My Photo
Blog powered by Typepad

April 2020

Sun Mon Tue Wed Thu Fri Sat
      1 2 3 4
5 6 7 8 9 10 11
12 13 14 15 16 17 18
19 20 21 22 23 24 25
26 27 28 29 30    

« Yishai of Israel writes: ‘An Open Letter to A citizen Of Gaza: I Am the Soldier Who Slept In Your Home’ | Main | Rep. James Clyburn turnaround: If more time is needed to get Iraq right, no need to hold fast to 16-month pullout deadline »

January 31, 2009


Sf Mead

The figures posted for military suicides are for both genders, but the figure for civilian suicides is male only.

But are there only men in the military? I think not.

In mid-2008, about 14 percent of the military were women, and civilian suicide rates among women are 4.4 per 100,000

So, using these figures in combination with your figures, we get:

(19.5 x 0.86) + (4.4 * 0.14) = 17.4 suicides per 100,000 in the military if they actually matched civilian suicide rates for men and women (I'm using the U.S. Army-provided rate for civilian men suicides here).

As such, 19.15 is much higher than the 17.4 we'd have if the military rates actually matched civilian rates for men and women.

Frank Warner

Even the 14 percent of women in the military aren't required to be in combat, so it would be deceptive to try linking combat-related experience to suicides of women in the armed forces. You might be able to include a female factor more validly by finding out what percent of all U.S. deaths in Iraq are women. I doubt that percentage is very high.

Frank Warner

Women account for about 2 percent of the Iraq war deaths, so we might fairly conclude women are exposed to about 2 percent of the combat action.

The expected suicide rate of 98 percent young civilian men and 2 percent young civilian women could be calculated this way:

(19.5 x 98%) + (4 x 2%), which equals
19.19 suicides per 100,000 civilians.

The armed forces last year had a suicide rate of about

19.15 per 100,000 troops exposed to combat.

Looks pretty close.

jj mollo

Uh, Frank, I think there's something wrong with that. You're talking about the expected rate among the victim population? Not the source population? Well, maybe ... I have to think about it.

Basically you're saying that so few women died of any cause in Iraq that they can be ignored. But suppose the overall population of women in Iraq was actually twice as many as men. Then such a small number of deaths would be very indicative of a discrepancy which muddles your conclusion.

I don't know about your logic. I do agree with your assessment of the numbers though. Something about combat was apparently keeping them from killing themselves. There is no statistical difference now with a reasonably comparable population and Sf Mead is just being captious.

My suspicion, though, is that some of the current suicides are really just postponed from earlier, and that the base suicide rate is still lower than stateside. These people are tougher than the average citizen. It would be interesting to know what the rate is for returned Iraq vets who were exposed to combat.

Frank Warner

It's not only about American women troops in Iraq. It's factoring in those U.S. women troops (about 2 percent of the U.S. armed forces) exposed to an equivalent amount of combat as men.

We're not trying to compare civilian suicides to suicides in a military population that lives an essentially civilian life. We're trying to compare U.S. civilian suicides to suicides of U.S. troops with a real risk of exposure to combat.

Frank Warner

It's comparing what you'd expect the suicide rate to be among young adult civilians to the suicide rate of troops, who are primarily young adults.


About 32,000 civilians commit suicide each year with many, many more attempting it. Source:CDC.
The data is very lumpy by age, gender, and race. About 73% of suicides are Caucasian males.
Your points are valid in that military and civilian ddmographics have to be controlled for.
The next point is that CDC takes several years to put data together so the latest CDC data is 2004, maybe 2005. Even in 2004 civilian suicide rates were trending up. But the Army is using current data.
See also Strengthening the Validity of Population-Based Suicide Rate Comparisons: An Illustration Using U.S. Military and Civilian Data


Looking at overall death rates matched by age, far more young men died in the US than in combat. Even comparing rates, there are occupations with higher death rates.

The most dangerous thing you can do is give a young male car keys and alcohol.

The comments to this entry are closed.