How Many Civilians Die in Iraq?

In response to an article written by Gil Eliot in the Guardian newspaper, the researchers of the Lancet Report respond to his misconstrual of the actual number of civilian deaths in Iraq.
--------
Taking responsibility for Falluja
Thursday November 17, 2005
The Guardian

Gil Elliot's critique of the Lancet study into Iraqi deaths (Letters, November 10) is misleading and adds little to the debate. The figure of 98,000 deaths is not extrapolated from 21 violent deaths, as Elliot suggests, but from all-cause mortality outside Falluja (89 deaths). The high mortality observed in Falluja should not be surprising given the bombardment the city endured, and the gender bias is acknowledged and discussed in the paper. It is important to remember that 98,000 deaths is based on conservative assumptions and careful sampling. It is statistically the most likely number of excess deaths. Given the data, there is a 90% chance that the true value is greater than 44,000 and a 50% chance that it is greater than 98,000. This study is now a year old and the violence continues unabated, claiming more lives. We must have a full and independent inquiry into excess mortality in Iraq.
Tom Yates, Dr Judy Cook, Dr jack Piachaud
Medact