Credit: Jasiek Krzysztofiak/nature

In 1975, when Prabhat Jha was growing up in Canada, his family received a report from India that his grandfather had died; the cause was unclear. Like many people living in rural India, Jha's grandfather had died at home, without having visited a hospital. Jha's mother was desperate for more information, so she returned to her home village to talk to locals. Years later, when Jha was at medical school, he reviewed his mother's notes and realized that his grandfather had probably died of a stroke. Now Jha, an epidemiologist at the University of Toronto, is nearing the end of an ambitious public-health programme to document death in India using similar 'verbal autopsy' strategies.

Google supercomputers tackle giant drug-interaction data crunch Twenty tips to help you interpret scientific claims How oral sex can cause throat cancer by transmitting HPV

The Million Death Study (MDS) involves biannual in-person surveys of more than 1 million households across India. The study covers the period from 1997 to the end of 2013, and will document roughly 1 million deaths. Jha and his colleagues have coded about 450,000 so far, and have deciphered several compelling trends that are starting to lead to policy changes, such as stronger warning labels on tobacco.

Public-health experts need mortality figures to monitor disease and assess interventions, but quality mortality data are scarce in most developing countries. Seventy-five per cent of the 60 million people who die each year around the globe are in low- and middle-income countries such as India, where cause of death is often misclassified or unreported. Groups such as the World Health Organization (WHO) typically base mortality estimates on hospital data, but in many developing countries most people die outside hospitals.

As global health researchers increasingly turn to indirect computer models, many applaud the MDS's low-tech, on-the-ground approach and see it as a model for assessing true health burdens in the developing world. “For countries like India, there will almost certainly continue to be a role for verbal autopsy,” said Colin Mathers, coordinator of mortality and burden of disease at the WHO. “It's a crucial source of information.”

Credit: SOURCE: P. Jha et al. PLoS Med. 3, e18 (2006)
Credit: Source: Prabhat Jha
Credit: Source: Prabhat Jha
Credit: SOURCE: P. Jha et al. PLoS Med. 3, e18 (2006); R. Dikshit et al. Lancet 379, 1807–1816 (2012); N. Dhingra et al. Lancet 376, 1768–1774 (2010); V. Patel et al. Lancet 379, 2343–2351 (2012); M. Hsiao et al. Brit. Med. J. Open 3, e002621 (2013); P. Jha et al. Brit. Med. J. 340, c621 (2010); B. Mohapatra et al. PLoS Negl. Trop. Dis. 5, e1018 (2011).
Credit: SOURCE: P. Jha et al. PLoS Med. 3, e18 (2006); R. Dikshit et al. Lancet 379, 1807–1816 (2012); N. Dhingra et al. Lancet 376, 1768–1774 (2010); V. Patel et al. Lancet 379, 2343–2351 (2012); M. Hsiao et al. Brit. Med. J. Open 3, e002621 (2013); P. Jha et al. Brit. Med. J. 340, c621 (2010); B. Mohapatra et al. PLoS Negl. Trop. Dis. 5, e1018 (2011).
Credit: SOURCE: P. Jha et al. PLoS Med. 3, e18 (2006); R. Dikshit et al. Lancet 379, 1807–1816 (2012); N. Dhingra et al. Lancet 376, 1768–1774 (2010); V. Patel et al. Lancet 379, 2343–2351 (2012); M. Hsiao et al. Brit. Med. J. Open 3, e002621 (2013); P. Jha et al. Brit. Med. J. 340, c621 (2010); B. Mohapatra et al. PLoS Negl. Trop. Dis. 5, e1018 (2011).
Credit: