Iraqi Death Study Needs Consideration

A recent study by researchers at the Johns Hopkins Bloomberg School of Public Health estimates that 654,965 Iraqis have died as a result of U.S. and allied intervention in the country since March 2003.
The study, which was published in the scientific journal ‘The Lancet,’ looked at yearly mortality rates before the U.S. invasion (5.5 deaths per 1,000 people) and compared them with rates of violent and nonviolent deaths in the years since (13.3 per 1,000).
Their findings, put bluntly, suggest that 2.5 percent of Iraq’s population has died as a result of the U.S. intervention.
George W. Bush was quick to dismiss the findings, saying that ‘six hundred thousand or whatever they guessed at’ is ‘not credible’ and denouncing the methodology of the study as ‘pretty well discredited.’
In December 2005, Bush estimated that 30,000 civilians had been killed in Iraq, but since the study was released, he has declined to provide a more recent estimate of his own, saying only that ‘a lot of innocent people have lost their life [sic].’
Bush’s assault on the methodology of the study and his insistent reliance on a much smaller estimate of Iraqi deaths betray either a misunderstanding of the new survey or a deliberate misrepresentation of it.
Firstly, the study’s methodology is not ‘pretty well discredited,’ but defended by several prominent statisticians.
Instead of relying on governmental records or media reports, as past estimates have done, the new study is based on personal interviews with 1,849 Iraqi households in randomly selected neighborhoods, a technique that has been used in other conflict situations, as in Congo, Rwanda and Darfur.
‘The sampling is solid,’ said John Zogby, the well-known pollster. ‘The methodology is as good as it gets.’
In fact, most of the biases that may have been introduced into the survey seem more likely to understate deaths than to overstate them.
For example, two governorates, representing five percent of Iraq’s population, were accidentally omitted from the study and were counted as reporting zero deaths.
Also, when surveyors found empty houses, they skipped them, even though one possible explanation for their vacancy is that ‘entire households … have been killed.’
Although it may initially seem that Iraqis could have over-reported the number of deaths in their families, according to the study, 92 percent of reported deaths could be corroborated with a certificate and ‘the pattern of deaths in households without death certificates was no different from those with certificates.’
If the methodology is sound, then how can the numbers this survey yielded be so far off from other estimates? The answer is simple: They are measuring totally different things.
The Johns Hopkins study is unique in that it does not differentiate between combatants and noncombatants and considers both violent and nonviolent deaths.
The Web site, which relies exclusively on media sources and reports only civilian deaths caused directly by U.S. or allied troops or collateral deaths ‘to which we can unambiguously hold our own leaders to account,’ estimates between 43,937 and 48,783 deaths in Iraq to date.
The Johns Hopkins study does not consider whether Iraqis killed were civilians, nor does it take into account who or what killed them.
Most of the reported deaths (601,027) were due to violence, most commonly gunfire, but some of the increase was attributed to ‘insufficient water supplies, non-functional sewerage and restricted electricity supply’ as well as ‘a deteriorating health service with insecure access.’
It doesn’t make sense to compare deaths of Iraqi civilians directly at the hands of American or allied troops to deaths of all Iraqis from causes including coronary disease and car bombings, but the United States cannot be held blameless for deaths caused by increased civil unrest and deteriorating quality of life in Iraq post-invasion.
‘All surveys have potential for error and bias,’ as the study admits, but the results of this particular survey should have been carefully considered before Bush chose to issue a knee-jerk reaction against them.
Even if top governmental officials can’t be bothered to read and understand the Johns Hopkins survey, you can read more about its methodology and findings online at (free registration is required).

Send comments to Please include your name, year and major.