All‐Cause and Suicide Mortality Among Lesbian, Gay, and Bisexual Veterans Who Utilize Care through the Veterans Health Administration

Abstract

Research Objective: Lesbian, gay, and bisexual (LGB) individuals have higher risk of suicide ideation and attempts than their heterosexual counterparts, but suicide data, and mortality data in general, about LGB populations are scant because sexual orientation is not included in mortality surveillance. Electronic health record (EHR) data provide opportunities to learn about mortality among LGB populations. This study investigated mortality among a cohort of Veterans with LGB sexual orientation documented within the Veterans Health Administration (VHA) EHR. Study Design: We conducted a retrospective cohort study based on documentation of LGB sexual orientation obtained through both natural language processing of clinical notes and extraction of structured administrative data for sexual orientation in the VHA EHR from October 1, 1999, to September 30, 2017. Mortality data were obtained from the National Death Index. Cause of death was defined by International Classification of Disease Tenth Revision codes. The ten top-ranked causes of death were compared between LGB Veterans and the general US population. Mortality rates were calculated for all-cause and suicide-related deaths. Person time was accumulated from first LGB documentation to either date of death or end of FY2017, whichever came first. Analyses were stratified by sex. Population Studied: LGB Veterans who utilize care through the VHA. Principal Findings: Throughout 18-year study period, there were 95,964 Veterans with at least one LGB documentation in the EHR (30% women, 61% men); of these, 12 277 died in the study period. Mortality follow-up averaged 4.5 years. All-cause mortality rate was 2323.4/100 000. Causes of mortality were similar to the US population; diseases of the circulatory system, cancer, and external injuries were the top three causes in both sexual minority male and female Veterans. Of the total deaths, 437 (3.6%) were due to suicide, and the average age at suicide death was 46.8 years. The suicide rate among LGB Veterans (82.7/100 000) was more than two times higher than rates observed for the general VHA population (37.7/100 000). Suicide rates among sexual minority male Veterans were 1.6 times higher than sexual minority female Veterans (95.01/100 000 vs 56.23/100 000, respectively). Conclusions: Findings corroborate previous limited research suggesting elevated rates of suicide among LGB individuals. These findings reflect mortality rates only for LGB Veterans who use the VHA and have documentation of sexual orientation recorded in the EHR. Implications for Policy or Practice: Suicide prevention is currently VHA's top clinical priority. Results indicate that LGB Veterans have greater risk of suicide death than the general VHA population. Currently, there are no evidence-based suicide prevention programs specific to sexual minority populations. Further research is needed to characterize the experience of LGB Veterans at risk for suicide and to determine how VA suicide prevention efforts reach vulnerable LGB Veterans. Tailoring to LGB Veterans may be necessary to enhance current treatment and prevention efforts. Implementing standardized data elements in the EHR that allow Veterans to self-report their sexual orientation, and fostering safe and affirming clinical environments in which they can do so, will be critical to facilitate intervention and health services research to address disparities in suicide.