Bush, Iraqis dismiss study on war deaths

CHICAGO — President Bush and Iraqi officials on Wednesday tried to discredit a new study suggesting that more than 600,000 Iraqis have been killed during the war in Iraq — a far higher death toll than previously estimated.

The report by researchers from Johns Hopkins University was published online Wednesday by The Lancet, a respected British medical publication.

Predictably, the study renewed concern about the human cost of the Iraqi war and the wisdom of pursuing it. Underscoring the report's political implications — the findings come just weeks before the November elections — Bush said at a news conference that "the methodology is pretty well discredited."

"I do know that a lot of innocent people have died and that troubles me, and it grieves me," Bush said. "And I applaud the Iraqis for their courage in the face of violence."

An Iraqi government spokesman, meanwhile, called the study "unbelievable" and said "these numbers are exaggerated."

Though some experts agreed the study had flaws, others said the results were probably a good, ballpark estimate.

"Our goal was not to produce the most precise number possible but rather to suggest the scope of war-related deaths," said lead author Gilbert Burnham, co-director of the Center for Refugee and Disaster Response at Johns Hopkins. "And what we show is that the numbers are very, very large."

The research method employed — known as "cluster sampling" — is broadly used in public health to estimate everything from the incidence of disease to the prevalence of death of war-torn areas such as Bosnia or Congo.

Researchers divided the country into 16 regions and selected specific geographic "clusters" in the regions for door-to-door surveys. Eight doctors traveled to these areas, asking 1,849 households how many members had died, when, and under what circumstances.

In 87 percent of interviews, doctors asked for a death certificate, and in 92 percent of those cases, one was provided.

Researchers then estimated the death rate in Iraq at 5.5 deaths per 1,000 people before the war and 13.3 per 1,000 afterward. Applying the latter figure to the Iraqi population of 26 million, they figured approximately 393,000 to 943,000 war-related "excess deaths" had occurred.

The middle of this range, about 655,000, translates into more than 16,000 fatalities a month since the war began in March 2003. The numbers include deaths of combatants and insurgents as well as civilians.

The latest Defense Department security assessment, released in September, put the number of deaths among Iraqi civilians and security forces at nearly 120 a day between May 20 and Aug. 11, up from 80 a day in the prior three months. In the public-health community, the Johns Hopkins report was greeted with a mixture of skepticism and admiration.

Mark van der Laan, professor of biostatistics at the University of California, Berkeley, said he wasn't sure he trusted the statistical adjustments the Johns Hopkins team performed to adjust for the small sample of households surveyed and varying levels of violence in the sample areas.

"Given the design of their study, it is likely [the data] should be taken with a large grain of salt," he said.

At the University of Chicago, Ronald Thisted also pointed out potential problems with the data analysis but said "there's pretty good evidence here that the number of deaths is in the hundreds of thousands." The range of estimated deaths was necessarily broad because of the small sample size, but still a significant finding, he said.