Nine Lives of India’s National Family Health Survey

When the US Agency for International Development (USAID) launched its Demographic and Health Survey (DHS) program in the 1980s to study reproductive behavior in developing countries, India viewed it with skepticism. The fall of the Berlin Wall in 1989 changed Indo-US relations, paving the way for the country’s first DHS survey in 1992–93. But this decision sparked controversy. India’s statistical establishment questioned the need for a new demographic survey, noting that India has a long record in generating such data on its own. This raised fears that well-funded surveys would drive up survey costs in a ratchet effect for all organisations. The most intense criticism came from left-wing activists, who portrayed it as an “imperialist” tool to control Third World populations. The claim was not completely different from reality. The infamous Kissinger Report of 1974 was by then declassified, and it exposed why the US was eager to fund fertility research: fears that an ‘explosion’ in Third World numbers would lead to hordes of disgruntled young people would fall prey to communism and undermine US commercial interests globally.

Even then, neither India’s statistical establishment nor the Left had much influence over the policy, and despite their objections, the survey went ahead. But doubts remained. The health ministry demanded a sample re-verification by an independent team, which USAID refused to fund. In the end, the ministry used its own funds for this exercise and decided to publish the results only after that second survey. Still, academics were wary of using survey data.

Academic opinion changed somewhat when two eminent demographers, Praveen Visaria and S Irudaya Rajan, viewed the NFHS as a landmark survey in an economic and political weekly article published in 1999. He explained how it filled critical data gaps in the country and argued that few surveys in India have shown such concern for verification and quality control.

The second round, held in 1998–99, attracted more attention than the first, but the third round of 2005–06 eventually brought the NFHS to prominence. The results for the third round were released in 2007, when India was in the midst of an unprecedented growth boom. The stagnation in the child malnutrition rate reported by it seems to have left many people behind. It was a section of the Left and in the Right to Food movement that played a key role in drawing the nation’s attention to the startling findings.

The government was embarrassed by India’s attention to nutritional deficiencies, and voices within the administration called for the survey to be closed. But sensible voices prevailed and the fourth round of NFHS was pushed forward with the additional mandate of preparing district level estimates. The resulting expansion in sample size, however, casts doubt on the comparison of results with previous rounds. The NFHS’s Technical Advisory Group was not able to fully persuade the National Statistical Commission (NSC) on this, but in the interest of time, an agreement was reached. The fourth round finally took place after a gap of 10 years. The latest round was interrupted by the pandemic, but thankfully not derailed.

There are four important lessons in the checkered history of the NFHS. First, it underscores the importance of transparency. From its inception, anonymous sample data was available free of charge to all. The National Sample Survey (NSS) unit level data was uncovered only after the NFHS arrived on the scene, and was not available free of cost until recently. The openness of the NFHS made it an outlier in the Indian data ecosystem in its early years, and helped establish its credibility.

The second important lesson is related to the politics of statistics. How and when a certain dataset is to be prepared is often determined by the social or global elite. But how it will ultimately be used is often beyond their control, especially if such datasets follow open-access policies. Nutrition was not the primary focus area of ​​the original DHS programme, but it was the nutrition component of the NFHS that attracted attention in India.

This brings us to the third important lesson. A major reason why nutrition received the most attention until then was the lack of regular and reliable nationally representative data on nutrition. Trends in fertility and mortality were still available from other sources. To be sure, state governments compiled monthly data on nutritional outcomes through Integrated Child Development Services (ICDS), but no one believed those numbers. It was also widely acknowledged by ICDS officials that the ‘real time’ data generated through the program was biased and varied greatly in quality and coverage across different regions. As NFHS gained acceptance, more and more states started relying on it for nutrition planning. The big lesson here is that in many situations, there is no substitute for an independent well-designed survey.

In the end, the NFHS survived and flourished in this country despite the tribulations caused by public ownership. The growing demand for reliable data in the world’s largest democracy ensured its existence. If we are not to stray from statistics, we must continue to closely track our data ecosystem.