{"id":208983,"date":"2025-01-09T09:52:20","date_gmt":"2025-01-09T19:52:20","guid":{"rendered":"https:\/\/www.hawaii.edu\/news\/?p=208983"},"modified":"2025-01-09T09:52:20","modified_gmt":"2025-01-09T19:52:20","slug":"ai-model-veteran-suicide-risk","status":"publish","type":"post","link":"https:\/\/www.hawaii.edu\/news\/2025\/01\/09\/ai-model-veteran-suicide-risk\/","title":{"rendered":"New AI<\/abbr> model helps predict veteran suicide risk"},"content":{"rendered":"Reading time: <\/span> 2<\/span> minutes<\/span><\/span>
\"two
(Photo credit: Israel Palacio\/Unsplash)<\/figcaption><\/figure>\n

Editor\u2019s note:<\/strong> If you or someone you know is struggling with suicidal thoughts or mental health matters, please call or text 988 to connect with a trained 988 Suicide & Crisis Lifeline counselor.<\/em><\/p>\n

U.S. Army veteran suicides can be predicted with “moderate to good accuracy” by applying artificial intelligence (AI<\/abbr>) to data available before veterans leave service, according to a new study led by Chris J. Kennedy at Massachusetts General Hospital and co-authored by a University of Hawaiʻi<\/span> at M\u0101noa Department of Psychology<\/a> faculty member. The study in JAMA<\/abbr> Psychiatry<\/em><\/a> applied machine learning—a subset of AI<\/abbr> that enables computers to “learn” from data.<\/p>\n

\"person
Santiago Papini<\/figcaption><\/figure>\n

“This research could potentially reshape how the military approaches mental health support for service members transitioning to civilian life,” said Assistant Professor and co-author Santiago Papini. “By identifying at-risk individuals early, we may be able to provide more targeted and timely interventions, potentially saving lives.”<\/p>\n

In its testing, the model identified a high-risk group comprising 10% of soldiers. This group accounted for 30.7% to 46.6% of actual suicides.<\/p>\n

The most influential factors in the model\u2019s predictions were sociodemographic data (male or non-Hispanic White—higher risk; older age—lower risk), Army career characteristics (combat related duties or less than 20 years of service—higher risk; honorable discharge—lower risk) and mental health factors (alcohol related outpatient visits, mental disorder inpatient admission and suicidal ideation while in service—all associated with higher risk). However, the researchers stress that none of these factors on their own can meaningfully predict suicide, which is why they are using AI<\/abbr> to look at complex combinations of factors, and a factor being associated with a higher risk of suicide does not mean that it is specifically causing the suicidal behavior.<\/p>\n

Researchers analyzed records of more than 800,000 U.S. Army soldiers who left the service between 2010 and 2019. The study, conducted from March 2023 to March 2024, developed a machine learning model for suicides occurring up to a decade after leaving active duty.<\/p>\n

By December 31, 2019, the cohort had experienced 2,084 suicides. The model\u2019s predictive accuracy varied over time. It performed best for the first month post-service, with decreasing, but still significant, accuracy over a 120-month (10-year) period.<\/p>\n

According to the 2023 National Veteran Suicide Prevention Annual Report<\/a> released by the U.S. Department of Veterans Affairs, in 2021, the suicide rate for veterans was 71.8% higher than for non-veteran adults, after adjusting for age and sex differences.<\/p>\n

Other authors on this study are from a variety of institutions, including Harvard University, Columbia University, University of Pennsylvania, Massachusetts General Hospital and more.<\/p>\n

The Department of Psychology is housed in the 糖心视频<\/abbr> M\u0101noa College of Social Sciences<\/a>.<\/p>\n

In a separate study out of 糖心视频<\/abbr> Hilo in 2023, researchers performed a historical comparison of U.S. Army and civilian suicide rates from 1900 to 2020, and discovered that societal factors are driving both military and civilian suicides. See this 糖心视频<\/abbr> News<\/em> story for more<\/a>.<\/p>\n","protected":false},"excerpt":{"rendered":"

In its testing, the model identified a high-risk group comprising 10% of soldiers.<\/p>\n","protected":false},"author":16,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[30],"tags":[301,1467,1363,637,1600,158,1026,9],"class_list":["post-208983","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-research","tag-college-of-social-sciences","tag-manoa-excellence-in-research","tag-manoa-research","tag-psychology","tag-public-impact-research","tag-publication","tag-social-science","tag-uh-manoa","entry","has-media"],"aioseo_notices":[],"aioseo_head":"\n\t\t\n\t\n\t\n\t\n\t\n\t\n\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t