Hicks and colleagues report in the April 16, 2009 issue of The New England Journal of Medicine regarding civilian non-combatant deaths during the current Iraq War period (2003-2008) was seriously flawed. Their stated goal was to analyze the causes of these deaths. To accomplish this purpose they used the database of Iraqbodycount.org that estimates 91,358 deaths due to “armed violence”. For this report the database was pared down to 60,481.

Several aspects of this are problematic. Iraqbodycount.org lists no publications in peer-reviewed biomedical journals. This is important since the methodology used by the authors on this website has been widely criticized. Their methodology lends itself to undercounting of the excess dead, because they require documentary proof of each death. This ignores an important and obvious problem that may be concisely phrased “the fog of war”. How this manifests in Iraq includes mass unmarked graves, rapid burials by Muslims of their relatives without notifying press or authorities, poor communications due to war-related destruction, etc.

Furthermore, to be included in this database, deaths must have been due to munitions. There is no mention of war-related water, food, or medical shortages or excess mortality due to the previous addition of chemical and radiological toxins in prior (sometimes lower level) war making (1991-2002) on Iraq by the U.S.

Other studies, using other methods, have estimated a much higher number of excess deaths due to the war, most notably those of the Johns Hopkins group, as published in The Lancet. Their mean estimate for the excess mortality March, 2003 to August, 2006 is about 655,000. They estimated mortality using a cross-sectional cluster sample survey. This group’s papers are not referenced by Hicks and coworkers nor is there any mention of this large discrepancy and the controversy it engendered. This, despite the fact that Dr. Hicks and Professor Spagat, a co-author, have extensively criticized the Johns Hopkins work elsewhere.

At the time of publication, the methods used by the Hopkins group were criticized bitterly by the Bush regime, though results using the same methods in the Congo were quoted approvingly by Colin Powell and Tony Blair. Clearly, epidemiology is not an exact science, but the Hopkins group approaches this fact scientifically, soberly working with the degree of imprecision inherent to the events being investigated. On the other hand, the current Iraqbodycount.org website does not make clear that though each and every fatality they list is certain, it is also certain that their total is inaccurately low.

It may be argued by the authors that their project is unrelated to the objections I bring forth here. After all, they sought to determine what sorts of munitions were responsible for acute munitions-related deaths. From that they could determine which party to the conflict is the proximate cause of death.
However it is important to place the data and conclusions there from in proper context. By referring only to the Iraqbodycount.org without background or further comment, the impression is given that there is no controversy surrounding the issue. As this is published as a Perspective in so distinguished a publication as The New England Journal of Medicine, an unearned imprimatur is given to the Iraqbodycount.org database.

Finally, all analyses and discussions of war mortality are enjoined from an ethical point of view, to fully consider, in context, the total excess mortality (and morbidity). Otherwise, a large number of the prematurely and wrongly dead are prematurely and wrongly forgotten.

I have sent a modified version of this comment to both The New England Journal of Medicine and The Lancet; they failed to publish it.

Note: The author, Morton S. Skorodin, M.D retains copyright of this article.