Symposia
LGBQT+
Nicholas Livingston, Ph.D.
National Center for PTSD
Westwood, Massachusetts
Kristine Lynch, PhD
Epidemiologist
VA Informatics and Computing Infrastructure
SALT LAKE CITY, Utah
Elise Gatsby, MPH
Analyst
VA Informatics and Computing Infrastructure
SALT LAKE CITY, Utah
Jillian Shipherd, PhD
Research psychologist
VA Central Office
BOSTON, Massachusetts
Scott DuVall, PhD
Director
VA Informatics and Computing Infrastructure
SALT LAKE CITY, Utah
Emily Williams, PhD, MPH
investigator
Center for innovation for Veteran-Centered and Value-Driven Care
seattle, Washington
Background: Unhealthy alcohol use is a leading cause of death in the U.S. (Mokdad et al., 2018). While it is clear that veterans (Hoggatt et al., 2021) and individuals with minoritized sexual orientations (MSO; Shipherd et al., 2021) experience disproportionate burden of alcohol-related harms, it remains unclear whether there is disproportionate risk at the intersection of these identities. To estimate the magnitude of alcohol-related mortality and disparity, we applied novel methods to examine rates of alcohol attributable death (AAD) and years of potential life lost (YPLL) per AAD among all veterans and separately among MSO veteran men and women. Since sexual orientation data is not routinely collected in Veterans Health Administration (VA), we used natural language processing to derive our MSO cohort from provider reports within patient medical records.
Method: We used VA electronic medical record data from 1999 to 2019. Our analytic sample included 102,085 MSO and 5,300,521 non-MSO veteran patients with a valid alcohol use screening between 2014-2018, used for descriptive purposes and linkage to National Death Index data. We adapted the CDC’s Alcohol-Related Disease Impact (ARDI) application (CDC, 2020) to qualify acute and chronic AAD, and prevalence across MSO and non-MSO veteran men and women. We then compared YPLL per AAD as a function of expected life expectancy across groups.
Results: Our sample included 34,172 MSO and 416,820 non-MSO women, and 67,913 MSO and 4,883,701 non-MSO men. Overall, 55% of MSO vs 56% non-MSO veterans reported any past year drinking. However, among veterans who drank, drinking level and severity were higher among MSO veterans. The standardized mortality ratio for AAD among MSO men was 1.82 compared to non-MSO men, and 3.01 among MSO women compared to non-MSO women. Average YPLL for all non-MSO veterans was 19 years versus 24 for MSO men and 33 for MSO women.
Conclusions: The prevalence of any past year drinking was similar for all veterans, but among veterans who drank, drinking severity was higher among MSO men and women. Rates of AAD were high for veterans overall but they were highest among MSO veterans. The YPLL per AAD was highest among MSO and non-MSO women, then MSO men, relative to non-MSO veterans. While the direction of these disparities was anticipated, the magnitude of the differences, particularly for MSO veteran women, far exceeded expectations and impress the exigent need for further research, outreach, and services for MSO veterans and women.