Math Model can Predict Risk of Suicide in Soldiers
A new screening system shows promise in predicting suicide risk in soldiers, a recent study reported.
"The Army is working through complex privacy and ethical concerns regarding the use of these types of data, as well as updating the data from today's soldiers," Lt. Col. Ben Garrett, an Army spokesman, said reported by USA Today. "Use of the risk model will enhance the Army's ability to provide services to soldiers at elevated risk."
According to the researchers, the system is a computer program that takes into account numerous factors, such as military enlistment age, history of violence, prescription drug use, military rank, access to weapons and history of run-ins with leadership. Other models that are used to predict suicide risk rely heavily on self-reports and do not have strong predictive power.
For this study, the team used the screening system when examining the medical records of 40,820 soldiers. All of the soldiers were hospitalized between 2004 and 2009 due to a mental health problem. In order to determine the soldiers with the highest risk of suicide, the team tested and retested more than 300 factors. In the end, they were able to categorize about five percent of the sample as high-risk. The researchers found that these soldiers were about 15 times more likely to commit suicide within the year after being discharged.
"The most impressive thing is that they identified this high-risk group in the hospital, and by just focusing on one in 20 of them, you're really dramatically improving your ability to predict," Dr. Mark Olfson, a professor of psychiatry at Columbia University, who was not a part of the study, commented according to the New York Times. "Clinicians don't do a very good job predicting suicide risk, even though we think we do."
Overall, in the high-risk group, there were 36 suicides. In the remaining group, there were 32 suicides. Some of the factors that were linked to higher risk of suicide were previous suicide attempts, severe traumatic brain injury, and a history of using weapons. Since the system accounts for factors related to the military, the researchers stated that it would not be useful in predicting suicide risk for civilians.
"According to their estimate, we could save four lives for every hundred people we treated," Lt. Gen. Eric B. Schoomaker, a former surgeon general of the Army and a professor of military and emergency medicine at the Uniformed Services University of the Health Sciences in Bethesda, MD, said. "This would be unparalleled, compared to almost any other intervention we could make in medicine. This study begins to show the positive effects big data can have, when combined with administrative health records."
The study, "Predicting Suicides After Psychiatric Hospitalization in US Army Soldiers," was published in JAMA Psychiatry.