National Family Health Survey (NFHS): Key Data for Health and Population Studies

When policymakers in Delhi want to know whether more babies are being born in hospitals, whether teenage girls in Bihar are anaemic, or whether women in Kerala have a say in household decisions, they don’t guess. They turn to the National Family Health Survey (NFHS). For over three decades, this massive household survey has been the closest thing India has to a health check-up for the entire nation, and its findings shape everything from village-level immunisation drives to crore-rupee central schemes.

What is the NFHS, really? 🔗

The NFHS is a large-scale, multi-round household survey that collects data on population, health, nutrition, and family welfare from a representative sample of households across the country. It is conducted under the Ministry of Health and Family Welfare (MoHFW), with the International Institute for Population Sciences (IIPS), Mumbai serving as the nodal agency. Technical assistance has historically come from the USAID-supported Demographic and Health Surveys (DHS) Program through ICF.

In simple terms, the NFHS is India’s chapter of a global series of DHS surveys that began in 1984 to track population and health trends in developing countries. Whether you are reading a newspaper headline about declining fertility or a research paper on anaemia, chances are the numbers came from the NFHS.

From NFHS-1 to NFHS-5: how the survey has grown up 🔗

The first round of the NFHS was launched in 1992-93, and the survey has been repeated roughly every six to seven years since. Each round has built on the previous one, adding new topics and expanding its scope to reflect changing public health priorities.

NFHS-1 (1992-93): laying the foundation 🔗

NFHS-1 was conducted across all states except Sikkim and covered around 88,000 households. Its primary aim was to provide reliable baseline estimates of fertility, mortality, family planning practices, and maternal and child health. For the first time, India had a standardised, internationally comparable dataset on these indicators at both national and state levels.

NFHS-2 (1998-99): adding women’s voices 🔗

NFHS-2 covered all 26 states at the time and expanded the questionnaire significantly. According to the World Bank’s microdata library, this round added information on reproductive health, women’s autonomy, domestic violence, women’s and children’s nutrition, anaemia, and salt iodization. This was an important shift because it treated women not just as bearers of children but as decision-makers and individuals with their own health concerns.

NFHS-3 (2005-06): men enter the picture 🔗

NFHS-3 maintained continuity with previous rounds while adding community-based HIV testing, which was a major step given the rising concerns about HIV/AIDS at the time. Crucially, it also included a separate men’s interview for the first time, recognising that men’s behaviours, attitudes, and health are central to family welfare outcomes.

NFHS-4 (2015-16): going district-level 🔗

NFHS-4 was a giant leap in scale. The sample size was expanded nearly fivefold compared to NFHS-3 so that, for the first time, the survey could produce district-level estimates in addition to state and national figures. It also introduced Clinical, Anthropometric, and Biochemical (CAB) testing, which meant field teams were not just asking questions but actually measuring height, weight, blood pressure, and haemoglobin levels.

NFHS-5 (2019-21): in the shadow of a pandemic 🔗

NFHS-5 covered 707 districts and over six lakh households. Conducted in two phases, with fieldwork interrupted by the COVID-19 pandemic, the round introduced new indicators on preschool education, disability, access to toilet facilities, death registration, menstrual hygiene practices, and methods and reasons for abortion. The scope of CAB testing was also expanded to include waist and hip circumference, along with a wider age range for blood pressure and blood glucose measurements.

NFHS-6, currently underway, is for the first time being coordinated entirely by IIPS, without ICF as a technical partner.

What the NFHS actually tells us 🔗

The NFHS is essentially a treasure trove of indicators on how people live, give birth, eat, fall sick, and seek care. Let’s look at a few headline findings from NFHS-5 to understand the kind of insights it generates.

Fertility has dipped below replacement level 🔗

One of the most discussed findings from NFHS-5 was that the Total Fertility Rate (TFR) declined from 2.2 in NFHS-4 to 2.0 children per woman, falling below the replacement level of 2.1. The urban TFR stood at 1.6 and rural at 2.1, with Uttar Pradesh, Bihar, Jharkhand, Madhya Pradesh, Meghalaya, and Manipur still above the replacement level. This single statistic has enormous implications for everything from school planning to old-age policy.

Infant and child mortality 🔗

The Infant Mortality Rate (IMR), which counts deaths of infants under one year per 1,000 live births, declined marginally between NFHS-4 and NFHS-5. Improvements in institutional deliveries, exclusive breastfeeding rates rising from 55% to 64%, and better vaccination coverage have all contributed. However, neonatal mortality and under-five mortality remain stubbornly high in several states, pointing to gaps in the first weeks of life when babies are most vulnerable.

It is important to note that the Maternal Mortality Ratio (MMR) is not directly estimated by NFHS due to sample size limitations for such a rare event. Instead, MMR is tracked through the Sample Registration System (SRS) of the Registrar General of India. However, NFHS provides crucial maternal health indicators that explain MMR trends, such as antenatal care visits, institutional deliveries, and skilled birth attendance.

Nutrition: a mixed bag 🔗

NFHS-5 showed modest improvements in child nutrition at the all-India level, with stunting declining from 38% to 36%, wasting from 21% to 19%, and underweight from 36% to 32%. But anaemia worsened, with more than half of all children and women, including pregnant women, found anaemic in multiple states. These numbers directly fuel programmes like Poshan Abhiyaan and the Anaemia Mukt Bharat strategy.

Gender, empowerment, and household dynamics 🔗

The NFHS also captures softer but equally important indicators such as women operating bank accounts, mobile phone ownership, exposure to spousal violence, and use of hygienic menstrual products. The share of women operating bank accounts rose from 53% to 79% between NFHS-4 and NFHS-5, reflecting the impact of financial inclusion drives like Jan Dhan Yojana.

How NFHS data shapes policy 🔗

The NFHS isn’t just an academic exercise. Its findings feed directly into the design, monitoring, and evaluation of major government schemes. Programmes like Janani Suraksha Yojana, Mission Indradhanush, Pradhan Mantri Surakshit Matritva Abhiyan, and the National Health Mission rely on NFHS data to identify priority districts, set targets, and measure progress.

NFHS data also helps in tracking India’s progress on Sustainable Development Goals (SDGs), particularly those related to good health, gender equality, and reduced inequalities. Researchers use it for academic studies, NGOs use it to design interventions, and journalists use it to hold the government accountable. A systematic review of NFHS-based studies on maternal health found over 130 peer-reviewed papers in the past three decades, showing how deeply embedded the survey is in health research.

The cracks in the data: limitations of NFHS 🔗

For all its strengths, the NFHS is not perfect. Anyone using its data, especially students and researchers, should be aware of its limitations.

The questionnaire is too long 🔗

Each NFHS interview can stretch for over an hour, sometimes longer. Critics argue that the questionnaire has grown so vast that respondent fatigue sets in, reducing the reliability of answers to questions asked towards the end. Some experts have flagged ethical concerns about asking so much from respondents without any incentive, and have suggested that the survey should be split into thematic modules instead.

Fieldwork quality varies 🔗

NFHS fieldwork is carried out by around 15 different field agencies across the country, and field investigators are often poorly trained, underpaid, and stretched thin. Training quality varies between agencies and between states, which means data quality can also vary. Interviewer bias and shortcuts taken in the field can introduce non-sampling errors that are hard to detect once the survey is over.

Methodological concerns 🔗

The National Statistical Commission has raised concerns about non-sampling errors creeping in due to the expansion of the survey from NFHS-4 onwards. Some specific concerns include unexplained fluctuations in immunisation coverage, the use of capillary blood samples for anaemia testing (which may overestimate prevalence compared to venous samples), and the reported sex ratio of 1,020 females per 1,000 males in NFHS-5, which sits at odds with other data sources like the Census and NSS.

Regional and contextual variations 🔗

Sample sizes at the district level, while a huge step forward, may still be too small for reliable estimates on rare events. Cultural context also matters: responses to sensitive questions on domestic violence, abortion, or sexual behaviour are influenced by what is considered socially acceptable to disclose in a given region. NFHS-5’s fieldwork during COVID-19 created additional comparability challenges between phase one and phase two states.

So how should you read NFHS data? 🔗

For students of population studies, public health, and the social sciences, the NFHS is an indispensable resource, but it should be used with caution. Always read the methodology section, check whether the indicator has been measured consistently across rounds, look for cross-validation with other sources like the Sample Registration System or the Census, and be especially careful with district-level estimates and sensitive indicators. The numbers tell a story, but the story is richer when you know where the data came from and how it was collected.

What do you think? If you were redesigning the NFHS to make it more reliable and respectful of respondents’ time, what is the one change you would prioritise, and why? And do you think district-level data is worth the trade-off in fieldwork quality and respondent fatigue?

References

1. https://www.nfhsiips.in/
2. https://www.dataforindia.com/nfhs-explainer/
3. https://microdata.worldbank.org/index.php/catalog/4482
4. https://www.nfhsiips.in/nfhsuser/aboutus.php
5. https://pmc.ncbi.nlm.nih.gov/articles/PMC10657051/
6. https://www.pib.gov.in/PressReleasePage.aspx?PRID=1847431
7. https://www.business-standard.com/article/current-affairs/india-s-fertility-rate-declines-to-2-1-show-nfhs-5-phase-2-findings-121112400981_1.html
8. https://www.ssph-journal.org/journals/public-health-reviews/articles/10.3389/phrs.2022.1604825/full
9. https://www.epw.in/journal/2024/35/national-family-health-survey-5/data-collection-and-quality-concerns-national.html
10. https://idronline.org/the-problems-with-the-national-family-health-survey-nfhs/