I waded through the pdf UN report to find out what they were doing to measure deaths, and it appears that they were including deaths fom all causes. This puts the Lancet study 3 to 5 times higher than the UN study, depending on the increase in death rate since the UN study.

More studies are definitely needed, but trying to compare the volumes of material in Iraq with those on global warming is silly.

The ILCS data has been derived from a question posed to households concerning missing and dead persons during the two years prior to the survey. Although the date was not asked for, it is reasonable to suppose that the vast majority of deaths due to warfare occurred after the beginning of 2003.

The question asked in ILCS was formulated and posed in a relatively standard way typical to large surveys and censuses (UN 1983). The question underestimates deaths, because households in which all members were lost are omitted. It is therefore common within demographic studies to use a correction for this, based on a number of assumptions derived from stable population theory (UN 1983). This has not been attempted here, as it is unlikely that the assumptions are satisfied. It is not common to make this correction in epidemiologically oriented studies,