The Corpus Callosum

Ministers were told…

is depressing.  Now the UK is misusing and distorting
scientific findings.  I suppose it is contagious.
 This pertains to the Lancet
that found an estimated 50% increase in the risk of
death in the immediate aftermath of the Iraq war (2003-2004).  

could understand Tony Blair or George Bush, or any other politician who
does not know anything about research methodology, making an
off-the-cuff statement questioning the validity of the study.
 But that is not what happened.

were told not
to rubbish Iraq deaths study

March 26, 2007


government advisers accepted as “robust” research that put the death
toll from the Iraq war 10 times higher than any previous estimate, new
documents have revealed.

study, by the Johns Hopkins School of Public Health, prompted worldwide
alarm when it was published in the Lancet medical journal in October
last year.

estimated that 655,000 Iraqis had died due to the violence in the
country. It has now emerged chief advisers warned ministers not to
“rubbish” the report.

the time, both the British and US governments were quick to dismiss the
peer reviewed study. The Foreign Office said it was based on a “fairly
small sample …extrapolated across the country”. Iraqi government data
was more likely to be accurate, it added.

US was more blunt. President George Bush said: “I don’t consider it a
credible report.”

according to papers obtained by the BBC World Service’s Newshour
programme under the Freedom of Information Act, senior officials warned
the methods used in the survey were “robust” and “close to best

the UK government at least had enough wits to ask for an official
review.  But when they did not like the outcome of the review,
they ignored it.  Instead, they went ahead with a public
pronouncement that they knew was inaccurate.

Persons unfamiliar with research methodology claimed that the method
was not valid.  They also took note of the wide range of
possibilities within the confidence interval [a range of 8,000 to
194,000, excess deaths, using a 95% CI (confidence interval)].

The study was widely debated
in the Blogophere
, but no one has come up
with a sound refutation of the study.  In fact, a second study
published last year confirms the results of the earlier study.
 While the range within the confidence interval is still wide
(392,979 to 942,636 excess Iraqi deaths), even the lowest number
indicates that the daily risk of dying is much greater in Iraq now,
than it was before the war.