times except when sleeping – even when swimming or
bathing. It records information about normal physical activity patterns without respondents having
to do anything special.
Respondents are also provided with materials to send a second urine sample from home to a
laboratory for nutritional analysis.
Error detection
Most editing of the data is performed at the time of the interview by the CAI application. It is not
possible for interviewers / HMS to enter out-of-range values and flow errors are controlled through
programmed skip patterns. For example, CAI ensures that questions that did not apply to the
respondent were not asked. Edits requiring corrective action were incorporated in CAI application
to deal with inconsistent responses. In addition, warnings not requiring corrective action were also
included to identify unusual (i.e., improbable rather than impossible) values as a means of catching
potential errors and allowing correction at source. At head-office, the data undergoes a series of
processing steps that results in some of the data being adjusted. As a final validation step, the CAI
175
edits are re-applied to the processed data. As a result, final data are complete and contain reserve
codes for responses of ‘less than limit of detection’, ‘valid skip’, ‘don’t know’, ‘refusal’, ‘not stated’.
Imputation
Questions on personal and household income were not asked in cycle 5 – instead, respondents
were asked permission to use administrative files to obtain their income. The process involved
finding personal income on administrative files for all members of the respondent’s household and
then to sum their income to derive the household income. However, in certain cases, the personal
income was not obtained for all members of the household because either the permission to use
administrative files was not given or it could not be linked / found for those who gave permission.
In these cases, the personal income was imputed.
➢ The personal income was set to zero for people aged 0-19 who gave permission, but for whom
no income was found on the administrative files
➢ For people aged 15-19 years old who did not give permission and people 20-39 years old who
refused or could not be found on the administrative files, their personal income was imputed
as the median revenue of their imputation class based on age, sex and collection site
➢ For people 40+ years old who did not give permission or could not be found, their personal
income were imputed using the median, but with imputation classes based on 5-year age
groups, sex and collection site
The personal income of all household members, whether found / linked / imputed, is then added
to derive the household income of the respondent. The personal income of the respondent is also
kept on the file.
Quality evaluation
One of the unique features of the CHMS is that 3 different sets of data are collected for the same
respondent – household interview data, physical measures data and laboratory results data. Each
set of data has to be processed on its own. Yet, they cannot be completely separated from each
other because, at various points during processing, the 3 sets of data have to be used together.
The processing of the household interview data was performed in a manner similar to that of other
health surveys at StatCan. The data are validated first at the record level, then at individual variable
level, followed by detailed top-down editing. During data collection, processing takes place on a
daily basis. The household interview responses have to be processed quickly in order for the data
to be available at the MEC in time for the respondents’ visit to the MEC.
176
Similarly, the processing of the physical measures data begins with the data being validated first
at the record level, then at individual variable level, followed by detailed top-down editing. Also,
because the laboratory tests are determined based on responses received at the MEC, the MEC
data are used to generate a file containing a list of tests for which laboratory results are expected
to be received. This laboratory control file is used in processing the laboratory results data.
The processing of laboratory data involves significant file manipulation due to the fact that several
different file types are received from the MEC and the various reference laboratories. As with the
household and physical measures data, the laboratory data are validated at the record level, then
at the individual variable level and several new variables are subsequently derived. The laboratory
data are processed as quickly as possible so that any results that have been identified as outside
of a normal range at the reference laboratories and the MEC are available in a timely fashion for
reporting to respondents.
Non-sampling errors
Much time and effort were devoted to reducing non-sampling errors in CHMS. Quality assurance
measures were applied at each stage of data collection and processing cycle to control the quality
of the data. In cycle 5 of CHMS, there was little partial non-response, since once the questionnaire
began, respondents tended to complete it. There was total non-response when the person selected
to participate in the survey refused to do so or could not be contacted by the interviewer. Cases
of total non-response were taken into account during weighting by correcting weights of persons
who responded to the survey in order to compensate for those who did not respond.
Response rates
In all, 8,539 dwellings were selected within the scope of CHMS cycle 5. Of these dwellings, 6,361
agreed to provide information on the composition of the household (response rate - 74.5%). From
respondent households, 8,847 persons were selected (1 or 2 persons a household) to participate
in the survey, of whom 7,944 responded to the questionnaire (response rate - 89.8%). Of these
persons, 5,786 then reported to the MEC for physical measurements (response rate - 72.8%). At
the national level, a combined response rate of 48.5% was observed.
Some respondents who attend the MEC are unable / unwilling to participate in the blood and urine
components. A response rate is derived for each of these components, which are supposed to be
done on the full sample respondents. The response rates for these measures use the full sample
response rates up to the MEC and derive the rest as follows – of 5,786 participants who reported
to the MEC for physical measurements, 5,482 participants provided blood and 5,688 urine. The
combined response rate for blood draw was 46.3%, for urine 47.7%. Likewise for other measures.
177
Canadian Health Survey on Children and Youth (CHSCY)
CHSCY explores issues that have an impact on physical and mental health of children and youth.
It is used by StatCan, Health Canada, PHAC, provincial and territorial health ministries as well as
other federal and provincial departments to monitor, plan, implement and evaluate programs to
improve the health of children and youth. Researchers from various fields are also interested in
the survey data and use it to conduct research into various factors that affect the health and well-
being of children and youth in Canada.
Target population
The 2019 CHSCY covers population aged 1 to 17 as of 31 January 2019, living in the 10 provinces
and 3 territories, covering at least 98% of the total target population in all provinces and 96% in
all Northern territories.
Instrument design
The survey content was developed based on consultation across Canada with key experts and
federal and provincial stakeholders. The goal of the consultation was to provide advice to StatCan
on what survey content would be relevant for programs and policies, and fill data gaps related to
children and youth. The questionnaire was developed by StatCan with PHAC and Health Canada.
Qualitative test by StatCan’s Questionnaire Design Resource Centre using face-to-face interviews
and focus groups was conducted during 2014-18.
Sampling
CHSCY is a sample survey with a cross-sectional design. Its sampling frame is the Canadian Child
Tax Benefit file. Sampling units are children and youth aged 1 to 17 years as on 31 January 2019.
In terms of geography, the sample is primarily stratified by province, and further into 3 age groups
– children aged 1 to 4 years, 5 to 11 years and youth aged 12 to 17 years. The sample size for the
survey was 92,170 raw units.
Data sources
Data was collected directly from the survey respondents. Respondents are given the opportunity
to complete an e-questionnaire. If it was not completed by 31 March 2019, a StatCan interviewer
called and asked the respondent to complete the questionnaire over the telephone.
178
Error detection
Some data editing was performed at the time of the interview within the electronic questionnaire.
The questionnaire had built-in checks for out-of-range or extreme values that prompt respondents
and interviewers to verify the recorded answer. Flow errors were controlled in the application via
programmed skip patterns – for e.g., questions that did not apply to a respondent were not asked.
149
Survey on Maternal Health (SMH)
The main objective of SMH is to collect information from biological mothers about their pregnancy
and postpartum experiences. Results from the survey are used by researchers and policymakers
to monitor and improve the health and wellness of women.
Target population
The target population for the survey is the set of biological mothers who have given birth between
1 January and 30 June 2018 in the 10 provinces. Persons living in institutions or on an Indigenous
reserve were excluded.
Instrument design
Content for the electronic questionnaire was drafted in consultation with PHAC. The questionnaire
underwent cognitive testing in the form of in-depth interviews in both official languages, conducted
by StatCan’s Questionnaire Design Resource Centre.
Sampling
SMH is a random, cross-sectional, targeted respondent survey. Its frame was stratified by province
and a simple random sample of mothers was selected independently in each province. Sufficient
sample was allocated to each of the provinces so that the survey could produce provincial level
estimates. An initial sample of 13,000 mothers was selected and sent to collection.
Data sources
Data was collected from the survey respondents either through an electronic questionnaire (EQ)
or through CATI (computer-assisted telephone interviewing).
149
https://www23.statcan.gc.ca/imdb/p2SV.pl?Function=getSurvey&SDDS=5233 (4/6/2020, 16:08 hours).
179
Response rate
The response rate for SMH 2018-19 was 54.74%.
150
Canadian Health Survey on Seniors (CHSS)
The CHSS is a supplement to CCHS – Annual component. It collects information related to health
status, health care services, supports as well as social and health determinants of population aged
65 and above. Its objectives are to –
1) Better understand what contributes to healthy aging by collecting data on the health and well-
being of seniors, including their use of health care services and supports, social, demographic,
geographic and economic determinants;
2) Produce estimates on the health of seniors aged 65 and over at the provincial, and for seniors
aged 85 and over at the national, levels;
3) Produce a cross-sectional dataset on the health of seniors that permits analysis on a range of
research questions and surveillance activities;
4) Evaluate changes on certain aspects of health from CCHS - Healthy Aging, 2008-09 survey.
<
The data collected in the survey will be used by StatCan, Health Canada, PHAC, provincial health
ministries as well as federal and provincial health planners across the country. The CHSS will help
policy makers, researchers and planners to make informed decisions regarding health care, social
services and support programs for the ageing population, which will affect all Canadians.
Target population
The CHSS covers the population 65 years of age and over living in the 10 provinces, with certain
exclusions.
Instrument design
Its instrument was developed in collaboration with Health Canada, PHAC and an expert advisory
group. The questions are designed for CAI – as questions were developed, the associated logical
flow into and out of the questions was programmed. This includes specifying the type of answer
required, the minimum and maximum values, on-line edits associated with the question and what
to do in case of item non-response. In collaboration with StatCan’s Questionnaire Design Resource
150
https://www23.statcan.gc.ca/imdb/p2SV.pl?Function=getSurvey&SDDS=5283 (4/6/2020, 16:38 hours).
180
Centre, the questionnaire was subjected to qualitative testing in 2018, which consisted of one-on-
one interviews. The objective was to evaluate respondent reactions to, and understanding of, the
survey as well as their willingness to respond to the questions.
Sampling
The CHSS is a sample survey with a cross-sectional design, a combination of the CCHS – Annual
component respondents from all provinces who are at least 65 years old along with an oversample
in all provinces, except Ontario and Quebec – no oversample was required to achieve the sample
targets in these provinces. The CHSS oversample is selected from a list frame of dwellings with a
valid telephone number that have at least one occupant aged 65 years or older.
Both the CCHS – Annual component for respondents 18+ and CHSS oversample for respondents
65+ have dwelling as the sampling unit. Once contact with the household has been established,
a roster of household members is taken, from which one person is randomly selected to complete
the survey. The CCHS oversample for CHSS purposes is stratified by province – 10,000 dwellings
a year. When combined with the CCHS – Annual component sample (15,000), it is estimated that,
overall, there will be approximately 25,000 respondents to the CHSS per year.
Data sources
Data is collected directly from survey respondents using CAPI and CATI software. Proxy reporting
is allowed, although some modules are skipped in that case. Respondents are initially offered to
complete the interview in either English or French. To remove language as a barrier to conducting
interviews, StatCan regional offices hire interviewers with a wide range of language competencies.
The average time to complete the survey is 15 minutes.
151
Canadian Survey on Disability (CSD)
CSD started out as Health and Activity Limitation Survey: Household Component (HALS) in 1986,
was renamed Participation and Activity Limitation Survey (PALS) in 2001, and finally CSD in 2012.
The purpose of CSD is to provide information about Canadian youth and adults (15+ years) whose
daily activities are limited due to a long-term condition or health-related problem. This information
is used to develop and evaluate policies, programs and services for those living with disabilities
to help enable their full participation in society. In particular, information on adults with disabilities
is essential for the effective development and operation of the Employment Equity Program. Data
151
https://www23.statcan.gc.ca/imdb/p2SV.pl?Function=getSurvey&SDDS=5267 (4/6/2020, 16:51 hours).
181
on disability are also used to fulfill Canada’s international agreement relating to the United Nations
Convention on the Rights of Persons with Disabilities. CSD information is used by all levels of the
government, associations for persons with disabilities as well as researchers working on disability.
The survey collects information on type and severity of disability, use of aids and assistive devices,
daily help received or required, use of various therapies and social service supports, educational
attainment, labour force participation details, requirements and unmet needs for accommodations
at school or work, the experience of being housebound, veterans of the Canadian Armed Forces
with disabilities, internet use, methods used to access government services and sources of income.
Target population
The population covered by the CSD was composed of all persons aged 15 and over (as of 10 May
2016 – Census Day) who reported difficulty ‘sometimes’ / ‘often’ / ‘always’ to one of the Activities
of Daily Living questions on the 2016 Census of Population long form. This included persons living
in private dwellings in the 10 provinces and 3 territories, with some exceptions.
Instrument design
The questionnaire was developed with Employment and Social Development Canada (ESDC). A
content review of the 2012 CSD was conducted with subject matter experts and stakeholders to
identify any potential data gaps. Input was obtained from ESDC’s Technical Advisory Group (TAG)
on disability, consisting of representatives from various community associations across Canada.
Specialized consultation was also provided by Veterans Affairs Canada and Service Canada.
New questions were added to the 2017 CSD, including 7 new modules in the areas of episodic
disabilities, general health, use of various therapies and social services supports, the experience
of being house-bound, veterans of the Canadian Armed Forces, internet use and accessibility of
government services. Two age-related questions were introduced for each of the 10 disability
types. The first asked respondents at what age they began having difficulty with a health-related
problem or condition; the second determined at what age the difficulty or condition began to limit
their daily activities. Also in 2017, CSD underwent transformation from a CATI to an EQ mode.
The labour force activities component of the questionnaire also underwent modifications to better
reflect standard employment indicators found in labour surveys by StatCan (harmonized content).
The 2017 questionnaire was tested in both official languages. Qualitative content testing was
conducted by Questionnaire Design Resources Centre at StatCan and in several off-site locations
across Canada. The EQ application underwent qualitative testing by the Centre along with ESDC.
An in-depth review of 2016 Census variables was undertaken for potential record linkage with the
CSD dataset, leading to addition of nearly 300 Census variables covering 15 subject areas linked
to the final CSD data files for 2017.
182
Sampling
CSD has a cross-sectional, stratified two-phase design based on the 2016 Census. The first phase
is the Census itself, corresponding to the sample of households selected to receive the long form,
with about 1 out of 4 households systematically selected across Canada. Phase 2 corresponds to
the sample of persons who reported having difficulty on the Activities of Daily Living questions on
the long form Census. Sampling unit for phase 1 (Census) is the household, phase 2 the person.
The total sample size for CSD 2017 was 50,000 persons. The sample was drawn using systematic
sampling with the frame being sorted by collection unit to minimize the chance of selecting more
than one person per household.
While CSD did not cover persons who responded ‘no’ difficulties or conditions on the Activities of
Daily Living questions on the Census questionnaire, a sample of these individuals (called the NO
sample) was nonetheless included in the final CSD data files. These people are all considered to
be people without a disability. This sample allows the computation of disability rates, which require
estimates for the entire population, not just persons with a disability.
An additional sample of nearly 5,000 persons was also drawn as part of a methodological research
project. This sample of persons was also drawn among persons who did not report any difficulties
or conditions on the Activities of Daily Living questions on the Census. The Disability Screening
Questions were asked of these persons to see if they had a disability or not to allow methodologists
to determine the extent to which new questions on the Census covered persons with a disability.
Data sources
Data were collected directly from survey respondents and also linked from the 2016 Census. Data
collection for CSD was done using an EQ, involving 2 types of collection methods – a self-reporting
method with the questionnaire completed directly online by the respondent (rEQ), an interviewer-
led method conducted via telephone (iEQ). All respondents received an invitation to participate in
the survey by mail, with the rEQ respondents receiving a link to the EQ and a secure access code,
while the iEQ respondents were informed that they would be contacted by telephone. Reminder
letters were sent to rEQ respondents nearly 2 weeks apart for the duration of collection. Collection
for rEQ and iEQ were done in parallel for the first part of collection, with all rEQ non-respondents
were transferred to iEQ for follow-up. Overall, nearly 40% of respondents completed the rEQ and
60% the iEQ. All survey responses were kept highly secure through industry-standard encryption
protocols, firewalls and encryption layers. Proxy interviews were allowed under some conditions.
The EQ was available in English and French. Interviews lasted about 35 minutes on average. To
reduce interview time, StatCan combined information from CSD with selected data from the 2016
Census. Data from other surveys or administrative data sources may be added later.
183
Error detection
All responses to the 2017 CSD questions were captured directly in the EQ application, both for
iEQ and rEQ components. For some questions, data underwent a preliminary verification process
when respondents were completing the survey. This was accomplished by means of a series of
edits programmed into the EQ. That is, where a particular response appeared to be inconsistent
with previous answers or outside of expected values, the interviewer or self-reporting respondent
was notified with an on-screen warning message, providing them with an opportunity to modify
the response. Once survey responses were transmitted to the head office, more extensive data
processing for CSD began. This involved a series of steps to convert the questionnaire responses
from their initial raw format to a high-quality, user-friendly database, involving a comprehensive
set of variables for analysis. A series of data operations were executed to clean files of inadvertent
errors, remove duplicate records, edit the data for consistency, code open-ended questions,
create useful variables for data analysis, and finally to systematize and document the variables for
ease of analytical usage.
Imputation
For CSD, discrepancies, logical inconsistencies and missing information were resolved, wherever
possible, by means of automatic, customized deterministic editing rules or manual interventions.
Quality evaluation
Quality assurance measures were implemented at each collection and processing step. Measures
included recruitment of qualified interviewers, training provided to interviewers for specific survey
concepts and procedures, observations of training of interviewers as well as interviews to correct
questionnaire design problems / instruction misinterpretations, procedures to ensure that coding
errors were minimized and edit quality checks to verify the processing logic. Data were verified to
ensure internal consistency and were also compared to other sources when available.
For 2017, the CSD included the full implementation of the Disability Screening Questions (DSQ)
used for identifying persons with disabilities. In 2016, the Activities of Daily Living question on the
Census, which serves to create the sampling frame for CSD, was replaced by new filter questions
taken from the DSQ framework. Qualitative and quantitative testing have shown that the new filter
questions allow for better coverage overall of persons with disabilities, and especially of persons
with less visible disability types (for e.g. pain-related disabilities, memory, learning, development
and mental health). One important consequence of this full implementation is that the disability
rates observed in the 2017 CSD are not comparable to those of the 2012 CSD, but are very much
consistent with what was expected and observed during testing.
184
Response rates
CSD 2017 had an overall response rate of 69.5%. Response rates for the provinces ranged from
66.5% for New Brunswick to 78.5% in Quebec; in territories, from 51.5% in Nunavut to 65.6% in
the Yukon; and by age group, from 62.1% among 15-24 year olds to 77% among 65-74 year olds.
152
Lessons for India
GoI / state / UT governments should consider the following lessons from the Canadian context –
® Although India does have a central statistical agency (MoSPI), it needs to be independent and
proactive like Statistics Canada to develop a rational, systematic and well-coordinated system
of data collection in the country. Fragmentation of India’s statistical landscape, leading to non-
essential data collection and wastage of precious limited resources, is a challenge that cannot
be tackled without doing so. MoSPI needs to step up to its mandate and act ‘as the nodal agency
for planned development of the statistical system in the country’.
153
It should seek assistance
from Statistics Canada in this regard, if needed – the presence of 2 persons of Indian origin at
the helm of the relevant Ministry and the agency should make it easier, even if not necessary.
Nevertheless, given the firmly entrenched federal system of governance in the country as well
as health being a state subject, MoSPI should facilitate – rather than take over – development
of rational, systematic and well-coordinated systems of data collection in the states / UTs too.
However, linkages between state Directorates of Economics and Statistics (DES) in particular
with MoSPI should be made more formal and organized than it is at the moment. MHA should
also consider yielding its own statistical ground to MoSPI. In any case, MoSPI at the central
and DES at the state level do send their statistical staff to various ministries. Why not have a
formally organized, coordinated system of data collection which is proactively led by them?
® Like StatCan, MoSPI and DES should be supported by a set of advisory groups which should
offer impartial and independent advice to them vis-à-vis various aspects of data collection as
well as review their performance from time to time.
® Beyond program MIS – which departments and ministries can manage internally, with MoSPI
/ DES oversight – all data collection should be supervised and quality checked by MoSPI / DES.
Right now, the DES’, in particular, are largely data aggregators without any statistical authority.
® Data linkage is one of the biggest lessons for India from the Canadian context. In order to avoid
enormous duplicity of data collection in the country and the resultant wastage of precious and
limited resources, this is something that MoSPI, DES as well as all government departments
152
https://www23.statcan.gc.ca/imdb/p2SV.pl?Function=getSurvey&Id=321555 (4/6/2020, 16:54 hours).
153
http://www.mospi.gov.in/about-ministry-0 (4/6/2020, 20:39 hours).
185
and ministries should seriously undertake. However, citizen consent and confidentiality of their
data needs to be paramount, another count on which India is hugely lacking. Public perception
and trust around consent and data confidentiality needs to be systematically built up through
positive action. Again, this is not possible without strong and independent central coordinating
statistical agencies both at the Central and state / UT levels.
® When India was liberalizing its economy in 1991, Canada was developing a Health Information
Roadmap (HIR) to tackle challenges with its health information system (HIS), highlighted by its
National Task Force on Health Information. It is time India consider establishing its own HIR to
address the manifold challenges with its HIS.
® If continuous surveys are not feasible (as many health surveys in the US are), India should at
least consider annual / biennial periodicity as in Canada (and also in the US).
® Improving survey response rates is something that Canada can learn from India. Nevertheless,
survey techniques adopted by Canada to ensure data validity, quality and representativeness
offer lessons for India.
® India should consider designing flexible surveys like the CCHS – Annual component with fixed,
variable and rapid content components, taking into cognizance the continuing and changing
needs of various stakeholders. Neither programs nor statistical instruments seem to have this
sort of flexibility, despite India being such a diverse and federal nation.
® India needs to consider developing specialized and dedicated surveys for various population
groups / themes like Canada (and the US). Rather than trying to broaden the scope of NFHS,
its focus should be sharpened and other health surveys included. With data linkages, we can
arrive at integrated views of health and its social determinants from a health-in-all perspective.
For this, even non-health surveys should include certain health components. A good place to
begin with would be a periodic survey of Employees State Insurance Schemes (ESIS), which
provides 6 social security benefits, including health, to its beneficiaries. A similar survey should
also be considered for beneficiaries of other social / government-sponsored health insurance.
® With an array of health surveys – and states doing their own set of surveys – there is no need
to have bloated sample sizes for a few surveys. Survey sampling needs to be smarter in order
to achieve representative data in the most focused, equitable and economically efficient style.
® The CCHS is somewhat like NFHS in terms of its broad population and geographical coverage.
However, what NFHS can learn from it is the breadth of thematic coverage – more importantly,
the flexibility of its instrument.
® India should consider developing a longitudinal survey like LISA, which would help study how
health and its determinants – as well as other aspects of adult life – change over time. It should
also consider developing instruments for comprehensive tracking of COVID-19 as in Canada.
186
11. United Kingdom
Like Canada, the UK has a central UK Statistics Authority (UKSA) – ‘an independent body at arm’s
length from government’. Its objective is to promote and safeguard the production and publication
of official statistics that ‘serve the public good’ – a) informing the public about social and economic
matters; b) assisting the development and evaluation of public policy; and c) regulating the quality
and publicly challenging the misuse of statistics. Its overall mission is to mobilize the power of data
to help the UK make better decisions.
The UK statistical system (UKSS) comprises the –
,,
1) Board of the UK Statistics Authority (BUKSA) – responsible for oversight of the UKSS
2) Office for Statistics Regulation (OSR) – the regulatory arm of the UKSA
3) Office for National Statistics (ONS) – UK’s National Statistical Institute and the largest producer
of official statistics in the country
4) Government Statistical Service (GSS) – a community of all those involved in the production of
official statistics in the country
The majority of official statistics are produced by statisticians operating under the umbrella of GSS
– working in either the ONS, UK’s government departments and agencies, or one of the 3 devolved
administrations (Northern Ireland, Scotland and Wales). Every public agency with a significant GSS
presence – vis-à-vis statisticians involved in the production or use of official statistics – has its own
Head of Profession for Statistics (HPS), while the 3 devolved administrations have their own Chief
Statisticians. The HPS and their staff in each agency are accountable to the National Statistician.
154
Health statistics in the UK
UK’s health statistical system is decentralized, with several agencies collecting and publishing data.
Some of the leading sources of health statistics in England are – ONS, NHS England, NHS Digital,
Department of Health & Social Care, Public Health England, Care Quality Commission, Ministry of
Defence, Department for Education and the Department for Work & Pensions.
Special mention needs to be made here of the NHS Digital, which is responsible for standardizing,
collecting and publishing data and information from across the health and social care system in
England. It has a National Indicator Library – the official hub of health and social care indicators in
England, with the methodology and other details for each indicator specified. It has a collection of
154
https://bit.ly/2MwqCaC and https://bit.ly/3gW33pH (5/6/2020, 11:44 hours).
187
over a thousand datasets for clinical indicators – covering a wide variety of subjects, ranging from
quality to population health and outcome of treatments – supporting clinical staff, commissioners,
researchers and others who need evidence to help with decision-making in health and social care.
Its Clinical Commissioning Group Outcomes Indicator Set (CCG OIS) is an important part of NHS
England’s systematic approach to quality improvement. Its NHS Outcomes Framework (NHS OF)
provides national level accountability for the outcomes the NHS delivers. Its Seven-day Services
are experimental statistics to help in effectively measuring both improvement and variation in care
provision across the week. Its Summary Hospital-level Mortality Indicator (SHMI) reports mortality
at trust level across NHS England, using a standard and transparent methodology. And finally, its
Compendium of Population Health Indicators is a wide-ranging collection of over 1,000 indicators
designed to provide a comprehensive overview of population health at the national, regional and
local levels
155
– somewhat like our own annual National Health Profiles, brought out by the Central
Bureau of Health Intelligence (CBHI), Ministry of Health and Family Welfare, Government of India.
However, not surprisingly, given the colonial legacy, both UK and India seem to be struggling from
a set of similar challenges as far as organization of their health statistical systems are concerned.
However, there are lessons that can be learnt in the way the UK is trying to diagnose and address
those challenges as well as from its existing health statistical system. In 2016, the UKSA convened
a roundtable of leaders from England’s health and care system to discuss how health statistics can
be better organized to facilitate evidence-based decision-making. Following were some of the major
conclusions of the roundtable –
1) Nearly 250 sets of health statistics are produced by 10 organizations, available on a ‘variety of
different websites, in different formats, with no single portal’. Lacking an effective coordinating
mechanism, the decentralised health statistics system in the UK is incoherent and inconsistent;
2) There are huge opportunities in combining administrative data from different sectors;
3) Data collection is significantly costly, burdensome for providers and bodies who produce data;
4) There is duplication of data collection between and within bodies;
5) The health statistical landscape is data-rich, but information-poor – the importance of analysis
has been neglected, as has been the support for analysts and researchers;
6) There are delays in data analysis and publication;
7) Although a coordinated system-wide approach is needed, there is no ‘magic bullet’ to address
these challenges;
8) An independent review of the state of health statistics should be commissioned.
155
https://bit.ly/30ciCnc and https://bit.ly/3gXXLK7 (5/6/2020, 13:41 hours).
188
Table 11.1: Major health surveys in the UK
Survey /
coordinating /
conducting
agencies
Objective Inception Frequency Methods Sample Representative Major themes
Health and
Lifestyle Survey
(HALS)
/ Health
Promotion
Research Trust
/ Social and
Community
Planning
Research
To assess self-reported
health, measured health,
cognitive function,
psychological well-being
and lifestyle of adults in
all areas of Great Britain -
changes in health and
circumstances of the
surviving respondents
1984 Occasional
Random sample
(England, Wales
and Scotland) -
longitudinal /
panel / cohort -
face-to-face
interviews - self-
completion -
psychological
and clinical
measurements
HALS 2 (1991-
92): 25+ years
surveyed in
HALS 1 (1984-
1985) - 5,352
interviews -
3,871 self-
completion -
4,483
measurements
National
Demographic,
working, social
circumstances -
self-reported
health - physical
measures -
cognitive
functioning -
dietary habits –
exercise, work,
leisure - alcohol
- smoking -
beliefs and
attitudes about
disease / health
Health Survey
for England
(HSE)
/ NHS Digital
/ NatCen Social
Research
To monitor changes in
health and lifestyles -
prevalence of specific
health conditions - policy
development - planning
services - monitoring and
evaluating policy
1991 Annual
Household
interview -
individual
interview (at
home) - paper
self-completion -
nurse visits
Around 8,000
adults (16+
years) and
2,000 children
(0-15 years)
England
Physical and
mental health
and well-being -
physical
measures
(nurse) - lifestyle
behaviours -
social care
189
Survey /
coordinating /
conducting
agencies
Objective Inception Frequency Methods Sample Representative Major themes
National Study
of Health and
Wellbeing –
Children and
Young People
/ NHS Digital
/ NatCen, ONS
To look at how common
different kinds of health,
developmental, emotional
disorders are - factors
associated with good and
poor health, well-being
and development - to
inform policy decisions
about the need for child
and adolescent mental
health services - to help
with planning services
1999 Occasional
Home interview
(parent / self)
10,500
children and
young people
(2-19 years) in
England
England
Health – well-
being and
development –
emotional /
conduct /
hyperkinetic /
autism spectrum
disorders -
social media use
– bullying
English
Longitudinal
Study of Ageing
(ELSA)
/ US National
Institute on
Aging (NIA), etc.
/ NatCen
Part of international
longitudinal survey of
ageing and quality of life
(for e.g. LASI in India and
LISA in Canada) which
explores the dynamic
relationships between
health, functioning, social
networks, participation
and economic position as
people plan for, move into
and progress beyond
retirement
2002 Biennial
Longitudinal /
panel / cohort -
multi-stage
stratified
random sample
- face-to-face
interview (CAPI)
- self-completion
(PAPI) - nutrition
questionnaire -
bio-measures
(nurse)
50+ year HSE
respondents -
18,000+
respondents
since 2002
England
Household and
individual
characteristics -
physical health
and activity -
psychosocial
health - social
participation -
work & pensions
- income and
assets - housing
- cognitive
function
190
Survey /
coordinating /
conducting
agencies
Objective Inception Frequency Methods Sample Representative Major themes
Health Survey
Northern Ireland
(HSNI)
/ Department of
Health, Northern
Ireland
/ Central Survey
Unit, Northern
Ireland Statistics
and Research
Agency (NISRA)
To provide a picture of
the health of the Northern
Ireland population - to
monitor the extent to
which government health
targets are being met - to
help authorities
effectively plan local
health services
2010 Annual
Systematic
random sample
- CAPI -
computer-
assisted self-
interviewing
(CASI)
16+ years -
2,866
households,
3,593
interviews
Northern
Ireland
Health - activity
- experience of
health & social
care - diets -
food security -
child health -
well-being scale
- mental and
sexual health –
smoking -
drinking -
measurements
Scottish Health
Survey (SHeS)
/ Scottish
Government
Health
Directorates
/ ScotCen Social
Research, ONS
To estimate prevalence
and monitor trends in
certain health conditions -
associated risk factors
and health behaviours -
cardiovascular disease
and related risk factors
remains the principal
focus - health inequalities
- help monitor progress
towards health targets
1995 Annual
Multi-stage
stratified
random sample
- cross-sectional
- face-to-face
CAI - self-
completion PAPI
- core questions
- rotating
modules -
clinical, physical
measurements
3,899
households -
6,793
respondents:
1,983 children
(0-15), 4,810
adults (16+),
1,204 adults
completed
biological
module
Scotland,
Health Board
and Local
Authority level
(4-year period)
General, mental
and respiratory
health and well-
being - CVD -
diet - physical
activity - obesity
alcohol -
smoking
191
Survey /
coordinating /
conducting
agencies
Objective Inception Frequency Methods Sample Representative Major themes
National Survey
for Wales (NSW)
/ Welsh
Government
/ ONS
Brings together 5 surveys
previously commissioned
by the Welsh government
(including the annual
Welsh Health Survey,
2003-15) - a key source
of information for the
Welsh government, public
sector organisations and
academics on the views
and circumstances of
people in Wales
2016 Annual
One-stage
stratified or
systematic
random sample
- cross-sectional
- 45-minute
face-to-face
interview
16+ year olds
- 11,922
respondents
Wales
Demographics -
health status -
mental well-
being - risk
factors - NHS -
social care -
material
deprivation –
education and
qualifications -
democracy –
culture
COVID-19
Infection Survey
(CIS)
/ Department for
Health and
Social Care,
ONS, University
of Oxford
/ IQVIA UK, The
National
Biosample
Centre
To understand how many
people of different ages
across the UK have
already had COVID-19 -
help the government
work out how to manage
the pandemic better
moving forwards and
protect the NHS from
being overwhelmed
2020 2020-21
Household (HH)
invites - HH can
call IQVIA for an
appointment -
questionnaire -
nose and throat
swabs, blood
samples - 16
home visits (15-
30 mins - once /
weekly 4 times /
monthly 11
times) by health
workers, nurses
ONS survey
respondents
(age 2+ years)
who agreed to
be contacted
again - phase
1: ca. 11,000
households in
England - next
year: 132,000
households
across the UK
UK (by age,
geography)
Antibody tests -
symptoms -
contact - gender
- ethnicity -
occupation -
date of birth -
general
physician (nose
and throat swab
results will be
sent to them) -
linkages with
ONS and NHS
data records
Source: UK Data Service and ONS websites as well as other sources. https://www.ukdataservice.ac.uk/ (6/5/2020, 21:58 hours). Developed by author. 192
The English Health Statistics Steering Group (EHSSG) was formed in 2016 as a remedial measure.
In 2018, it was handed over the responsibility to improve the coherence and accessibility of health
and social care statistics in England by removing the duplication of statistical releases, harmonizing
definitions and methodologies, increasing user engagement and aligning publication dates. To do
so, it collaboratively produced a Work Plan 2019-24, established several theme groups that cover
the breath of the health and care statistical system and developed the ‘Health and Care Statistics
Landscape for England’, providing an overview as well as links to all key official health and social
care statistics to help users find relevant statistics on specific topics and cross-cutting themes in
one central place.
156
Table 11.1 above provides an overview of key health surveys in the UK, including a recent one on
COVID-19. One leading survey from each administration (England, Northern Ireland, Scotland and
Wales) is discussed in detail below.
Health Survey for England (HSE)
Among the evidence provided to the Sir Donald Acheson-led ‘Committee of Inquiry into the Future
Development of the Public Health Function’ in England (1988), ‘the most important … was a lack
of co-ordinated information on which to base policy decisions about the health of the population
at national and local levels’.
157
Consequently, a Central Health Monitoring Unit was established in
the Department of Health in 1989, and an annual survey on health and nutrition was commissioned
to fulfill the purpose. The survey, which initially focused on cardiovascular disease and associated
risk factors, came to be known as the HSE. Started in 1991, HSE is a series of annual surveys, the
2018 survey being the 28th one. It provides regular information that cannot be obtained from other
sources about public’s health and health-related behaviour. Each round includes core questions
covering general health, hypertension, diabetes, social care and health-related behaviours as well
as measurements such as blood pressure, height and weight measurements and analysis of blood
and saliva samples. Additionally, there are modules on specific issues that vary from year to year.
Sometimes, the core sample is augmented by additional samples from a specific population sub-
group such as minority ethnic groups, older people or children.
HSE provides information on children (0 to 15 years) and adults (16 and above), living in private
households in England. It consists of an interview in person, followed by a visit from a nurse, who
takes a number of measurements and samples. A total of 2,072 children and 8,178 adults were
interviewed in the 2018 survey – of them, 1,103 children and 4,825 adults had a nurse visit. The
sample is designed to represent the whole population within practical constraints (time, cost, etc.).
156
https://bit.ly/2XB9tTx, https://bit.ly/3h2tWbk, https://bit.ly/2BLqkYc, https://bit.ly/2UeI7AD (26/10/2019, 19:49 hours).
157
https://api.parliament.uk/historic-hansard/lords/1988/apr/27/public-health-in-england (6/6/2020, 14:57 hours).
193
Sample design
HSE 2018 had a multi-stage, stratified, random probability sample, designed to be representative
of the population living in private households in England. Those living in institutions (such as care
homes) were outside the scope of the survey. The sampling frame was the small user Postcode
Address File (PAF). The very small proportion of households living in addresses not on PAF (less
than 1%) was not covered. The sample consisted of 9,612 addresses selected at random in 534
postcode sectors. All HSE surveys cover the adult population aged 16 and over (up to a maximum
of 10 adults per household). From 1995, the survey has included children aged 2 to 15, and from
2001, infants aged under 2. Up to 4 children a household were interviewed (up to 2 aged between
0 and 12, up to 2 aged between 13 and 15). Where there were 3 or more children in an age band,
2 of the children were selected at random to limit the respondent burden for parents.
Data collection
Data collection involved both interviews and self-completion.
The household interview included questions on household size, composition and relationships;
type of dwelling, tenure, and the number of bedrooms; car ownership; smoking within the home;
the economic status and occupation of the household reference person; and household income.
Adults were asked to participate in a face-to-face interview which included a self-completion
questionnaire. The contents of the self-completion booklets varied by age – young adults aged 16
to 17 were asked about smoking and drinking behaviour as well as other questions. Interviewers
also had the option of using this booklet for those aged 18 to 24 if they felt that it would be difficult
for anyone in this age group to give honest answers to questions face-to-face with other household
members present.
Children aged 0 to 15 years were also interviewed, and were eligible for a nurse visit. During the
interview, those aged 13 to 15 answered themselves, while parents answered on behalf of children
aged 0 to 12. In addition, children aged 8 and over answered questions on some sensitive topics
within a self-completion questionnaire.
Self-reported longstanding conditions among adults and children were examined using data from
the 2017 and 2018 surveys. Participants were asked – ‘Do you have any physical or mental health
conditions or illnesses lasting or expected to last 12 months or more?’. If the answer was positive,
it was further asked – ‘What is the matter with you?’ and their answers for up to 6 conditions were
recorded verbatim. These were grouped into 14 categories of WHO’s ICD-10, covering infectious
and non-communicable diseases of the body and mind.
Interviewers also measured the weight of all participants and the height of everyone aged 2 and
over. In 2018, for the first time, only a proportion (89%) of addresses were eligible for nurse visits.
194
In these households, nurse visits were offered to all participants who were interviewed (both adults
and children). The nurse visit included questions about prescribed medicines and, for adults, folic
acid and nicotine replacement products. Nurses took waist and hip measurements for those aged
11 and over and measured the blood pressure of those aged 5 and over. Adults were also asked
to provide non-fasting blood samples for the analysis of total cholesterol and HDL cholesterol, and
glycated haemoglobin. Samples of saliva were taken from children aged between 4 and 15 for the
analysis of cotinine (a derivative of nicotine, showing recent exposure to tobacco / smoke).
Table 11.2: Content of interview (including self-completion questionnaires) by age group (in years)
Theme 0-1 y 2-4 y 5-15 y 16-64 y 65+ y
General health, long-standing illness,
limiting long-standing illness
Diagnosed hypertension and diabetes
Breathing problems
Receipt of social care
Fruit and vegetable consumption
Smoking, e-cigarettes, etc.
Exposure to second-hand smoke
Drinking alcohol
Economic status and occupation
Educational attainment
National identity
Ethnic origin
Height and weight measurements
Consent to link data to health records
Source: HSE 2018 Quick Guide. https://bit.ly/375Mvad (6/6/2020, 16:44 hours).
195
Table 11.3: Content of self-completion booklets by age group (in years)
Theme 8-12 y 13-15 y 16-17 y 18+ y
Smoking
E-cigarettes
Other nicotine delivery products
Exposure to second-hand smoke
Drinking alcohol
EQ-5D
158
(general health)
General Health Questionnaire
Gambling behaviour
Physical activity (IPAQ)
Sexual orientation
Religion
Source: HSE 2018 Quick Guide. https://bit.ly/375Mvad (6/6/2020, 16:44 hours).
Response rate and interview time
A household response rate of 59% was achieved, with an individual response rate of 54% of adults
and 60% of children in all eligible households. Average interview time for an adult was 40 minutes,
while nurse visit for adults who took part in all measurements averaged 32 minutes. When children
were interviewed without adults (aged 8 to 15), average interview time was 14 minutes and nurse
interview was 16 minutes.
159
158
EQ-5D is a widely known and used instrument in population health surveys. It is a standardised instrument for the
measurement of a person’s health status and has 2 parts – a descriptive system and a visual analogue scale (EQ VAS).
The descriptive system consists of 5 dimensions – mobility, self-care, usual activities, pain or discomfort, and anxiety or
depression. In the original version, each dimension has 3 severity levels – having no problems, some problems or severe
problems. To increase sensitivity to changes in health, a new version with 5 levels was developed (HSE 2018: Methods).
https://bit.ly/3eVxFWp (6/6/2020, 15:55 hours).
159
https://bit.ly/2Ubhf4r and https://bit.ly/375Mvad (6/6/2020, 15:53 hours).
196
Health Survey Northern Ireland (HSNI)
The HSNI is a Department of Health survey that runs every year on a continuous basis. The survey
covers a range of health topics that are important to the lives of people in Northern Ireland. It has
been conducted since 2010/11 with separate modules for different policy areas in different years.
HSNI is the single source of a range of population-based health and social care data, and is used
to inform a wide range of Department of Health’s strategies and indicators. Outside the department,
key users include the Public Health Agency, Health and Social Care trusts, other departments and
Arm’s Length Bodies, local government and the voluntary sector. Additionally, its statistics are of
interest to the local media, academics and the general public. It is sometimes used to compliment
the administrative data sources, and does not capture the same information collected elsewhere.
Department data needs are established on an annual basis during the questionnaire development
phase. This involves a series of meetings to enable policymakers to share their requirements.
Steps are taken to maximise usefulness of the data whilst mindful of the burden on respondents.
The department participates in the survey control process whereby surveys are assessed in terms
of the burden they place on respondents vis-à-vis time taken to complete the survey. Additionally,
any new survey which is proposed is considered within the context of the health survey and other
existing surveys, in attempt to reduce duplication and increase the re-use of existing data sources.
HSNI follows ONS guidance on harmonized standards for social surveys; where possible, includes
questions that have been agreed as standard in other UK countries (for e.g. on physical activity).
Further, recognized scales / instruments are used to allow comparisons to be made more readily.
Nevertheless, though steps are taken to encourage standardization and commonality in approach
across surveys, differences in sampling, weighting, etc. make it broadly rather than fully comparable.
The 2018/19 HSNI included questions on general health, mental health and well-being, antibiotics,
obesity, smoking, drinking alcohol and sexual health. It had a sample size of 3,593 individuals aged
16 and above. It had a systematic random sample of addresses from the Northern Ireland Statistics
and Research Agency (NISRA) Address Register (NAR). The NAR is developed within NISRA, and
is primarily based on the Land and Property Services (LPS) POINTER database. A total of 6,240
addresses were selected for interview. From an eligible sample of 5,448 addresses, 53% or 2,866
households participated. In each household, everyone aged 16 or over was selected to participate.
Measurements of height and weight were sought from individuals aged 2 and over in participating
households – data was obtained from 501 children (2-15 years) and 2,723 adults (16 and above).
The survey first results report and trend tables are published within a year of fieldwork completion,
most typically within 7-8 months. The date is preannounced on the department’s statistical release
calendar. In the majority of cases, the target publication date is met.
160
160
https://bit.ly/3cHW4gG and https://bit.ly/2UhSK5B (6/6/2020, 18:25 hours).
197
Scottish Health Survey (SHeS)
The SHeS has been carried out annually since 2008 – earlier, it was carried out in 1995, 1998 and
2003. The 2018 survey was the 14th in the series. Commissioned by Scottish Government Health
Directorates, it provides regular information on aspects of public health and related factors which
cannot be obtained from other sources. The SHeS series was designed to –
1) estimate the prevalence of particular health conditions in Scotland;
2) estimate the prevalence of certain risk factors associated with these health conditions, and to
document the pattern of related health behaviours;
3) look at differences between regions and sub-groups of population in the extent of their having
these particular health conditions or risk factors, and to make comparisons with other national
statistics for Scotland and England;
4) monitor trends in the population’s health over time;
5) make a major contribution to monitoring progress towards health targets.
Each survey in the series includes a set of core questions and measurements (height and weight,
blood pressure, waist circumference and saliva sample) and modules on specific health conditions
and risk factors that vary from year to year. Each year, the sample is augmented by an additional
boosted sample for children. Since 2008, NHS Health Boards
161
(HB) also have the opportunity to
boost the number of adult interviews.
Cardiovascular diseases (CVDs) and related risk factors remain the principal focus of the survey.
The main components of CVD are ischemic heart disease (IHD) or coronary heart disease (CHD)
and stroke, both of which are clinical priorities for NHS Scotland. Many of the key behavioural risk
factors for CVDs are of special interest to health policymakers and NHS on their own. For example,
smoking, poor diet, lack of physical activity, obesity and problematic alcohol use are all the subject
of specific strategies at improving health in Scotland. SHeS has detailed measures on all of them.
Topics covered in the 2018 to 2021 surveys were agreed following a consultation carried out in
2017. Many of the topics and questions included in earlier years of the survey were included again
to continue the time series. The 2018 survey included the same rotating topics as 2016 and 2014
surveys, although in 2018, a number of questions were removed or made less frequent to shorten
the survey and reduce the respondents’ burden. Some questions were added to the 2018 survey;
nevertheless, like earlier surveys, it continued to have a focus on CVDs and associated risk factors.
161
NHS Scotland consists of 14 regional NHS Boards – responsible for protection and improvement of their population’s
health and the delivery of frontline health care services – 7 Special NHS Boards and 1 public health body which support
regional NHS Boards through a range of specialist and national services. https://bit.ly/2UeMXxu (6/6/2020, 20:07 hours).
198
Table 11.4: Contents of the 2018 SHeS survey
CORE SAMPLE – Main interview outline
Version A Version B
Household questionnaire including household composition
General health (0+ years) including caring (4+)
Respiratory symptoms 16+ -
General CVD (16+)
Use of health services (0+)
Asthma (0+)
Asthma additional 16+ -
Physical activity adults (16+) and children (2-15)
Sedentary activity adults (16+) and children (2-15)
Additional physical activity questions 2+ -
Eating habits adults 16+ -
Eating habits children (2-15)
Fruit and vegetable consumption (2+)
Smoking and drinking (16+) [16-19 through self-completion]
Dental health (16+)
Economic activity (16+)
Education (16+)
Ethnic background, religion and country of birth (0+)
Self-completion (13+ and parents of 4-12 year olds)
Height and weight (2+)
Data linkage and follow-up research consent (0+)
- Biological module (16+)
Source: SHeS 2018: Volume 2 – Technical Report. https://bit.ly/2Y6Lomx (6/6/2020, 20:46 hours).
199
Main interview
Information was collected at both the household and individual level. Table 11.4 summarizes the
content of individual interviews – the topics covered for a participant depended on their age and
the sample type to which their address had been allocated. Version A households accounted for
67% of the main (core) sample – their questionnaire included the core questions and the questions
included in Version A rotating module. Version B households accounted for the remaining 33% of
the main (core) sample – only core questions were asked during main interview, with participating
adults also eligible to complete the biological measures module.
Sample design
The survey is designed to yield a representative sample of the general population living in private
households in Scotland every year.
Since 2008, the sample has been designed to be representative of adults – and since 2012 of the
population – at the HB level and since 2018 for adults at Local Authority (LA) level as well, following
4 years of data collection. HBs and LAs with sufficiently large sample sizes may be able to analyze
their data with fewer years of data collection.
In 2018, the SHeS design was coordinated with the designs of the Scottish Household Survey and
the Scottish Crime and Justice Survey as part of a survey efficiency project to allow the samples
of the 3 surveys to be pooled for further analysis.
In 2018, a random sample of 6,080 addresses was selected from the Postcode Address File (PAF),
using a multi-stage stratified design. If an address was found to have multiple dwelling units, 1 was
randomly selected; if multiple households were found at a dwelling unit, 1 was randomly selected.
Each person in a selected household was eligible for inclusion. If there were more than 2 children
in a household, 2 were randomly selected to limit burden on households. Individuals interviewed
at these addresses formed the main sample. At each selected household in the main sample, all
adults and a maximum of 2 children were eligible for interview. Two further samples were selected
for the 2018 survey – a child boost sample (5,448 addresses) in which up to 2 household children
were eligible to be interviewed, but adults were not, and a HB boost sample (224 addresses) for
HBs that opted to boost the number of adults interviewed.
Data collection
A letter stating the purpose of visit was sent to each sampled address in advance of the interviewer
visit. Interviewers sought the permission of each eligible adult in the household to be interviewed,
and both parents’ and children’s’ permission to interview up to 2 children aged 0-15 years.
200
Interviewing was conducted using a combination of CAI and self-completed paper questionnaires.
Adults and children aged 13-15 completed interviews themselves. Parents of children aged 0-12
completed the interview on behalf of their child.
Those aged 13 and over were also asked to complete a short paper self-completion questionnaire
on more sensitive topics during the interview. Parents of children aged 4-12 selected for interview
were also asked to fill in a self-completion booklet about the child’s strengths / difficulties designed
to detect behavioural, emotional and relationship difficulties.
Towards the end of the interview height and weight measurements are taken from those aged 2
and above. In a household sub-sample, interviewers seek permission from adults to participate in
an additional biological module. Participants are asked questions by specially trained interviewers
about prescribed medication, anxiety, depression, self-harm and suicidal attempts. In addition, the
interviewer takes participants’ blood pressure, saliva sample and measures weight. Data from the
biological module is reported every second year to allow 2 years of survey data to be combined.
Response rates
In 2018, interviews were held in 3,899 households with 1,980 children (949 as part of main sample
and 1,031 as part of the child boost sample) and 4,810 adults. Of these, 1,204 adults completed
the biological module.
For the combined main and HB boost sample, 57% of all eligible households responded, with all
individual interviews complete at 45% of households. For the child boost sample, around 3/4ths
of the households were ineligible as they did not contain any children. In eligible households, 64%
responded, with all individual interviews complete at 64% of households.
Adult response rate was 47% for men, 53% for women, 50% total. There was a further differential.
For both men and women, younger age-groups had a lower response rate (51% for men and 71%
for women aged 16-24 years) vis-à-vis the elderly (92+% for men and 95+% for women over 65).
Response rates were highest among children aged under 11 years (93-99% for boys and 95-99%
for girls), while the response rate for children aged 11-15 years was slightly lower at 92% for boys
and 91% for girls.
162
National Survey for Wales (NSW)
NSW is conducted by the Welsh Government (WG), succeeding 5 surveys earlier commissioned
by WG and 3 of its sponsored bodies – including the Welsh Health Survey, which was discontinued
in 2015. Before the survey was launched in 2016, a large-scale pilot survey and a small-scale field
162
Scottish Health Survey 2018: Vols. 1 and 2. https://bit.ly/3dLnxzC and https://bit.ly/2Y6Lomx (6/6/2020, 21:39 hours).
201
test on the final questionnaire were conducted. The survey is designed to be representative of all
adults aged 16 and above, in private households in Wales. The annual sample is, therefore, set up
as a stratified, single-stage random selection of addresses across Wales.
In its third round (2018-19), carried out by ONS, a sample of 24,762 addresses was initially drawn
from the Royal Mail Small Users Postcode Address File. The sample was stratified by 22 Welsh
LAs, with survey effort approximately proportional to the LA population size and oversampling to
ensure a minimum effective sample size of 250 in smaller LAs and 750 in Powys. It involved 11,922
interviews – the number of interviews achieved in each LA ranging from 312 in Isle of Anglesey
to 1,057 in Cardiff and 1,394 in Powys. Respondents received a £10 shopping voucher as a ‘thank
you’ for taking part in the survey, which was sent to them after the interview.
Face-to-face interviews were conducted using portable computers for 73 main modules to cover
the entire range of topics specified by WG and its sponsored bodies. Key themes included –
➢ Health
➢ Childcare and child education
➢ Climate change and environmental action
➢ Visits to the outdoors, participation in arts events and sports activities
➢ Use of / satisfaction with public services
➢ Material deprivation and income
➢ Well-being and loneliness
➢ Use of / attitudes towards the use of the Welsh language
➢ Internet access and use
➢ Tax devolution
Health and social care questions were focused on the following themes –
➢ BMI
➢ Diet
➢ Alcohol
➢ Smoking and e-cigarettes
➢ Physical activity
➢ Child screen time
➢ Mental well-being
➢ Hearing impairment
➢ Eye care
➢ Use of / satisfaction with GP and hospital services
➢ Use of / satisfaction with out of hours GP services
➢ Overall satisfaction with health services
➢ Views on social care services
202
The survey questionnaire and supporting materials were available as standard in both Welsh and
English (pilot and mainstage surveys), with interviews also offered in other languages with the help
of interpreters, if required. Furthermore, for some topics, sub-sampling was used to collect robust
estimates at the national level as efficiently as possible.
Questionnaire
The survey design is based on a single questionnaire administered to one randomly selected adult
aged 16 and above in each sampled household. The questionnaire content was developed by WG
and the 3 sponsored bodies. Annual questionnaire review and stakeholder consultation is carried
out by WG to determine new policy needs and the questionnaire content for the new survey year.
Further, a feedback questionnaire is sent to interviewers to get their views on how well the survey
process worked in the field, whether there were any specific issues with the questionnaire content
or flow and whether there is room for improvement. WG had considered the interviewer feedback
during the development of the 2018-19 questionnaire. A small scale pilot survey was conducted
by ONS in January 2018 to test the questionnaire and fieldwork processes.
In 2016-17, questions for the survey were largely taken from the 5 predecessor surveys, with some
questions also taken from other large-scale surveys. Changes for consecutive years include small
updates to individual questions which were continued from year to year, discontinuing or pausing
certain questionnaire modules that do not need to be asked every year as well as introducing new
questionnaire modules and individual questions.
Response rate and interview time
The planned response rate for 2018-19 was 56%, based the previous year’s achievement and the
additional measure put in place to increase response. The final response rate at the national level
was 54.2%. The number of interviews achieved was at / above target only in 9 LAs.
ONS closely monitored the progress of survey response performance over the course of fieldwork
period and applied very strict performance management measures to ensure that targets are met.
Performance at the start of the survey year was a little volatile across LAs, potentially influenced
by short-term capacity issues (e.g. sick leave, annual leave, mentoring for new interviewers) being
addressed inconsistently. However, performance became stronger and more consistent from the
second quarter onwards.
However, despite 73 main modules, median interview length was around 46.6 minutes, with 50%
of interviews lasting between 35 and 60 minutes.
163
163
https://bit.ly/2Yc7ZOu and https://bit.ly/2AbUoPp (6/6/2020, 22:27 hours).
203
Lessons for India
® The UK Statistics Authority (UKSA) is ‘an independent body at arm’s length from government’.
The Statistics Wing of MoSPI – the National Statistical Office (NSO) – should also be the same.
Independent, high-quality data has a critical role to play in a democracy, and statistical offices
should be answerable to the public at least as much as they are to elected governments. The
NSO, like the UKSA, should, first of all, aim at informing the public about social and economic
status and development, and then at assisting the development and evaluation of public policy
and programs as well as statistical coordination and regulation. One could also argue that the
independence and quality of a country’s statistical system is a core indicator of the strength of
its governance and democracy. In all the 3 countries included in this study, statistical systems
are focused on / answerable not just to governments, but also citizens. In India, no need seems
to be felt to make data user-friendly and accessible to citizens, and even researchers have to
struggle to access it and make sense of it. Data collection and dissemination in general, in the
sphere of health in particular, should function with a democratic ethos, be citizen-oriented and
-friendly. Only then will mechanisms such as community-based monitoring of health services,
as envisaged under schemes like the NHM, be realistically possible.
® The CBHI (MoHFW) should try to model its annual National Health Profile reports on the lines
of NHS Digital’s ‘Compendium of Population Health Indicators’, with a wide range of indicators
(and not just numbers) to provide a comprehensive overview of population health and care in
the country in a citizen-friendly manner so that it can be used by ordinary citizens to hold their
elected governments accountable.
® The UK statistical system suffers from several challenges similar to India’s – for e.g. duplication
of data collection, fragmented / incoherent / inconsistent / data-rich, information-poor statistics
(without much analysis to help citizens or governments) and dissemination delays. India could
learn from the remedial initiatives of the English Health Statistics Steering Group (EHSSG) and
even partner with it in addressing its own challenges.
® Given the NHS, a number of surveys in the UK support the evaluation of health care utilization,
quality and satisfaction. Certain health surveys in India, including the NFHS, provide general
evidence regarding the utilization of certain aspects of public and private health care, but there
is very limited scheme-specific evidence from their side to directly evaluate their performance.
Health surveys should gather more policy- and scheme-specific as well as general evidence
(both quantitative and qualitative), not just on health care utilization, but also on the quality of
health services and, most importantly, people’s satisfaction and expectations to enable action
towards making health systems in the country increasingly citizen-centric and -accountable.
® Health surveys in UK as well as in the other countries studied here also collect information on
mental health and well-being of the population. Only one round of the National Mental Health
204
Survey in India has been conducted in India so far, with no further round being planned (until
our meeting in October 2019 with NIMHANS, Bengaluru, which conducted the first round). If
not a dedicated survey, India should incorporate questions on mental health and well-being in
existing / proposed health surveys, especially given widespread coronavirus-instigated anxiety
and distress. A young population, ridden with personal and professional identity crisis, makes
a focus on mental health and well-being critical even more generally. ‘National Study of Health
and Wellbeing – Children and Young People’ survey is an example to consider.
® India needs to consider having a longitudinal / panel / cohort health survey, even if with a very
limited sample size, to be able to track changes in health and health care services over time
in an in-depth manner. It has LASI focused on health as well as other themes for adult / ageing
population, but one is needed focused on health for a broader age-group. It can be occasional
in its periodicity like HALS in the UK.
® Once again, we see that most health surveys are either of annual / biennial periodicity. SRS is
the only health survey in India with annual periodicity. However, it offers rudimentary birth and
death data only. All health surveys in India should be of preferably annual or maximum biennial
periodicity, with data dissemination within a maximum of 6 months. Smart sample size should
be considered, which is representative at the national and state / UT level. States / UTs should
conduct their own health surveys which are representative at state / district / sub-district levels,
with similar periodicity and dissemination timeframe. However, there should be consistency in
survey definitions, methodologies, etc. to ensure comparability, even if broadly and selectively.
® All the 3 countries studied here have multiple health surveys. An integrated health survey – or
for that matter, an integrated general survey – is only possible in small geographies like Wales.
Indian UTs could consider the Welsh survey model and have one integrated survey – but with
annual periodicity – which focuses on health as well as other themes. This would also help in
studying the social determinants of health (SDH). In fact, the inclusion of health themes in non-
health specific surveys should be encouraged to have SDH evidence. However, the demands
of comparability / interoperability need to be met.
® Surveys should have computer-assisted field investigator interviews as well as self-completion
instruments for sensitive questions. Pictorial computer-assisted instruments can be developed
for the less literate / illiterate populations. Work done on obtaining informed consent for clinical
trials for such population groups could be referred to for this purpose.
® Likewise, children should be directly interviewed, wherever possible, to know from them about
them. Their parents / guardians can still respond to more complex questions related to them.
® Disease-related questions / tests in surveys should be designed such that they can be mapped
to ICD-10. The Health Survey for England (HSE) does that to some degree. Mapping of survey
205
questions with WHO Family of International Classifications (FIC) should be ideally considered.
At the moment, only SRS-CoD survey does ICD-10 mapping.
® UK’s COVID-19 Infection Survey framework should be considered by India. US and Canadian
surveys appear more desirable since they focus on the social and economic impact of COVID-
19 as well, which is important given its widespread impact. Nevertheless, the UK has adopted
a robust, long-term testing strategy as part of its survey, which should be considered by India,
given that COVID-19 seems to be here to stay. At least, its health and wider impacts would be.
® The fieldwork model followed by the NFHS, involving a range of different private agencies for
different rounds, should be completely done away with. There should be a dedicated unit and
teams for conducting health – as well as other – surveys. Where external agencies have to be
involved, there should, first of all, be sufficient internal regulatory / monitoring capacities within
the parent organization and only agencies (including research organizations / universities) that
can deliver high-quality data should be involved. A UK-like system of ONS, NatCen, ScotCen,
etc. should be developed in India, with NSO in the lead from a statistical perspective and DoHR
from the domain perspective. There are huge conflicts of interest involved in the way the NFHS
is organized at the moment – neither should DoHFW, which manages MoHFW’s schemes, be
coordinating the survey nor should an external agency (IIPS) be in the lead, involving several
private agencies, whose primary motive is profit rather than data rigor and quality. Dedicated
surveyors need to be trained and employed – temporary arrangements should be disbanded.
206
12. Conclusions
Let us conclude the study with key recommendations and a table which characterizes the potential
respective features and themes which the 6 national health surveys that we propose could cover.
® The Government of India (GoI) should adopt a definition of health,
164
which can guide the design
and assessment of all health-related activities, particularly its health information system (HIS).
® In line with India’s health transition, health data collection should also shift from a demographic
to a predominantly health orientation – according due importance to the population dimension.
At the same time, we need to ensure that the emergent health orientation is not exclusively /
predominantly biomedical, and is sufficiently focused on the broader determinants of health.
® GoI should develop a National Health Data Policy (NHDP) and a National Health Data Advisory
Committee (NHDAC) with members from relevant ministries / departments of central and state
/ UT governments (health, statistics, planning); national organizations (ICMR, IIPS, NIMHANS,
ICSSR, etc.); multilateral agencies (UNSC, WHO, UNDP, UNFPA, UNICEF, World Bank etc.);
leading international health statistical agencies like the NCHS (US), Statistics Canada and NHS
Digital (UK); national / international health experts; industry and civil society representatives.
® The NHDAC should develop a health systems framework and health-related goals, targets and
indicators with timelines like SDGs – a National Reference List (NRL) of core health indicators,
like the WHO’s GRL, which is periodically revised to incorporate emerging concerns. For every
indicator, there should be a rationale, standardized definition, numerator, denominator, method
of measurement and estimation, disaggregation, frequency, preferred and other data sources,
baseline value, etc. State / UT governments should, likewise, develop SHDPs, SHDACs and
SRLs. NRLs and SRLs should guide interoperable data collection through a variety of sources.
® The NRL / SRL should be developed vis-à-vis core indicators of national / state health policies
and programs, international data reporting requirements (including health-related SDGs) and
WHO’s Family of International Classifications (WHO-FIC). HIS should be revised accordingly.
® HIS comprises a variety of data sources, and they are all required for monitoring various health
system components, with a preferred respective role for each. In India, since surveys are seen
as compensating for the weakness of administrative sources, there are high expectations from
them. All data sources should be strengthened in order to have rational expectations from each.
164
India is a signatory to WHO’s constitution, and it could be argued that it affirms the definition of health enshrined in it.
207
® Health surveys should focus on monitoring the vision / goals / objectives of health policies and
programs to periodically ensure that they are being fulfilled. Program MIS / other mechanisms
(ground assessments by DGHS Regional Offices, PRCs, review missions, local communities,
etc.) should be strengthened for regular program monitoring and evaluation. Surveys should
not be expected to help in MIS data validation beyond a few core indicators; indicator definitions,
population coverage, etc. should be harmonized in such cases.
® The division of labor between the 4 relevant ministries could be the following. MHA looks after
all population-related indicators through census (decennial), CRVS (continuous) and the SRS
(annual enumeration-cum-survey) – as it already does. MoHFW should look after public health
surveillance – as it already does. However, it should be Department of Health Research (DoHR)
in MoHFW, rather than its Department of Health and Family Welfare (DoHFW) – as is presently
the case – which should lead and coordinate all public health surveillance activities, with the
exception of policy- and program-based MIS. DoHFW, being the operational wing of MoHFW,
should manage various MIS in an integrated, consistent and coordinated manner. The MHA and
MoHFW could collaborate for a cause of death survey, given that it requires domain expertise
which the MHA lacks. DoHR / ICMR institutions should be involved in this case. MoSPI should
oversee all health surveys in consultation with MoHFW and MHA. All statistical activities should
strictly be conducted under its statistical guidance, coordination and supervision – and, in the
case of health, under the domain-related guidance of DoHR (MoHFW). For this, MoSPI needs
to be independent, both from political interference as well as from the IAS-led bureaucracy.
® There is a serious conflict of interest that the DoHFW, which manages various health schemes,
also commissions and manages the independent NFHS, conducted by an agency (IIPS) which
itself ‘is under the administrative control’ of the MoHFW. Not only this, the MoHFW’s Statistics
Department, which manages HMIS, also manages the NFHS. These are very serious conflicts
of interest which should be addressed immediately. If need be, the DoHR should be renamed
as the Department of Health Research and Surveillance (DoHRS) – research and surveillance
go hand in hand – and all health surveillance activities, including surveys, should be carried
out under its domain supervision and MoSPI’s statistical supervision. Budgets and staff in both
these organizations should be enhanced accordingly. The DoHR / ICMR already has a network
of leading centers across the country, which could be leveraged for this purpose. However, it
should go beyond its biomedical – and adopt a much more broad-based – approach to health.
® While data collection is important, analysis is also part of data generation hemisphere, followed
by interpretation and response under the data use hemisphere (figure 1.5). All four data-related
frameworks need to be strengthened at the central / state / local-most levels – it cannot be the
exclusive prerogative of researchers / statisticians on sidelines (DES, NIHFW / SIHFWs, PRCs,
etc.) or at the top (ICMR, IHME, etc.) to analyze / interpret data. Central / state / local capacities
208
need to be strengthened and IT tools leveraged for the entire data life-cycle. In fact, those who
collect data at local levels can sometimes contextualize and contextually analyze it better than
those who do not know / understand the local context in which the data was collected. This is
also in keeping with the spirit of decentralization inherent in the conceptualization of the NHM.
® There has to be a clearly defined framework for data collection, processing, synthesis, analysis
and use for the design and assessment of policies and programs as well as course-correction.
In the absence of such a framework – despite humongous data collection and ‘reporting’ within
the system as part of MIS and accountability of various functionaries – data ‘use’ for the design
and assessment of policies and programs as well as course-correction is not seen as important
and becomes an arbitrary / whimsical activity. Statisticians and IT can provide the tools, but it
is eventually the domain officials who have to use the data from a policy / program perspective.
This is seriously missing across the country – from the national to the local levels, including in
states like Kerala (field interactions).
® Ease of data use should be facilitated for policymakers as well as other stakeholders. This is a
huge challenge at the moment. The STATcompiler customization tool of DHS surveys and the
visualization hub of GBD data with causes of death are two excellent examples.
® Not just survey schedules, but fact sheets, at least, should also be prepared in local languages.
They should be made available as well as painted on the walls of SCs, PHCs, CHCs and district
hospitals in local languages. This can also be done for key indicators from non-survey sources.
This would not only help in the democratization of official data, but also enhance accountability.
® A mixed methods approach should be adopted to health surveys in the country. For guidance,
we could refer to health surveys conducted in India pre- and post-independence and The DHS
Program, of which the NFHS is a part, for instance. The richness of the notion and experiences
of health, well-being and health care utilization and satisfaction can best be captured by means
of qualitative methods. Qualitative information could also imbue the otherwise dry quantitative
data with an intimate, human sense and help in making health systems more people-oriented.
® The DHS Program also has several types of surveys and not just the standard DHS, according
to which the NFHS has been modelled. Beyond the standard DHS surveys – with large sample
sizes, typically conducted every 5 years, to allow comparisons over time – ‘interim DHS’ focus
on select indicators, are conducted between standard DHS, have shorter questionnaires and
sample sizes, but are nationally representative. There is also ‘continuous DHS’, as part of which
data is collected and reported annually by a permanent DHS office and field staff. There are
‘in-depth DHS’ and ‘mini DHS’ too. India should adopt a dynamic approach to health surveys,
and consider the various options available in the light of its requirements.
209
Proposed surveys
® All the 3 countries studied here have multiple health surveys. An integrated health survey – or
for that matter, an integrated general survey – is only possible in small geographies like Wales.
Indian UTs could consider the Welsh survey model and have one integrated survey – but with
annual periodicity – which focuses on health as well as other themes. This would also help in
studying the social determinants of health (SDH). In fact, the inclusion of health themes in non-
health specific surveys should be encouraged to have SDH evidence. However, the demands
of comparability / interoperability need to be met.
® Table 12.1 below outlines a health survey strategy with 6 surveys and their major characteristics
that we recommend GoI should adopt at the national level. A similar strategy can be developed
for the states / UTs for representative data at the state, district and subdistrict levels. However,
there should be consistency in survey definitions, methodologies, etc. to ensure comparability,
even if broadly / selectively, between surveys conducted at the national and state / UT levels.
® Once the challenge of COVID-19 diminishes, the CIS could be dropped. However, the template
of an emergency survey which reflects the health as well as its broader socioeconomic impact
should be kept for the future. The CIS will also enable us to monitor potential future pandemics
in a better way. With CIS / like being an emergency survey, we have a total of 5 regular surveys.
® The IHS would be the comprehensive health survey providing an overview of the health of the
nation. The IHMS would be an in-depth survey, with a smaller sample size, aimed at collecting
clinical information from physical and biomedical tests and measurements – like the NHANES
in the US and CHMS in Canada. Such tests and measurements should not be appended to an
existing survey (as has been done in the case of NFHS) since testing and measurements need
to be done much more professionally and with much more caution than the general field work
investigators can possibly ensure. The MCHS would be exclusively focused on maternal and
child health in a broader – unlike the typical traditional RCH – framework. SRS will continue to
be the same, but with added features and professional dissemination practices of the NFHS.
It will be the only survey source of vital statistics. RGI should give up the cause of death survey,
for which a more specialized agency like ICMR is better suited. It can support it with its death
statistics from the CRVS, MCCD, SRS, etc. The ICMR should enhance the sample size for the
CoD survey to yield representative data at the national and state / UT levels. The frequency of
this survey should be annual for better respondent recall.
® The proposed surveys cover major data points which health surveys should provide, including
the core indicators of health policies to monitor progress on them. Many of the core indicators
of health policies overlap with those of health programs, hence we have not included the latter
210
here. In any case, the table below is more illustrative, to provide a potential blueprint of a health
survey strategy at the national level. The proposed NHDAC / SHDACs can work out the details.
® India should consider designing flexible surveys like the Canadian CCHS – Annual component
with fixed, variable and rapid content components – taking into cognizance the continuing and
changing needs of various health sector stakeholders. Neither health programs nor statistical
instruments have this sort of flexibility, despite India being such a diverse and federal country.
® With an array of health surveys – and states doing their own set of surveys – there is no need
to have bloated sample sizes for a few surveys. Survey sampling needs to be smarter in order
to achieve representative data in the most focused, equitable and economically efficient style.
® Health surveys in US, Canada and UK are either continuous or of annual / biennial periodicity.
India should adopt a mixed strategy – with continuous surveys or annual / biennial periodicity
(see table 12.1 below for details).
® Data linkage is one of the biggest lessons for India from the Canadian context to avoid enormous
duplicity of data collection and resultant wastage of precious and limited resources. However,
as in Canada, citizens’ consent / confidentiality should be paramount – no data linkage should
happen without their explicit consent and with utmost care to keep it confidential.
® Surveys should have computer-assisted field investigator interviews as well as self-completion
instruments for sensitive questions. Pictorial computer-assisted instruments can be developed
for the less literate / illiterate populations. Work done on obtaining informed consent for clinical
trials for such population groups could be referred to for this purpose.
® Likewise, children should be directly interviewed, wherever possible, to know from them about
them. Their parents / guardians can still respond to more complex questions related to them.
® The fieldwork model followed by the NFHS, involving a range of different private agencies for
different rounds, should be completely done away with. There should be a dedicated unit and
teams for conducting health – as well as other – surveys. Where external agencies have to be
involved, there should, first of all, be sufficient internal regulatory / monitoring capacities within
the parent organization and only agencies (including research organizations / universities) that
can deliver high-quality data should be involved. A UK-like system of ONS, NatCen, ScotCen,
etc. should be developed in India, with NSO in the lead from a statistical perspective and DoHR
from the domain perspective. Dedicated surveyors need to be trained and employed –
temporary arrangements should be disbanded. 211
Table 12.1: Proposed national health surveys and themes
Survey characteristics
India Health
Survey (IHS)
India Health
Measures
Survey (IHMS)
– IHS sample
Maternal and
Child Health
Survey (MCHS)
– IHS sample
Sample
Registration
System (SRS)
Cause of Death
Survey (CDS) –
SRS sample
COVID-19
Impact Survey
(CIS) – IHS-
SRS samples
Design and logistics
Conducting / coordinating agency NSO ICMR IIPS ORGI DoHR ICMR, NSO
Target age-groups (in years) 6+ 6+
0-5 children,
15-49 women
0+ 0+ 15+
Representativeness (data pooling for lower level representativeness) National, states / UTs National
Research design
Mixed
methods
Quantitative Mixed methods Quantitative Mixed methods
Mixed
methods
Survey design
Cross-
sectional
Panel Cross-sectional Panel Cross-sectional Panel
Oversampling Vulnerable and under-served populations
Survey mode (CAPI: Computer-assisted personal-interviewing;
CASI: … self-interviewing; CATI: … telephone-interviewing)
CAPI, CASI
CAPI, physical
examinations
CAPI, CASI CAPI CAPI
CAPI, CATI
(follow-up)
Periodicity Biennial Annual Biennial Annual Annual Continuous
Duration – Interview 1 hour 45 minutes 1 hour 30 minutes 1 hour 15-30 minutes
Duration – Data collection (in months) 12 6 12 6 12 Continuous
Duration – Data dissemination (in months, following data collection) 6 3 6 3 6 Quarterly
Duration – Final report (in months, following data dissemination) 6 3 6 3 6 Biannual
212
Survey characteristics
India Health
Survey (IHS)
India Health
Measures
Survey (IHMS)
– IHS sample
Maternal and
Child Health
Survey (MCHS)
– IHS sample
Sample
Registration
System (SRS)
Cause of Death
Survey (CDS) –
SRS sample
COVID-19
Impact Survey
(CIS) – IHS-
SRS samples
Local area characteristics (local area questionnaire – to be filled in by the area survey team based on documentation, observation and interactions)
Type of area (rural / urban)
Physical infrastructure (including from a physical activity perspective)
Social infrastructure (health care, public health, educational facilities)
Environmental characteristics (pollution, WASH, etc.)
Economic and employment characteristics
Health facility characteristics (health facility questionnaire – to be filled in by the area survey team based on documentation, observation and interactions)
WHO health system building blocks (availability and quality)
Household characteristics
Local area score (based on local area characteristics)
IHS scores can be used by all surveys
Health facility score (based on health facility characteristics)
Housing characteristics (cooking fuel, electricity, drinking water,
sanitation, ventilation, number of rooms, etc.)
Household economic status (including asset ownership)
Social characteristics of the household (caste / tribe, religion, etc.)
Household composition (relation, age and gender)
213
Survey characteristics
India Health
Survey (IHS)
India Health
Measures
Survey (IHMS)
– IHS sample
Maternal and
Child Health
Survey (MCHS)
– IHS sample
Sample
Registration
System (SRS)
Cause of Death
Survey (CDS) –
SRS sample
COVID-19
Impact Survey
(CIS) – IHS-
SRS samples
Respondent characteristics
Demographic characteristics
Both of living
and deceased
Both of living
and deceased
Civil registration (births and deaths, Aadhar / PAN / BPL card, etc.)
Educational characteristics
Employment characteristics
Health practices and behaviors (sanitation, physical activity, dietary
patterns, salt intake, tobacco and alcohol consumption, etc.)
Exposure to mass media
Respondent’s health status
NHP 2017 – ‘Attainment of the highest possible level of health and well-being for all at all ages’
General health and well-being – physical, mental and social
165
(self-
reported) – incl. musculoskeletal, sense organ diseases, disabilities
Disease, disability, risk factor prevalence (measurements / tests)
166
COVID-19 prevalence (symptoms and tests)
Maternal and child health and nutrition
165
Health as defined in WHO’s constitution; social health and well-being as defined at https://www.nhp.gov.in/social-health_pg (9/6/2020, 17:35 hours).
166
List of health measurements and tests included in the NHANES (US) – https://bit.ly/2XMsKS8 – and CHMS (Canada) – https://bit.ly/2MKjOq0 (9/6/2020, 17:42 hours) – surveys.
214
Survey characteristics
India Health
Survey (IHS)
India Health
Measures
Survey (IHMS)
– IHS sample
Maternal and
Child Health
Survey (MCHS)
– IHS sample
Sample
Registration
System (SRS)
Cause of Death
Survey (CDS) –
SRS sample
COVID-19
Impact Survey
(CIS) – IHS-
SRS samples
Cognitive and emotional health
Adverse drug reaction (ADR)
Fertility
Mortality
Injuries (transportation, falls, poisoning, self-harm and violence, etc.)
Respondent’s views on health system performance (public and private)
NHP 2017 – ‘Expand preventive, promotive, curative, palliative and rehabilitative services provided through the public health sector with focus on quality’
Coverage of essential health services (SDG 3.8.1 tracer indicators –
‘service capacity and access’ indicators to be covered under health
facility questionnaire above)
Financial protection when using health services (SDG 3.8.2)
Accessibility / affordability / utilization of safe, efficacious and quality
preventive, promotive, curative, palliative and rehabilitative services
Accessibility / affordability / utilization of comprehensive primary
health care (CPHC), including Health and Wellness Centres (HWCs)
Access / affordability / utilization of secondary and tertiary health
care and linkages with CPHC
Accessibility / affordability / utilization of various systems of medicine
215
Survey characteristics
India Health
Survey (IHS)
India Health
Measures
Survey (IHMS)
– IHS sample
Maternal and
Child Health
Survey (MCHS)
– IHS sample
Sample
Registration
System (SRS)
Cause of Death
Survey (CDS) –
SRS sample
COVID-19
Impact Survey
(CIS) – IHS-
SRS samples
Maternal and child immunization
Access to / utilization of health insurance and public health schemes
Access to free drug and diagnostic services through public facilities
Screening and management of noncommunicable diseases (NCDs)
– facility- and community-based (including interlinkages / referrals)
Patient safety and system to report adverse drug reactions (ADRs)
Performance and attitudes of health care and public health staff
Awareness and attitudes towards antimicrobial resistance (AMR)
and antimicrobial use (AMU), self-medication, AMU in agriculture
IEC for small family norm
IEC for healthy choices
Trust in public and private health care facilities
Interface of public and private health care facilities
Health care satisfaction and expectation
Child care, training and soft skill development
Out-of-pocket health care expenditure (especially catastrophic)
Socioeconomic impact of poor health
216
Survey characteristics
India Health
Survey (IHS)
India Health
Measures
Survey (IHMS)
– IHS sample
Maternal and
Child Health
Survey (MCHS)
– IHS sample
Sample
Registration
System (SRS)
Cause of Death
Survey (CDS) –
SRS sample
COVID-19
Impact Survey
(CIS) – IHS-
SRS samples
Specific outcomes / outcome indicators (MoHFW’s policies)
NHP 2017 (general), NPP 2000, National Multisectoral Action Plan for Prevention and Control of Common NCDs, 2017-22 (NMAP-NCDs), SDG 3 (wherever specified)
Life expectancy at birth
Disability-adjusted life years (DALYs) (together from IHS, IHMS, SRS)
Total fertility rate (TFR)
Maternal and child mortality (stillbirth, neonatal, infant under-5, MMR)
Prevalence of stunting among under-5 children
Premature mortality from cardiovascular diseases, cancer, diabetes
or chronic respiratory diseases – probability of dying between ages
30-70 from these 4 diseases and cancer incidence by type of cancer
(NMAP-NCDs)
Disease prevalence / incidence (HIV/AIDS; leprosy, kala-azar and
lymphatic filariasis in endemic pockets; tuberculosis; blindness) –
preventable morbidity, disability and mortality due to NCDs (NMAP-
NCDs); malaria, neglected tropical diseases, hepatitis, water-borne,
other communicable diseases (SDG 3.3)
Risk factor prevalence / incidence (blood pressure, blood sugar,
tobacco use, etc.) – alcohol consumption, obesity, physical activity,
salt intake, cooking fuels, fruit and vegetable consumption (NMAP-
NCDs); substance abuse, including narcotic drug abuse (SDG 3.5)
Occupational injury among agricultural workers – deaths and injuries
from road traffic accidents (SDG 3.6)
217
Survey characteristics
India Health
Survey (IHS)
India Health
Measures
Survey (IHMS)
– IHS sample
Maternal and
Child Health
Survey (MCHS)
– IHS sample
Sample
Registration
System (SRS)
Cause of Death
Survey (CDS) –
SRS sample
COVID-19
Impact Survey
(CIS) – IHS-
SRS samples
National Vaccine Policy 2011
Morbidity and mortality due to vaccine-preventable diseases (VPDs)
Childhood infectious diseases of public health importance
Impact of existing interventions
National Mental Health Policy of India 2014
Distress, disability, exclusion, morbidity and premature mortality
associated with mental health
Prevalence and impact of risk factors related to mental health
Risk and incidence of suicide and attempted suicide
Respect for rights and protection from harm of person(s) with mental
health problems
Stigma related to mental health problems
Other outcome indicators
Fertility indicators (including birth rates and sex ratios)
Mortality indicators (including death rates)
Cause-specific morbidity and disability (WHO-FIC)
Cause-specific mortality (WHO-FIC)
218
Survey characteristics
India Health
Survey (IHS)
India Health
Measures
Survey (IHMS)
– IHS sample
Maternal and
Child Health
Survey (MCHS)
– IHS sample
Sample
Registration
System (SRS)
Cause of Death
Survey (CDS) –
SRS sample
COVID-19
Impact Survey
(CIS) – IHS-
SRS samples
Specific processes / process indicators (MoHFW policies)
NHP 2017 (general), NPP 2000 (specified)
Antenatal care coverage
Skilled attendance at birth – institutional deliveries by trained
persons (NPP 2000)
Newborns fully immunized by one year of age – universal
immunization of children against all vaccine preventable diseases
(NPP 2000)
Met need of family planning – unmet needs for basic RCH services,
supplies and infrastructure as well as access to information /
counseling and services for fertility regulation and contraception
with a wide basket of choices (NPP 2000)
Registration of births, deaths, marriage and pregnancy (NPP 2000)
Prevention and control of communicable diseases (NPP 2000)
Hypertensives and diabetics maintain ‘controlled disease status’
Integrated Indian systems of medicine (ISM) for reproductive and
child health (RCH) services and household outreach (NPP 2000)
NMAP-NCDs
Adults receiving drug therapy and counselling to prevent heart
attacks and strokes
219
Survey characteristics
India Health
Survey (IHS)
India Health
Measures
Survey (IHMS)
– IHS sample
Maternal and
Child Health
Survey (MCHS)
– IHS sample
Sample
Registration
System (SRS)
Cause of Death
Survey (CDS) –
SRS sample
COVID-19
Impact Survey
(CIS) – IHS-
SRS samples
Availability and affordability of quality, safe and efficacious essential
NCD medicines including generics and basic technologies in both
public and private facilities
Access to palliative care assessed by morphine-equivalent
consumption of strong opioid analgesics per death from cancer
Vaccination coverage against hepatitis B virus monitored by number
of third doses of Hep-B vaccine administered to infants
Women aged 30-49 screened for cervical cancer at least once
Women aged 30 and above screened for breast cancer by clinical
examination by trained health professional at least once
High risk persons (using tobacco, smoking and smokeless and betel
nut) screened for oral cancer by examination of oral cavity
Capacity of individuals, families and communities to make healthier
choices by creating healthy environments that promote health and
reduce the risk of NCDs
Accessible and affordable good quality care to people with disease
or risk factors through primary health care approach
National Mental Health Policy of India 2014
Enhanced understanding of mental health
Universal access to quality health and social care for mental health
(including prevention services, treatment, care and support
services) throughout the lifespan within a rights-based framework
220
Survey characteristics
India Health
Survey (IHS)
India Health
Measures
Survey (IHMS)
– IHS sample
Maternal and
Child Health
Survey (MCHS)
– IHS sample
Sample
Registration
System (SRS)
Cause of Death
Survey (CDS) –
SRS sample
COVID-19
Impact Survey
(CIS) – IHS-
SRS samples
Access to mental health services for vulnerable groups, including
homeless persons, persons in remote areas, difficult terrains,
educationally / socially / economically deprived sections
Enhanced availability and equitable distribution of skilled human
resources for mental health
Enhanced financial allocation and improve utliisation for mental
health promotion and care
Identify and address the social, biological and psychological
determinants of mental health problems
Electronic Health Record (EHR) Standards for India 2016
Promote interoperability and where necessary be specific about
certain content exchange and vocabulary standards to establish a
path forward toward semantic interoperability
Source: Developed by author. 221
Annexure A – List of interviewees
Sn. Name Designation Division / Department / Ministry / Organization Government of
NEW DELHI
1 Dr Nivedita Gupta + Team Chief Director (Statistics)
Statistics Division, Department of Health and Family
Welfare (DoHFW), Ministry of Health and Family
Welfare (MoHFW)
India
2 Ms Anjali Rawat Director (HMIS) Statistics Division, DoHFW, MoHFW India
3 Mr Birendra Kumar Mishra Deputy Director
Central Bureau of Health Intelligence (CBHI),
Directorate General of Health Services (DGHS),
DoHFW, MoHFW
India
4 Ms Anu Nagar + 1 Joint Secretary Department of Health Research (DoHR), MoHFW India
5 Dr D K Shukla Consultant; Former Director In-charge
National Institute of Medical Statistics (NIMS), Indian
Council of Medical Research (ICMR), DoHR, MoHFW
India
6 Dr Harpreet Singh Scientist 'E' & Head
Informatics, Systems & Research Management Cell,
ICMR, DoHR, MoHFW
India
7 Dr Ashoo Grover Scientist 'F' & Head Research Methodology Cell, ICMR, DoHR, MoHFW India
8 Ms Sandhya Singh Deputy Registrar General
Vital Statistics Division, Office of the Registrar
General of India, Ministry of Home Affairs
India
9 Dr Ashutosh Ojha + Team Deputy Director General
Social Statistics Division, Ministry of Statistics and
Programme Implementation
India
10 Ms Shruti Pandey Assistant Director (P&E) Planning and Evaluation Division, Ministry of AYUSH India
11 Dr Harshad P Thakur + 1 Director
National Institute of Health and Family Welfare
(NIHFW), MoHFW
India
222
Sn. Name Designation Division / Department / Ministry / Organization Government of
12 Dr Ved Prakash Yadav National Consultant (Health Systems) WHO (Country Office)
13 Mr Jorge Coarasa + 1 Program Leader (Human Development) The World Bank (India)
14 Mr Luigi D' Aquino Chief of Health UNICEF (India Country Office)
15 Mr Venkatesh Srinivasan + 1 Assistant Representative UNFPA (Country Office)
16 Ms Suneeta Krishnan + 1 Country Lead (Measurement, Learning & Evaluation)
Bill and Melinda Gates Foundation (India Country
Office)
17 Ms Moutushi Sengupta Director MacArthur Foundation (India)
18 Prof Lalit Dandona Distinguished Research Professor Public Health Foundation of India (PHFI)
19 Dr Bhaswati Das Associate Professor
Centre for the Study of Regional Development,
School of Social Sciences, Jawaharlal Nehru
University
UDAIPUR
20 Dr B L Nagda Former Joint Director
Population Research Centre (PRC) Udaipur (Mohanlal
Sukhadia University)
India
21 Dr Pooran Mal Yadav Additional Charge PRC Udaipur India
22 Dr Julfikar Kazi Joint Director (Udaipur Zone)
Department of Medical, Health and Family Welfare
(DoMHFW)
Rajasthan
23 Dr G S Rao District Program Manager (Udaipur) DoMHFW Rajasthan
24 Mr Pratap Singh District Monitoring and Evaluation Officer (Udaipur) DoMHFW Rajasthan
223
Sn. Name Designation Division / Department / Ministry / Organization Government of
25 Mr Punit Sharma Deputy Director (Udaipur) Directorate of Economics and Statistics (DES) Rajasthan
JAIPUR
26 Dr Deepak Saxena + Team Senior Regional Director
Regional Office of Health and Family Welfare
(RoHFW), DGHS, DoHFW, MoHFW
India
27 Ms Seema Mishra + 1 Deputy Director CBHI, DGHS, DoHFW, MoHFW India
28 Shri Rohit Kumar Singh + 1 Additional Chief Secretary DoMHFW Rajasthan
29 Shri Naresh Kumar Thakral Special Secretary & Mission Director (NHM) DoMHFW Rajasthan
30 Dr R S Chhipi Director (Family Welfare / FW) DoMHFW Rajasthan
31 Dr K K Sharma
Director (Public Health) and Commissioner (Food and
Safety)
DoMHFW Rajasthan
32 Mr Sujan Kumar Saha Assistant State Programme Manager
State Programme Management Unit (SPMU, NHM),
DoMHFW
Rajasthan
33 Mr Raushan Kumar Jha State Data Manager SPMU (NHM), DoMHFW Rajasthan
34 Mr Vikas Meena + 1 Demographer (FW) DoMHFW Rajasthan
35 Dr Ramesh Chandra Gupta Project Director
National AYUSH Mission, Department of Ayurved and
Indian Medicine (DoAIM)
Rajasthan
36 Dr Renu Bansal + 2 Director Homeopathy Chikitsa Vibhag, DoAIM Rajasthan
37 Dr Joga Ram District Collector (Jaipur) Rajasthan
38 Dr Mamta Chauhan + 1 Associate Professor State Institute of Health and Family Welfare (SIHFW) Rajasthan
224
Sn. Name Designation Division / Department / Ministry / Organization Government of
39 Dr Shiv Dutt Gupta + 1 Chairman
Indian Institute of Health Management Research
(IIHMR)
40 Dr Arindam Das Associate Professor IIHMR
41 Dr Ruchit Nagar + 1 CEO Khushi Baby
LUCKNOW
42 Dr Krishan Kumar Mittra + 1 Senior Regional Director RoHFW, DGHS, DoHFW, MoHFW India
43 Prof Arvind Mohan Director PRC Lucknow (University of Lucknow) India
44 Ms Jasjit Kaur + 3 Additional Mission Director & Additional Executive Director
NHM & State Innovations in Family Planning Services
Project Agency (SIFPSA)
Uttar Pradesh
45 Dr Vikasendu Agarwal Joint Director (IDSP) Directorate of Medical and Health Services Uttar Pradesh
46 Mr Arvind Kumar Pandey Director DES Uttar Pradesh
47 Dr Narendra Agarwal + 1 Chief Medical Officer (Lucknow) DoMHFW Uttar Pradesh
48 Mr Satish Kumar District Program Manager (Lucknow) DoMHFW Uttar Pradesh
49 Dr Mahesh Nath Singh Assistant Professor SIHFW Uttar Pradesh
50 Prof Nomita Kumar + 1 Assistant Professor Giri Institute of Development Studies (GIDS)
51 Mr Waseef Naqvi + 1 Senior Research Analyst Academy of Management Studies (AMS)
225
Sn. Name Designation Division / Department / Ministry / Organization Government of
PATNA
52 Shri Manoj Kumar Mission Director NHM Bihar
53 Mr Ranjan Kumar Assistant Director (HMIS and MCTS) NHM Bihar
54 Dr Tabrez Akhter Lari State Programme Officer Bihar AYUSH Society Bihar
55 Mr Banshidhar Mishra Joint Director DES Bihar
56 Dr Hemant Shah Chief of Party (Bihar Technical Support Program) CARE India
57 Dr Sanchita Mahapatra Epidemiologist
Centre for Health Policy (CHP), Asian Development
Research Institute (ADRI)
GUWAHATI
58 Dr Parthajyoti Gogoi Regional Director RoHFW, DGHS, DoHFW, MoHFW India
59 Ms Mallika Medhi Director Directorate of Health Services (Family Welfare) Assam
60 Dr Lakshmanan S Mission Director NHM Assam
61 Mr Rahul Dev Chakraborty State MIS Manager NHM Assam
62 Dr Jyotirmoy Choudhury Consultant Directorate of AYUSH Assam
63 Dr R M Dubey Professor and Head Centre for Sustainable Development Goals (CSDG) Assam
64 Dr Madhulika Jonathan Chief UNICEF India (Guwahati Field Office)
226
Sn. Name Designation Division / Department / Ministry / Organization Government of
65 Dr Ashoke Roy Director Rural Resource Centre for North Eastern States
66 Dr Joydeep Borua Associate Professor O K D Institute of Social Change and Development
PUNE
67 Dr V L Gokak Senior Regional Director RoHFW, DGHS, DoHFW, MoHFW India
68 Dr Madhuri Thakar Scientist 'F'
Immunology and Serology (I&S), National AIDS
Research Institute (NARI), ICMR
India
69 Dr Ashwini Shete Scientist 'D' I&S, NARI, ICMR India
70 Dr Vini Sivanandan + Team Assistant Professor
PRC Pune (Gokhale Institute of Politics and
Economics / GIPE)
India
71 Dr Nitin Bilolikar + Team Deputy Director of Health Services (Pune Region) Public Health Department (PHD) Maharashtra
72 Dr Pradip Awate State Surveillance Officer (IDSP) PHD Maharashtra
73 Mrs P P Telkhade District Statistical Officer (Pune) DES Maharashtra
74 Dr Anjali Radkar Professor GIPE
THIRUVANANTHAPURAM
75 Dr Ali Manikfan Abdullage + 1 Senior Regional Director
Regional Office of Health and Family Welfare,
Directorate General of Health Services
India
76 Team PRC Thiruvananthapuram (University of Kerala) India
77 Dr Rathan U Kelkar Mission Director / Secretary
NHM / Department of Agriculture Development and
Farmers' Welfare
Kerala
227
Sn. Name Designation Division / Department / Ministry / Organization Government of
78 Dr Sreehari M State Nodal Officer (Child Health and RBSK) NHM Kerala
79 Dr Raju V R + 1 Additional Director (FW, Planning and e-health) Directorate of Health Services (DHS) Kerala
80 Mr Preeth State Data Officer / Demographer Health Information Cell, DHS Kerala
81 Mr V Ramachandran + Team Director DES Kerala
82 Dr K S Shinu + 3 In Charge / Executive Director
Kerala SIHFW / State Health Systems Resource
Centre (SHSRC)
Kerala
83 Dr Preetha + Team District Medical Officer (Thiruvananthapuram) DHS Kerala
84 Mr Anish Kumar B Deputy Director (Thiruvananthapuram) DES Kerala
85 Dr Sankara Sarma P Professor and Head
Achutha Menon Centre for Health Science Studies,
Sree Chitra Tirunal Institute for Medical Sciences and
Technology
86 Prof Irudaya Rajan + 1 Professor Centre for Development Studies (CDS)
87 Prof K R Thankappan Professor
Department of Public Health and Community
Medicine, School of Medicine and Public Health,
Central University of Kerala
OTHERS
88 Prof K S James + NFHS Team Director and Senior Professor
International Institute for Population Sciences (IIPS) –
Mumbai (email interaction)
India
89 Dr Prashant Mathur Director
National Centre for Disease Informatics and Research
(NCDIR), ICMR, DoHR, MoHFW – Bengaluru
India
90 Dr B N Gangadhar + Team Director
National Institute of Mental Health and Neuro-
Sciences (NIMHANS) – Bengaluru
India