Primary sources of health care among LGBTQ+ veterans: Findings from the Behavioral Risk Factor Surveillance System

OBJECTIVE: This study examined the primary source of health care between veterans with lesbian, gay, bisexual, queer and similar identities (LGBTQ+) and non‐LGBTQ+ veterans. DATA SOURCES AND STUDY SETTING: Veterans (N = 20,497) from 17 states who completed the CDC's Behavioral Risk Factor Surveillance System from 2016 to 2020, including the Sexual Orientation and Gender Identity and Health Care Access modules. STUDY DESIGN: We used survey‐weighted multiple logistic regression to estimate average marginal effects of the prevalence of utilization of Veteran's Health Administration (VHA)/military health care reported between LGBTQ+ and non‐LGBTQ+ veterans. Prevalence estimates were adjusted for age group, sex, race and ethnicity, marital status, educational attainment, employment status, survey year, and US state. DATA COLLECTION METHODS: Study data were gathered via computer‐assisted telephone interviews with probability‐based samples of adults aged 18 and over. Data are publicly available. PRINCIPAL FINDINGS: Overall, there was not a statistically significant difference in estimated adjusted prevalence of primary use of VHA/military health care between LGBTQ+ and non‐LGBTQ+ veterans (20% vs. 23%, respectively, p = 0.13). When examined by age group, LGBTQ+ veterans aged 34 and younger were significantly less likely to report primary use of VHA/military health care compared to non‐LGBTQ+ veterans (25% vs. 44%, respectively; p = 0.009). Similarly, in sex‐stratified analyses, fewer female LGBTQ+ veterans than female non‐LGBTQ+ veterans reported VHA/military health care as their primary source of care (13% vs. 29%, respectively, p = 0.003). Implications and limitations to these findings are discussed. CONCLUSIONS: Female and younger LGBTQ+ veterans appear far less likely to use VHA/military for health care compared to their cisgender, heterosexual peers; however, because of small sample sizes, estimates may be imprecise. Future research should corroborate these findings and identify potential reasons for these disparities.