The number of people dying from malaria in India may be more than 10 times higher than World Health Organization figures suggest, a new study says.
But WHO has disputed the findings.
In the study, international experts examined 122,000 deaths from 2001 to 2003 in 6,671 parts of India thought to be representative of the entire country. Trained workers interviewed family members or friends about the deaths, asking questions about things like the symptoms people had before they died and if they got any treatment. They found 90 percent of people who died were in rural areas and 86 percent were not treated in any health facility.
Those reports were then e-mailed to two of 130 physicians who determined whether or not the person had died from malaria. Researchers found among people aged one month to 70 years, 3.6 percent of them probably died from malaria.
Based on their findings, experts guessed from 125,000 to 277,000 people die of malaria in India every year. Official numbers from the World Health Organization put that figure at about 15,000. The study was published online Thursday in the journal Lancet. It was paid for by the U.S. National Institutes of Health, the Canadian Institute of Health Research and the Li Ka Shing Knowledge Institute.
"When I first saw the results, I thought, 'This can't be right,'" said Prabhat Jha, director of the Centre for Global Health Research at the University of Toronto, one of the study authors. "But the more we looked, the more the pattern seemed to fit," he said.
Jha and colleagues also found a geographical link between the deaths the doctors attributed to malaria and government-reported malaria deaths, patterns of the disease's seasonal spread, among other indicators.
Jha said past estimates of malaria deaths have mostly been done in treated patients in hospitals or health clinics, so experts are unlikely to find many deaths. He said adult deaths have been seriously underestimated, which could change global strategies to fight the disease.
Robert Newman, director of the agency's Global Malaria Programme, said WHO had serious doubts about the Lancet study and questioned the validity of diagnosing malaria based on the recollections of family members or friends. He said studies by WHO and partners showed as few as four percent of deaths attributed to malaria using this method were actually caused by malaria.
"While routine reports of malaria cases and deaths in India are certainly incomplete, the new estimate of 200,000 deaths appears too high," Newman said. "The findings of this study cannot be accepted without further validation."
Malaria has long been a scourge in India, where problems such as crowded living conditions and stagnant pools of water allow the mosquito-borne disease to thrive. It is curable with drugs, if caught early, but can be fatal without treatment, particularly in children.
Others called the Lancet study results startling.
"Evidence is increasing that the scale of (India's malaria) burden has been greatly underestimated," wrote Bob Snow, a professor at KEMRI-University of Oxford and the Wellcome Trust and co-authors in an accompanying commentary.
Snow and colleagues said the study emphasized the inadequacies in how WHO reports malaria cases, which often overlook people not admitted to hospitals. They said similar problems could exist in Burma, Bangladesh, Pakistan, Afghanistan and Indonesia.
Richard Tren, a director of Africa Fighting Malaria, an Africa- and U.S.-based advocacy group, doubted the numbers discrepancy would apply to Africa. He said the bigger problem was tracking how malaria money is spent and that authorities often don't ensure that bed nets and medicines actually make it to the people who need them.
Jha said he hoped the study would prompt Indian and global health officials to reconsider malaria's impact, particularly on adults.
"In India, they take cricket fever pretty seriously," he said. "So now I hope they'll take malaria fever seriously too."