Crunching the Numbers on the Rate of Suicide Among Veterans

320px-veterans_dayBy Matt Wood

In a recent column in the New York Times, Nick Kristof pointed out a startling statistic: for every American soldier killed in combat this year, 25 will commit suicide. A report from the Center for a New American Security says that from 2005 to 2010, service members took their own lives at a rate of one every 36 hours, and the Department of Veterans Affairs says that 18 veterans commit suicide every day.

Further analysis of this epidemic uncovers even more troublesome data. In a recent study published in the American Journal of Public Health, a statistician from the University of Chicago Center for Health Statistics found that the youngest group of veterans, ages 17-24, were almost four times as likely to commit suicide than nonveterans.

Robert Gibbons, PhD, professor of biostatistics in the departments of Medicine and Health Studies has long been interested in statistical analysis of suicide data to uncover patterns that could point to potential intervention strategies and treatment policies. He was on the Food and Drug Administration (FDA) committee in 2004 that issued a “black box warning” about increased suicide risk in children taking antidepressants. He actually voted against the warning, citing concerns about ambiguous data, and earlier this year published his own statistical analysis of the clinical trials the FDA used for their warning and found that the drugs were quite effective in decreasing suicide risk in adults and neither increased nor decreased suicide risk in children and adolescents.

In his current work on veteran suicide, Gibbons and his colleagues tackled another statistical disagreement. The editors of the American Journal of Public Health asked him to write an editorial about two studies on the rate of suicide among veterans that showed conflicting results. He and his colleagues weren’t able to replicate the results of either study, so they decided to do their own using publicly available data.

“We realized there was just no way we were going to be able to be Solomon-like and opine about who was right or wrong,” Gibbons said. “So we decided to see if we could figure out a way to shed some light on the answer using stuff off the internet, something a 10-year-old could get.”

Those data are from the Centers for Disease Control and Prevention National Violent Death Reporting System, which provides detailed data on all violent deaths, including suicides, from 16 states from 2005 to 2008.

“It’s a very important data source that very few people know about or use, so we’re really happy to use it. It’s a great national resource,” Gibbons said. “I think what’s cool about it is that it shows you can actually learn something from data that are completely publicly available, and learn something that is probably right.”

The key difference in how Gibbons and his colleagues analyzed the data versus the two previous studies is that they looked at how the rate of suicide changed with age. Statistical analysis of epidemiological data commonly controls for age, but a more detailed stratification by age is required to tease out the effects of recent military service and the associated emotional and psychological traumas of war.

The problem with the two conflicting earlier reports was that they were based on national surveys that did not control for the time at which the veterans actively served in the military.  One study included veterans with more recent military service and identified an increased risk of suicide risk. The other study looked at veterans further removed from military service and did not find an effect.  Gibbons and his colleagues explained these differing conclusions in the editorial using their new results.

“The bottom line as it turns out there’s a really big risk of increased suicide right when you get out of your military service,” Gibbons said. The risk of suicide in the youngest group of veterans was almost four times higher than nonveterans. That number drops off after the age of 24, but remains around one and a half times higher.


suiciderateThe Department of Veterans Affairs recently announced that it will be hiring 1,600 additional psychiatrists, psychologists, social workers and clinicians to address veterans’ mental health needs, and there is evidence that vets with post-traumatic stress disorder could be suffering from the same kind of brain disease associated with sports concussions. Gibbons said he hopes studies like this will underscore the problem and add urgency to these efforts.

“I think it’s the kind of thing that will drive dollars to make sure that veterans get the kind of treatment that they need for what obviously is a very life-changing experience,” he said. “The National Institutes of Health and the Department of Defense have launched a huge initiative to study suicide and the psychological and biological basis of suicide in the military. That’s a good idea because clearly there’s something going on.”

He pointed to the big spike on the graph of veteran suicide rates in the paper (shown above) and said, “You don’t go from this to this by chance.”

=====
Gibbons, R., Brown, C., & Hur, K. (2012). Is the Rate of Suicide Among Veterans Elevated? American Journal of Public Health, 102 (S1) DOI: 10.2105/AJPH.2011.300491

About Matt Wood (318 Articles)
Matt Wood is the editor of the Science Life blog and the social media specialist for the University of Chicago Medicine.

8 Trackbacks & Pingbacks

  1. A Note for Veteran’s Day 2012 | Mesopotamian Marine
  2. Older Veterans Also Have Mental Health Needs | Health
  3. Veterans Day 2012 | Sean Huze
  4. Older Veterans Also Have Mental Health Needs |
  5. Michael Friedman, L.M.S.W.: Older Veterans Also Have Mental Health Needs - Health Medicine Network
  6. Fuck The Vote. Riot. « After The Massacre
  7. The Disturbing Trend of Military Suicide | The Morningside Post
  8. Year in Review: UChicago Research 2012 « SCIENCE LIFE

Comments are closed.

Follow

Get every new post delivered to your Inbox.

Join 1,266 other followers

%d bloggers like this: