100,000 Iraqis dead since the invasion

[Blog]

For relative risk, anything above 1 is an increased likelyhood of the event happening. In this case I'm assuming its dying a violent death at the hands of evil American overlords. The higher the number, the greater the chance, but just because a number is above one doesn't mean that its really anything to write home about.

The RR of a 1/2 pack a day smoker getting lung cancer over a non smoker is ~10, of a smoker getting mouth/throat cancer ~5, of an asbestos worker who smokes getting cancer is 92. The RR of you getting prostate cancer if your dad has it is 2.1, if your brother has it 2.8. The RR of getting any form of cancer if you eat red meat once a week is 1.5 (compared to someone who doesn't eat red meat at all), if you eat 1 or less servings of fish a week its 2.0. I used to have this neat little table I wish I could find it, but the higher the RR the more likely it is to mean something.

Here is an example of the calculation:

http://scangis.dhec.sc.gov/scan/coho...cohortcalc.htm

There are other things that can mess up the numbers, like publication bias (generally, only positive RR's are published), but that is probably not an issue here.

Theres more general stuff here about the study:

http://www.boston.com/dailynews/303/..._000_Ir:.shtml

Including:

''I emailed it in on Sept. 30 under the condition that it came out before the election,'' Roberts told The Associated Press. ''My motive in doing that was not to skew the election. My motive was that if this came out during the campaign, both candidates would be forced to pledge to protect civilian lives in Iraq.

Getting an scientific article published in under a month is insane, as well as having the author stipulate conditions and have the journal meet them.