By MALAVIKA VYAWAHAREMAY 22, 2014
The New York TimesMAGADI, India — M. R. Gundappa, 60, died the way most Indians do: with no doctor present, no monitors beeping by his side and no written record. The only person present was his wife, Sushilamma, 48, who spent the day of his death trying to get him admitted to a government hospital where he could be treated for abdominal pain.
On a recent afternoon, Sushilamma spent an hour trying to retrieve her memories of the fateful day in 2013 for an official from the Office of the Registrar General of India, who sifted through her story for clues to what had caused her husband’s death.
Nearly 70 percent of deaths in India, five million in all each year, take place in the absence of medical supervision, according to the office, which is responsible for registering births and deaths.
To fill this gap, a new survey, the Million Death Study, is trying to turn the clock back on a million premature deaths that took place between 2001 and 2014, sifting through evidence provided by families and caregivers. By assigning causes to these deaths, based on the accounts of witnesses, the study hopes to identify the major causes of premature death in India.
“The idea is to show, with evidence, that many of these deaths are preventable,” said Prabhat Jha, a professor at the University of Toronto, who conceived of the project.
While full results are not expected for four to five years, some preliminary findings have been released, and those have stirred controversy. The survey’s estimate of total malaria deaths in India is more than 10 times the World Health Organization’s. Its figure for deaths related to H.I.V. infections is significantly lower than what the United Nations predicted, and the Indian government, which has spent heavily to control the spread of the disease, may take that into account as it settles on future medical priorities.
Approaching strangers to ask about the deaths of their loved ones is no simple endeavor in India, a fact that was brought home to Ashok Kumar, a registrar official, when he knocked on a door in the town of Magadi, in the southern state of Karnataka, on a recent day.
Seeing his credentials, Sushilamma opened the door, allowing him to settle into the cramped living room, under the watch of a pantheon of family gods perched on the pale walls. A garlanded photograph of her husband was a recent addition, and her eyes were drawn to the photograph as she discussed the details of his death with the visitor.
Then the neighbors came, to watch, to listen, to offer sympathy and sometimes to fill in some gaps when her account faltered.
As the interview stretched and Mr. Kumar went into the details of the death, her answers became vague as his questions became more pointed.
“Did he have a pain in his heart?” he asked.
“I don’t think so,” she replied.
“Are you sure? Was there any pain in this region?” he asked, gesturing to his chest.
No. No, she insisted.
“Try and remember,” he said.
“I really don’t remember,” she said in an apologetic tone, shaking her head, a weariness creeping in.
Mr. Kumar’s questioning was part of an unusual form of medical investigation, a verbal autopsy, that traces its roots to 17th-century London, a city haunted by epidemics, where “death searchers” regularly showed up at homes after someone died.
In India, two pioneering studies were conducted in the 1950s and ’60s, but were more limited in scope.
“We are literally chasing down death in millions of homes over a decade now,” said Suresh Rathi, a senior researcher who manages the project. “Where would we find the doctors for that?”
Twice a year, senior registrar officials like Mr. Kumar visit households across India that reported deaths in the previous six months, carrying verbal autopsy forms. Scans of the completed forms are shared with the Bangalore office of the Center for Global Health Research, an international nonprofit group that is collaborating with the registrar’s office on the project.
Each verbal autopsy form is sent to two doctors from a pool of about 300, who independently assign a probable cause of death. If their verdicts match, the cause of death is made final. Otherwise, a senior doctor is asked to arbitrate and make a decision.
This part is completed fairly quickly, Professor Jha said. The real challenge is getting the verbal autopsies.
In urban areas, Mr. Kumar said, “we may be welcomed into homes, but are not always welcome to discuss details of a loved one’s death.”
Village residents are more willing to share their stories, according to Professor Jha. “It is almost cathartic,” he said. “For many it is a chance to vent about the treatment or the lack of it at government and even private hospitals.”
Language poses an additional obstacle: Autopsy narratives are recorded in regional languages, and though the project employs doctors fluent in 18 of India’s 20 officially recognized languages, 2,000 verbal autopsies have had to be thrown out because they were incomprehensible.
Some 42,000 autopsies have been gathered and analyzed, each on a paper form. Because of the study’s reliance on paper, it could take anywhere from four to five years before the final results are available. That reservoir of information will be valuable to public health specialists, but will probably bring little to the families who were its subjects. The results will never be shared with them, and the deaths will remain unexplained.
Still, the researchers persist. On a hot afternoon in the village of Uragahalli, Mr. Kumar was trying to obtain an account of the death of a girl who had survived barely an hour after her birth.
The girl’s mother had lost consciousness after the delivery, and remembered very little. So Mr. Kumar relied on the baby’s aging grandparents, who struggled to put what happened into words.
“What’s the use talking about this now?” the grandmother asked in the local Kannada language, her irritation clear.
The interview with Sushilamma, the widow, was equally exhausting. The end of the interview had the quality of a bandage being slowly unwrapped from an open wound: She described many hours of pleading with doctors to tell her what was wrong with her husband.
“The worst part is I still don’t know what happened,” she said, her voice tinged with something more than grief.