Live Science Plus
Soldiers At High Suicide Risk Can Be Identified with Math Model
It may be possible to identify which Army soldiers are at high risk for suicide by using a new mathematical model, a new study suggests.
Researchers analyzed information from more than 40,000 Army soldiers who were hospitalized for a psychiatric condition between 2004 and 2009.
It's known that people who are admitted to hospitals with a psychiatric diagnosis are at increased risk for suicide after they are released. But even among this high-risk group, suicide is relatively uncommon, and so it would not be practical for everyone released from a psychiatric hospitalization to undergo an intensive suicide prevention program, the researchers said. It would be more feasible to target intensive programs to those most at risk for suicide.
In the new study, 68 soldiers died by suicide within a year of being released from the hospital. That translates to a rate of 264 suicides per 100,000 hospitalized soldiers per year, compared with the rate of 18.5 suicides per 100,000 soldiers per year among all U.S. Army soldiers. [5 Myths About Suicide, Debunked]
The researchers fed information from Army and Department of Defense administrative files into a computer program to look for factors that predicted suicide risk. Previous research has shown that computer algorithms are much more accurate at predicting a person's suicide risk than are doctors. Unlike a person, a computer model can consider hundreds of potential risk factors at once.
In the new study, the researchers' program looked at 131 variables linked with suicide risk, from basic factors such as gender and age to details such as whether the person had access to a firearm, was previously treated for a psychiatric illness or currently had post traumatic stress disorder.
The study found that the 5 percent of soldiers who were predicted by their model to have the highest risk of suicide after their hospital discharge accounted for more than half of the suicides in the study.
Get the world’s most fascinating discoveries delivered straight to your inbox.
"The high concentration of suicide risk in the 5 percent of highest-risk hospitalizations is striking," study co-author Ronald Kessler, a professor of health care policy at Harvard Medical School, said in a statement. What's more, this 5 percent was also at high risk for other adverse outcomes following the individual's hospital release, including dying from an unintentional injury, attempting suicide or being readmitted to the hospital.
The strongest predictors of suicide risk included being male, enlisting at a later age, possessing a registered firearm, attempting suicide in the past, as well as aspects of prior psychiatric treatment — such as the number of antidepressant prescriptions filled in the last 12 months, and disorders diagnosed during the hospitalization.
The suicide rate among Army soldiers has been on the rise since 2004. "Although interventions in this high-risk stratum would not solve the entire U.S. Army suicide problem, given that post-hospitalization suicides account for only 12 percent of all U.S. Army suicides, the algorithm would presumably help target preventive interventions," the researchers wrote in the Nov. 12 issue of the journal JAMA Psychiatry.
However, further research is needed before doctors could use the model. For instance, because the model was based on information from only 68 suicides, further testing will be required using more recent data to confirm how well the model predicts suicide risk, the researchers said.
Researchers also need to consider the potential for harm from the model, because undergoing an intensive suicide prevention program might lead to "undue scrutiny" that could affect a soldier's career, the researchers said.
Editor's note: This article has been updated to correct the number of people in the study. Information from more than 53,000 hospital visits was used in the study, and these visits involved 40,820 U.S. soldiers.
Follow Rachael Rettner @RachaelRettner. Follow Live Science @livescience, Facebook & Google+. Original article on Live Science.
Rachael is a Live Science contributor, and was a former channel editor and senior writer for Live Science between 2010 and 2022. She has a master's degree in journalism from New York University's Science, Health and Environmental Reporting Program. She also holds a B.S. in molecular biology and an M.S. in biology from the University of California, San Diego. Her work has appeared in Scienceline, The Washington Post and Scientific American.