Skip to content

US Military Fatality Statistics: Have Suicides Increased?

All statistics were gathered from this DOD site. 2010 numbers were current as of 5 June this year; they were not part of the plots below.

In 1983, the year in which yours truly entered into servitude with his Uncle Sam, the United States military fatality rate was 0.1%. That is, roughly 1 out of every 1000 service members handed in their helmets early. This rate dropped rapidly, reaching a low in 2000, when about 1 out of every 10,000 died in uniform.

That rate, as to be expected unfortunately, has risen since the start of the last two wars. There was also an upward blip during the First Gulf War, as shown in this picture (from 1980 to 2009):

US Military Fatalities overall

Interestingly, the rate for the last two years, even though we are engaged in two wars, is less than it was at the start of the 1980s.

The 2010 data only includes data until 5 June: but the rate so far is 3 out of every 10,000; and is projected to be lower than 2009.

The distinct causes of death are more interesting, as shown in this picture:

US Military Fatalities by type

Accident fatalities have dropped dramatically, resurging during the surge, then tempering back to their low levels, presumably after the new recruits were trained.

Naturally, the fatality rate due to hostilities increased by a bunch with the start of the wars. But just look at that rapid fall off after the surge in 2007! Fatalities in 2008 and 2009 were about 2 out of every 10,000, a halving from the previous year.

You might have heard other reports from on high, but as this picture shows quite dramatically, the surge worked.

The homicide rate of the United States—as a whole—is about 5.4 per 100,000, a rate which has held somewhat steady over the last decade. But the homicide rate for the military has been—and still is—lower than in the population.

In 2009, the military rate, while high, was still lower than in the population. And, so far in 2010, it looks to be returning to a level of about 2 per 100,000.

Deaths due to illness also decreased until the wars, as might be expected. Well, maybe not as expected. The drop from 1980 to 2000 is itself interesting and noteworthy. The rate, even in 2009, is still lower than the US population. (But this might be due to age: military members are relatively young and the young do not die from disease at the same rate as the old, of course. I do not have time to compile age-matched statistics to compare.)

The most unfortunate, and non-ignorable, statistic is the suicide fatality rate, or “Self Induced Fatalities”, to put it in government-speak. It might not seem odd that this rate has increased during the current wars, but then why the increase in the mid-1990s? After all, the First Gulf War was already over four years before that rate peaked. I don’t have an answer.

The good news is that the current rate, as of 5 June, looks to be coming in at half last year’s rate. The other item of significant note is that the current high rate (as of 2009) puts the military on roughly equal terms with the population as a whole. That is, for most years, the suicide rate for service members is lower than in the population.

Finally, the fatality rate due to terrorism is low. The Marine barracks bombing in Lebanon sticks out and cannot be ignored, but apparently the terrorist attack of Maj. Nidal Malik Hasan, who shot up Fort Hood in the name of a belief which shall remain nameless.

Hasan murdered 13 in what all but members of the current administration call a terrorist attack. Yet the official statistics say “0” died in 2009 due to terrorism. Surely an oversight?

7 thoughts on “US Military Fatality Statistics: Have Suicides Increased? Leave a comment

  1. Regarding Hasan, I can’t see how is rampage can be considered terrorism.

    Terrorism is the use of terror as a means of coercion. Killing or bombing to achieve some sort of political end, or to create a climate of fear. One loony with a gun is not a terrorist. And, shouting ‘Allah Akbar’ doesn’t change that.

    Was the Virginia tech shooter a terrorist? The Columbine kids? Terrifying, but not terrorists.

    As for the other numbers, when the Iraq war began, we would hear the daily and monthly casualty figures. A colleague pointed out to me, that the number of service men that died in 2003 was not significantly higher than the numbers that had died in previous years when there was not a war going on. However, rather their deaths being calked up to “accidents” they were KIA.

  2. I agree with Doug on this.

    Terrorism is an act that has a goal. I don’t honestly believe that Hasan had any goal other than to cause mayhem and take his loony misery out on others.

    I dislike the liberal use of the word terrorism largely because I think we should be careful with its application. It’s become a buzzword.

    Bear in mind that I don’t think that this isn’t terrorism because it was against a military target. It’s clear that attacks on military targets can be terrorism. I think it wasn’t terrorism because even if he claimed to be acting as an agent of some bizarre form of Islam, his “political goals” were muddy at best. Furthermore, it seems that he acted alone:

    http://www.fbi.gov/pressrel/pressrel09/forthood111109.htm

    Terrible? Violent? Yes. Not terrorism. Just meaningless violence from a sick individual.

  3. I could be wrong, but I’m pretty sure the increase in “self-induced fatalities” in the mid 1990’s was due to the release of the “Jagged Little Pill” album by Alanis Morrissette and especially her music videos, which were broadcast globally.

  4. SM,

    I might have guessed Beatles reunion tour, but I won’t say you’re far off.

    Ari, Doug M,

    Can we come up with a robust definition of terrorism? Just having a goal seems to fit our boy in this situation.

  5. Back in the Nam days, the chance of coming home was around 90% IIRC and I understand that was quite an improvement over the Korean war. Gosh a 0.1% fatality rate!

    What exactly is a terrorist in a war? We just had enemy back then. Is this the new word for ‘enemy’ since this is the war against ‘terrorism’? Oh, wait! We also have “hostilities”. What’s the difference? Somebody setting off a bomb in Saigon was called “Charlie”. It was still considered enemy action.

    The suicide rate is interesting. These people weren’t dragged there kicking and screaming (i.e. drafted) but signed up to go. Maybe they found out war isn’t quite the same as a game of Doom? Maybe the reason is just plain boredom?

    What is homicide in a war? Presumably war is all about homicide. I suppose what is meant is not-so-friendly friendly-fire. During the Vietnam war most of the ‘homicides’ occurred during leave and usually over women or gambling though some of our so-called friendly-fire deaths on patrol were actually murders. Perhaps the resurgence of ‘homicide’ is also indicative of boredom?

    But damn the fatality rate is still 1 per 1000 ! They can send aman to the moon; why can’t they conduct a safe war?

  6. DAV,

    The Civil War was the deadliest war in American History. A Safe War? Even a shouting match can injure people mentally and physically.

    No War..

    Peace.
    ^_^

  7. “The most unfortunate, and non-ignorable, statistic is the suicide fatality rate, or “Self Induced Fatalities”, to put it in government-speak. It might not seem odd that this rate has increased during the current wars, but then why the increase in the mid-1990s? After all, the First Gulf War was already over four years before that rate peaked. I don’t have an answer.”

    Some portion of alcohol related accidents have always been suicides.

    Someone gets depressed, liquored up and then drives into a tree at 100 MPH. These deaths are almost always counted as ‘accidents’.

    Social acceptability of ‘drunk driving’ plummeted in the early 1990’s.
    In the current conflicts, Centcom General Order #1 prohibits alcohol consumption.

    Hence, the ‘suicides classified as accidents’ has gone down.

Leave a Reply

Your email address will not be published. Required fields are marked *