New Scientific Algorithm Better at Predicting Army Suicides
A scientific algorithm based upon medical records and actuarial data can better help identify those most at risk for committing suicide than other methods, according to a new study.
Previous research has found that such actuarial data was found to be a better predictor of soldiers at higher risk for suicide than a doctor’s clinical judgment.
The suicide rate in the U.S. Army remains at all-time high levels and exceeds the rate among civilians — nearly 30 deaths per 100,000 people (versus 25 deaths-per-100,000 in the civilian population).
Researchers wanted to better understand how they might identify those at the highest risk for suicide in order to develop better prevention strategies for them in the future. By targeting such strategies at soldiers at greatest risk, the researchers believe they can help reduce the Army’s suicide rate.
Researchers examined the medical records and actuarial data for 53,769 psychiatric hospitalizations of active duty soldiers over the course of five years from 2004 through 2009. They looked at over 130 different variables linked with suicide risk, ranging from basic demographics (like age and gender) to things like whether the person had access to a gun or had prior psychiatric treatment.
Scientists found that 68 soldiers committed suicide within 12 months of being released from the hospital. The model proposed by the researchers could identify 36 of those people.
Researchers found the strongest predictors for increased risk of suicide included sociodemographic factors such as being male, late-age of enlistment, criminal offenses, weapons possession, prior suicidality, aspects of prior psychiatric treatment (such as the number of antidepressant prescriptions filled in 12 months) and disorders diagnosed during the hospitalizations.
Soldiers in the highest predicted suicide risk group had seven unintentional injury deaths, 830 suicide attempts and 3,765 subsequent hospitalizations within 12 months of hospital discharge.
More than 50 percent of the suicides could be accounted for in the study by just 5 percent of soldiers who were predicted by the new algorithm to be at highest risk of suicide.
“The high concentration of risk of suicides and other adverse outcomes might justify targeting expanded post-hospitalization interventions to soldiers classified as having highest post-hospitalization suicide risk,” noted the researchers. “The high concentration of suicide risk in the 5 percent of highest-risk hospitalizations is striking.”
The high risk group predicted by the researchers’ new model also were at higher risk for other adverse events, including hospital readmissions, attempting suicide, and death from an unintentional injury.
It is believed that if future research confirms the validity of the researchers’ scientific algorithm, prevention strategies can be targeted at the highest-risk group.
The research group was led by Ronald Kessler, a professor of health care policy at Harvard Medical School.
The study appears in the latest issue of JAMA Psychiatry.
Source: JAMA Psychiatry
Grohol, J. (2018). New Scientific Algorithm Better at Predicting Army Suicides. Psych Central. Retrieved on September 30, 2020, from https://psychcentral.com/news/2014/11/13/new-scientific-algorithm-better-at-predicting-army-suicides/77294.html