An ambitious survey currently underway in India will monitor the health of 1.3 million households, taking regular physical measurements--such as blood pressure--chronicling risk factors and identifying causes of mortality.
"By documenting how people die, you can actually improve public health programming," said Prabhat Jha of St. Michael's Hospital and the University of Toronto, the leader of this massive project.
Comprehensive health statistics can illuminate new diseases and quantify the consequences of behaviors like tobacco and alcohol use. In India, most deaths occur in the home, away from medical attention, where there is no one qualified to certify their cause.
Advertisement
"Routine statistics are what picked up lung cancer and AIDS," said Jha. "They picked up a big increase in lung cancer, and that led to work on smoking. They picked up an increase in immune deaths in young men, and that said there was a new infection going on, which was AIDS."
A literal cast of thousands will be supporting this project: There are 7,000 part-time staffers, who each month identify vital events —births, deaths or pregnancies— in their communities. There are 800 full-time field staffers, employed by the government, who conduct interviews, visiting each of the million households every six months. These investigators collect details about a family's health.
The information is input into a web-based system, keeping the identity of each household anonymous. One hundred doctors then look at the data and assign causes when a death has occured in a family. Each death is ascribed a cause by two doctors. If there is disagreement, a third doctor will be consulted to arbitrate. About 5 to 10% of the work is repeated randomly as a check of the overall system.
The current round of sampling began in 2002-2003. While comprehensive statistics regarding causes of death will not be available for another 6 to 8 months, the project has already demonstrated that the massive system works.
"We found that for causes of death before old age—before 70 plus—the sytem seems to generate good information on the causes of death," said Jha. "It produces results that are independently verifiable."
Reliable mortality measurements should help identify environmental and genetic factors that constitute risk factors for various diseases. Once risk factors are identified, public health organizations can work to stop preventable deaths.
It is Jha's hope that other developing countries will adopt the methods of the study in India.
"Developing countries can use this model, which is a very simple but scientifically robust way to improve their own health statistics," Jha said. "The amount of information they can gather is enormous."