Comparing Suicide Rates Across Subpopulations of Veterans Receiving Veterans Health Administration Care, 2017-2021: The Suicide Risk Encyclopedia

Adm Policy Ment Health. 2026 May 25. doi: 10.1007/s10488-026-01510-0. Online ahead of print.

ABSTRACT

This study presents suicide surveillance findings from the Veterans Health Administration (VHA) Suicide Risk Encyclopedia, a novel population-based suicide rate comparison tool, demonstrating which Veteran VHA patient subpopulations experience the most elevated suicide rates. For cohorts of Veteran VHA patients alive as of the last day of 2016, 2017, 2018, 2019, and 2020 and with VHA encounters in the prior 2 years, we assessed subpopulation size, number of suicide deaths in the subsequent calendar year, and the suicide rate for 482 measures of Veteran characteristics, diagnoses, and services utilization. We compared suicide rates across measures and years and rank-ordered subgroups by suicide rates. For subgroups with at least 20 suicides from 2017 to 2021, the measures with the highest suicide rates were: prior-year suicide attempts (suicide rate: 380.9/100,000 person-years), high-risk Veterans Crisis Line (VCL) calls (i.e., calls rated as high risk by the VCL responder; 354.1/100,000 person-years), moderate-to-high-risk VCL calls (276.6/100,000 person-years), VHA suicide high risk flag receipt (263.8/100,000 person-years), and suicidal ideation diagnostic codes (257.5/100,000 person-years). For measures with a prevalence of at least 10%, the measures with the top suicide rates were alcohol use disorder diagnoses (99.8/100,000 person-years), anxiolytic prescriptions (87.9/100,000 person-years), non-psychotherapy and medication management mental health outpatient encounters (e.g., 82.7/100,000 person-years), unspecified anxiety disorder diagnoses (75.0/100,000 person-years), and other-specified depression diagnoses (73.5/100,000 person-years). Findings identify high risk VHA patient subpopulations, provide information on trends, and include data regarding subpopulation sizes. These findings highlight the potential value of health systems developing population-based risk comparison tools.

PMID:42183954 | DOI:10.1007/s10488-026-01510-0