To: email@example.com Subject: Correction for "Exact Death Toll of Iraqis Remains Murky", Jim Krane, Fri Mar 10, 12:53 PM ET
I wanted to reply with a request that AP clarify or correct its statements regarding civilian death tolls in Iraq in the future.
In the aforementioned article, Mr. Krane states that "the staff (of the Baghdad mortuary) has photographed and catalogued more than 24,000 bodies from the Baghdad area alone since 2003..." and that "President Bush has said he thinks violence claimed at least 30,000 Iraqi dead as of December, while some researchers have cited numbers of 50,000, 75,000 or beyond."
That said, Krane also states in passing, almost dismissively, that "in late 2004, a study published in the Lancet medical journal estimated the war had caused some 98,000 civilian deaths. But the British government and others were skeptical of that finding, which was based on extrapolations from a small sample."
The truth of the matter is that the extrapolations for the Lancet report weren't from a small statistical sample at all. Approximately 1000 Iraqi households throughout the country were surveyed extensively throughout Iraq to create the Lancet report, which any real statistician will tell you is quite significant. That's a larger statistical sampling than many of the political polls cited regularly by the press, infact.
Another issue with Krane's article is a glaring lack of clarity on his part as to what a civilian casualty actually is. Surveys like Iraq Body Count tally only on those Iraqi civilian deaths which are reported by multiple Western news sources. This goes a long way to explain why their count for civilian casualties is only about 35,000 for all of Iraq, while the Baghdad morgues alone have counted over 24,000 bodies in a city that contains only about 1/5th of the country's population.
Even the Baghdad coroners or the people over at Iraq Body Count will tell you, however, that they do not and cannot tally all Iraqis who die from increased violence due to the war. The people at IBC have basically admitted that there are probably tens of thousands of Iraqi civilians who have died whose deaths were not reported in news accounts, and who were buried by their families and probably never even arrived at the mortuaries in question.
Compare the methodologies behind these lower civilian death totals with the Lancet report, and the large gap in the numbers becomes obvious. The Lancet report tallies increased statistical morbidity across ALL CAUSES OF DEATH, and it made very clear that the rate of mortality has gone up for pretty much everything because of the conflict. (i.e. A broken medical system, lack of sanitation, polluted water, increased rate of communicable diseases, lack of electricity, lack of ambulances and emergency medicine, phone outages, and a simple inability for emergency response teams to get to and respond to patients due to conflict, roadblocks, etc.)
In other words, it's not necessarily violence that will kill you in Iraq. It's failure to get treatment for childhood diseases, or lack of air conditioning for the sick, weak, or elderly. There's a surge in infant mortality rates, and simple things like the inability to get an ambulance when you have a treatable heart attack can and do needlessly kill Iraqis every day. All of these causes are well above pre-war levels, which is why the Lancet report can conservatively say about 100,000 civilians died above the norm and be correct, even as Iraq Body Count can cite newspaper reports on civilian deaths through violence and also be correct.
Violence has, according to the Lancet report, become the #1 killer in Iraq, but there are many, many other increased causes of death, and those deaths are probably the most underreported ones in the Iraq conflict.
The leader of the Lancet report study was the same person who did similar statistical estimates in places like the Congo massacres, which Tony Blair repeatedly cited. It would be nice if Krane cited research from an statistician, rather than the say-so of the British or US government, when dismissing the Lancet Report's findings offhand.
The Lancet report, incidentally, corresponded very closely to the findings of another study at John Hopkins University, where returning soldiers were asked by psychologists whether they believed they killed Iraqi civilians, and if so, how many. The percentage of soldiers who returned who reported such deaths was around 25%, if I remember right. If you do the math, you'd find that the number they estimated, if they applied it across the board to returning soldiers, corresponds very closely to the Lancet figures of estimated civilian deaths caused by U.S. troops.
The truth is, the best estimate anyone can give as to actual civilian fatalities in Iraq would be to do something like the following:
1> Take the Lancet report -- which statistically calculated increased morbidity over a 19-month survey period. 2> Divide 98,000 by 19 months, for an average increased morbidity of 5158 Iraqis per month. 3> Add in an estimate for the 17 months of the war the Lancet hasn't reported. 5158x17=87686 ... and add it to 98,000. In other words, about 185,000 dead Iraqi civilians above the statistical pre-war norm of sanctions-era Iraq.
Realise that this is a conservative estimate, and that if Iraq Body Count reports that civilian violent deaths have doubled over the past year, which they did recently, any Lancet estimate would also likely increase, both as a result of increased violent deaths and as a result of the effects of such conflict on the nation's infrastructure. Also note that since the original Lancet estimate, some of the most violent incidents have occurred, such as the invasion of Fallujah.
"Please understand how extremely conservative we were: we did a survey estimating that ~285,000 people have died due to the first 18 months of invasion and occupation and we reported it as at least ~100,000."
You can contact Dr. Roberts. Maybe he can give you an interview and point out to you just how incredibly offbase you people are at times. Really, it shouldn't be too hard tracking him down, given that he teaches at Johns Hopkins. I would be glad to give you his phone number, if you would like.
I look forward to the Associated Press doing some actual work of their own in the future to determine more of the truth behind the casualty reports, rather than citing wildly erratic statistics that do not accuately reflect Iraqi civilian casualties as a result of the war. Your reporting on this matter is, frankly, appalling, considering that you could easily be dismissing about 150,000 dead Iraqis in such a myopic, uninformative, and unscientific manner.
You're a major news organization. Why don't *YOU* commission a scientific survey of the casualties if you feel that more information is required, rather than make it clear to the public how little you know about the subjects you're reporting on? Must others do all the work for you?