Decoding the Million Death Study

The lack of reliable, cause-specific mortality statistics is considered a major obstacle to the improvement of public health in many low- and middle-income countries. Researchers and government officials in India have set up the Million Death Study to address this situation. First, how the study produces quantitative estimates of the burden of mortality in India is explored by collecting symptomatic data, using that data for diagnostic purposes, and aggregating those diagnoses into an overview of mortality in India. Second, the limitations of the perspective on public health based on discrete and specific diseases that result from this approach are addressed. Numbers alone cannot solve the public health issues India faces, rather cognitive justice towards a broader range of perspectives on major public health problems is required to develop effective political interventions.

Research for this paper was funded by the Marie Skłodowska-Curie Actions of the European Commission, grant number 272996. The author would like to thank all interview respondents for their time and insight, and Ashawari Chaudhuri, Poonam Pandey and Ambuj Sagar for their fieldwork support.

“How do you count the dead?” This question, posed by Science in June 2012 (Vogel 2012), appears to be straightforward enough. Nevertheless, many researchers and administrators consider it to be one of the most important and challenging questions in the context of public health in low- and middle-income countries (Jha 2012). Economics Nobel laureate Angus Deaton, for example, emphasises the issue in The Great Escape, where he argues that “not the least of the health problems faced by the poor countries of the world today is the lack of good information on the number of people who die, let alone on what causes their deaths” (Deaton 2013: 16). Many public health professionals point to the absence of consistent and comprehensive registration of causes of death as one of the main obstacles to obtaining this “good information.” They have therefore established various initiatives to produce more accurate estimates of death and its causes around the world (Jha 2014). Prominent among these is the Million Death Study (MDS) in India. Health scientists and epidemiologists consider the production of reliable mortality statistics particularly challenging in India, due to its large and diverse population, the variety in disease burdens within the population and the general lack of insight into what this burden consists of (India State-Level Disease Burden Initiative Collaborators 2017). The MDS aims to address these challenges by gathering cause of death data outside of the health system, relying on India’s census infrastructure and household interviews instead—a method called verbal autopsy (VA; see Jha et al 2006). At the same time, the study explicitly seeks to affect the health system by providing evidence for evaluating and improving public health programmes and policy.

In this paper, I explore how the joint objectives of producing knowledge on mortality and improving public health are woven together in the study. I demonstrate how this relation is not self-evident by discussing researchers’ ambivalence towards the tensions between scientific accuracy and the political significance of the numbers they produce. I conclude that recognising this ambivalence is informative for how to think about the limitations of numerical evidence in the politics of public health.