Posted by: Lister | January 7, 2008

Death toll in Iraq

I’ve looked into this a few times. Including last September when a poll was published claiming Civilian toll in Iraq may top 1 million. That survey was not nearly as accurate as the two published in the Lancet. But it opened up the whole debate. (btw, the Just Foreign Policy website extrapolates the Lancet results — it is not a new survey. The JFP assumes a constant rate of under-reporting by Iraq Body Count. The ORB poll is in line with JFP. Although neither, of course, are of the standard of the Lancet survey.)

We have 3 polls — each come up with the same conclusion: “The total number of deaths is far higher than any count of bodies”. You have to assume they are either very unlucky in the stats they collect or they have an intention to collect stats that don’t represent what is going on. Or… The stats they collect do reflect what is going on.

So are they
1) unlucky
2) conpsiring to deceive
3) correct

The British government looked into it. And, in spite of what Tony said in public, the government’s experts came down on the side of the Lancet:

The British government was advised against publicly criticising a report estimating that 655,000 Iraqis had died due to the war, the BBC has learnt. Iraqi Health Ministry figures put the toll at less than 10% of the total in the survey, published in the Lancet.

But the Ministry of Defence’s chief scientific adviser said the survey’s methods were “close to best practice” and the study design was “robust”.

Another expert agreed the method was “tried and tested”.

[…] a memo by the MoD’s Chief Scientific Adviser, Sir Roy Anderson, on 13 October, states: “The study design is robust and employs methods that are regarded as close to “best practice” in this area, given the difficulties of data collection and verification in the present circumstances in Iraq.”

All war zones find it difficult to count deaths. In Bosnia, which was by no means as chaotic as Iraq, they counted about 30-40% of deaths during wartime. That’s unusually high, according to Les Roberts. 10% isn’t so far from what’s normally seen.

1) unlucky sample
Can we just dismiss two surveys published by the Lancet by saying that they were really bad at selecting their samples? The odds don’t add up. The ORB poll was entirely independent and its results are much more in sync with the Lancet studies than with Iraq Body Count.

And it’s not just the number of deaths that the Lancet were unlucky to discover. It’s the mode of death. The CIA World Factbook lists Iraq as having a death rate of 5.26 deaths/1,000 population (per year). Population of Iraq is 26,783,383 (2006 est).

5.26*26783 = about 141000 per year

Even the IBC total for the whole war doesn’t match that. So if IBC is correct, then far less than half the deaths in Iraq are due to violence.

Lancet-2, however, recorded 547 post-war deaths, about 300 of them violent. (Amongst 12,801 people in the survey)

Surely the ratio of violent:non-violent deaths in a year ought to give us a clue whether IBC or the Lancet report is more accurate. But I can’t find that figure on the internet.

2. conpsiracy to deceive
I suppose anything can be faked. But in this case it would take more than just one person. The ORB survey was independent of the Lancet-report’s authors. Of course, we get very few surveys of the standard published in the Lancet. That in itself is a question that needs to be asked. Why do we have so little data? If we had more, it would be easier to judge the Lancet reports.

Also, note the point at the end of the above: the ratio of violent to non-violent deaths is a measure that could blow Lancet-2 out of the water. It seems an easy figure to get hold of, but no-one seems to have done so yet — even though there are people who wish to discredit the Lancet report.

But the figures I requested would indeed be hard to fake. Morgues all over the place would have to give false data. That would require loads of people to be involved in the cover-up. A count of death certificates is something only a government ministry can do. And, in a war zone, they tend to be bad at making such counts — without intent, it’s just a difficult job.

In Lancet-2 (and possibly Lancet-1, also) families that reported a death were asked for death certificates. More than 80% provided one. Here’s the kicker: they were asked at the end of the interview — not the beginning. After they had already reported the death, they were asked for the certificate.

And for those that didn’t provide a certificate, the numbers of deaths they gave matched with the 80%+ who did give certificates. (I don’t know the extent of the match).

This implies, of course, that more death certificates were issued than were officially counted. I think this can be done without intent to deceive — it’s not the first time that a government has been unable to accurately count deaths in a warzone.

So why are death certificates issued?
There are so many things that a bereaved family member cannot do without a death certificate for whoever died. Can they even bury their loved one in a cemetery without a death certificate? Can you just turn up with a dead body and say “Here, would you bury this for me?”

Then there’s insurance, inheritance, etc.
So the bereaved will make sure they get a death certificate. What happens after that? The counting isn’t automated, as far as I know. How many other jobs does an Iraqi doctor have to do. He HAS to give out the death certificate, because the family won’t leave him alone until he does. What needs to be done afterwards for that certificate to be counted? I don’t know.

3. The stats are correct
Well if people were going to accept that, I wouldn’t be summarizing old debates so that the points are easier to get hold of in the future.

Iraq morgue ‘receives 1100 dead’ (Aug 2005):

THE number of dead Iraqi civilians counted at the Baghdad morgue hit 1100 in July, the highest toll in recent history, a British newspaper reported today, blaming the daily violence.

[…] The death toll was up from about 800 in July last year and 700 during the same month in 2003, according to the left-wing daily.

By comparison, equivalent figures for July 1997, 1998 and 1999 – during the leadership of former Iraqi president Saddam Hussein – were all below 200, The Independent said.

Many of today’s corpses were badly mutilated, meaning that between 10 and 20 per cent of them were never identified, the newspaper said.

Since January, the medical authorities have buried 500 nameless bodies.

From 200 to 1100.
The Lancet-2 report only reckons the death rate went from about 5.5 to 13.3

So how many morgues closed? Why this big increase in the arrivals at one morgue? And that was from 2005 — the level of violence we’d all slap ourselves on the back for returning to.

The UN report
Les Roberts, quoted at scienceblogs:

B) This UNDP survey covered about 13 months after the invasion. Our first survey recorded almost twice as many violent deaths from the 13th to the 18th months after the invasion as it did during the first 12 (see figure 2 in the 2004 Lancet article).

The second survey found an excess rate of 2/1000/year over the same period corresponding to approximately 55,000 deaths by April of 2004(see table 3 of 2006 Lancet article). Thus, the rates of violent death recorded in the two survey groups are not so divergent.

Especially when you take into account Pedersen’s belief that his own survey under-estimated the deaths:

The JHU study, [Pedersen] noted, asked Iraqis only about mortality. The U.N. study asked Iraqis about many aspects of their living conditions. Pedersen said his study probably underestimated deaths caused by the war because the interviews did not focus on the issue, while the Lancet article probably overstated them because no other subject was discussed.

For deaths up to April 2004:
Lancet-2 gives 55,000
Against Pederson’s UNDP: 18,000 to 29,000 is an under-estimate.

Some of you may be thinking: “Hold on. That’s unlikley. If someone asked me 100 questions or 2 questions… And one of those was ‘has anybody died?’ It wouldn’t change my answer.”

But the UN had some awsome resources. When their figures for infant mortality didn’t tally with what they were expecting from previous surveys, they sent interviewers back out to talk to the SAME people. As a result, they corrected (I suppose) their undercount of infant mortality.

Download the report:

When the ILCS was conducted, data were evaluated as they came in, and it soon became apparent that the infant and child mortality rates would turn out considerably lower than expected. Interviewers were therefore asked to pay particular attention to the birth history section of the interview.

Nevertheless, when first estimated, the mortality rates turned out to be lower than those published here. Therefore, it was decided to conduct control interviews in order to check if all births and deaths were recorded. First, a sample of 500 households in Baghdad that had already been interviewed were interviewed again. The questionnaire used consisted of just the birth history and parts of the labour force section. Once it turned out that there were indeed some omissions of births and deaths, it was decided to re-interview all households again with the small questionnaire.

Isn’t that amazing? I include it because it teaches not to trust expectations. If you thought that people would remember a death in the family regardless of how long the interview was, then maybe you think that the media would report most of the car-bombs that go off in Iraq. But then you are arguing from your expectations, and not from data.

Patrick Cockburn in the Independent on the difficulty of reporting from Iraq:

The sectarian warfare in Baghdad is sparsely reported but the provinces around the capital are now so dangerous for reporters that they seldom, if ever, go there, except as embeds with US troops.

Iraq’s Health Ministry ordered to stop counting civilian dead from war (Dec 2003):

Iraq’s Health Ministry has ordered a halt to a count of civilians killed during the war and told its statistics department not to release figures compiled so far, the official who oversaw the count told The Associated Press on Wednesday.
The health minister, Dr. Khodeir Abbas, denied in an email that he had anything to do with the order, saying he didn’t even know about the study.

Dr. Nagham Mohsen, the head of the ministry’s statistics department, said the order was relayed to her by the ministry’s director of planning, Dr. Nazar Shabandar, who said it came on behalf of Abbas. She said the U.S.-led Coalition Provisional Authority, which oversees the ministry, also wanted the counting to stop.

“We have stopped the collection of this information because our minister didn’t agree with it,” she said, adding: “The CPA doesn’t want this to be done.”

Les Roberts (one of the authors) defends himself at the Independent.



  1. The NYTimes answers one of my above questions:

    […] At the same time, Iraqi officials have asserted that they made improvements in their ability to track fatalities using morgue counts and other means. One shortcoming has always been that the corpses of many victims, if they are identifiable, are taken by family members straight to the cemetery, bypassing the morgue and hospital. Yet Iraqi authorities say that relatives still have an incentive to obtain a death certificate because it is required for inheritance, for government compensation, and for other purposes.

    My post on the WHO’s Iraq survey.

  2. Robert Fisk on business at the Baghdad mortuary (Aug 2005):

    July was the bloodiest month in Baghdad’s modern history – in all, 1,100 bodies were brought to the city’s mortuary; executed for the most part, eviscerated, stabbed, bludgeoned, tortured to death.

    […] Of the dead, 963 were men – many with their hands bound, their eyes taped and bullets in their heads – and 137 women.

    […] The figures for this month cannot, of course, yet be calculated. But last Sunday, the mortuary received the bodies of 36 men and women, all killed by violence. By 8am on Monday, nine more human remains had been received. By midday, the figure had reached 25.

    […] in July 2003 – three months after the invasion – 700 corpses were brought to the mortuary in Baghdad. In July of 2004, this rose to around 800. The mortuary records the violent death toll for June of this year as 879 – 764 of them male, 115 female. Of the men, 480 had been killed by firearms, along with 25 of the women. By comparison, equivalent figures for July 1997, 1998 and 1999 were all below 200.

    […] In November 2004, the Foreign Secretary, Jack Straw, supported an estimate from Iraq’s ministry of health that 3,853 civilians were killed and 15,517 injured between April and October. This gives an annual death rate of 7,700, a third of the IBC estimate.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s


%d bloggers like this: