March 20, 2013


Jim Stubenrauch is a CHMP senior fellow.

It’s fitting that the 10th anniversary of the invasion of Iraq, a milestone passed this week, would occasion some soul-searching among those who supported the war as well as those who opposed it. How many of the war’s stated goals—other than the removal of Saddam Hussein—have been accomplished? And at what cost, in blood and treasure, to the people of Iraq and the nations comprising the coalition, the U.S. foremost among them? It’s debatable whether Americans are safer than we were in 2003; are the Iraqis who survived the war any better off?

It seems obvious that a clear understanding of how many Iraqi civilians were killed, wounded, and displaced in the conflict would be necessary to answer these questions, and it’s symptomatic of our current political predicament that these basic facts remain elusive. How can policymakers and the American people judge the success or failure of the mission if we haven’t taken accurate measure of the human cost of the war—or come to grips with the most accurate measures we have? And have mainstream news organizations done all they should in making the facts known?

When the last U.S. combat troops left Iraq in December 2011, even the New York Times and National Public Radio—two news organizations often charged with having a “liberal bias”—settled on “more than 100,000” when reporting the number of Iraqi civilians killed in the war (NPR’s All Things Considered repeated the same figure this past weekend). But several studies that used the best epidemiologic methods available had already reported much higher estimates.

The March 16 issue of the Lancet (subscription necessary) contains several important articles that aim “to crystallise Iraq’s current health situation, to clarify its most pressing health problems, and to offer a prognosis for the future health of the country.” A short article by Frederick Burkle, Jr. and Richard Garfield, “Civilian Mortality after the 2003 Invasion of Iraq,” performs a great service in untangling the knotty history of several highly contested mortality counts of the past decade. They show how the studies’ findings were resisted for political reasons and they compare the methods and estimates of the most important studies, including the ongoing Iraq Body Count Project (which uses passive surveillance methods, primarily media reports of killings) and the Lancet’s own controversial mortality surveys of 2004 and 2006 (which used active surveillance methods). (Burkle, now at the Harvard Humanitarian Initiative, was the first interim minister of public health in Iraq in 2003; Garfield, at Columbia University’s Schools of Nursing and Public Health, was a coauthor of the 2004 Iraq mortality survey.)

The two Lancet reports were based on random cross-sectional cluster sampling surveys, the standard epidemiologic approach to obtaining casualty estimates in conflict zones. The first, published in 2004, found that 100,000 Iraqis had already been killed in the war; the risk of violent death was 58 times higher than it was before the war, and the majority of the dead were women and children killed in air strikes. The second study, published in 2006, estimated that 655,000 excess deaths had occurred because of the war; of these, more than 600,000 were violent deaths. And while the proportion of deaths attributed to coalition forces had diminished, the actual numbers of those killed by the coalition had increased each year. (While there were some problems with the methodology of the two studies, the findings of the second, say the authors, are remarkably consistent with the first. Here’s one of many contemporaneous news articles about the controversy ignited by the publication of the 2006 report.) In light of these statistics, the estimates repeated by the Times and NPR—in 2011 and just the other day—look woefully inadequate.

Burkle and Garfield note that Iraq continues to struggle with a public health emergency. At the conclusion of their article, they write

In truth, because of the politicization and perceived weaknesses of the methods of the Iraq studies, all the studies of civilian death have been discounted or dismissed, yet if half a million civilians have perished, that information should be known. The only accurate death records are of US and coalition forces. Public health data, once untouchable, are increasingly controlled by political decision makers. They cannot have it both ways in defining the ground truth; in every war, combatant forces of states and the leaders they serve must be accountable.