Substance use rates are elevated across all non-heterosexual identity groups, including queer and questioning youth

For years, research on substance use disparities among sexual minority populations focused on three groups: gay, lesbian, and bisexual individuals. This framing reflected the categories available in national surveys. As those surveys have expanded, so has the picture they reveal. A study published February 18, 2026 in the American Journal of Psychiatry draws on the 2023 National Survey on Drug Use and Health - the first nationally representative U.S. dataset to include sexual identity options beyond the traditional triad, and the first to ask respondents aged 12 to 17 about their identity - to examine substance use patterns across the fuller spectrum of sexual identities.

Who was studied and what was measured

More than 52,000 people participated in the 2023 NSDUH. The survey included sexual identity options for queer, pansexual, asexual, and "something else" in addition to lesbian, gay, bisexual, and heterosexual, and added a "still figuring it out" category for respondents uncertain of their identity. Substance use data covered cannabis, hallucinogens, inhalants, methamphetamine, cocaine, prescription drug misuse, and other controlled substances.

The research team, led by Kevin Yang, MD, resident physician in the Department of Psychiatry at UC San Diego School of Medicine, compared rates across all identity groups against heterosexual respondents.

Elevated rates across the board

Every non-heterosexual group showed significantly higher use of every substance category measured compared to heterosexual individuals. The elevation was not confined to one or two substances or one or two identity groups - it was consistent across the full dataset.

Within that broad pattern, specific groups showed particularly elevated risks for particular substances. Bisexual respondents had the highest odds of using inhalants, hallucinogens, cocaine, and methamphetamine compared to heterosexual peers. Gay and lesbian respondents also showed substantially higher rates across most substances. Queer, pansexual, and asexual individuals - groups largely absent from prior national research because they were not included as response categories - showed elevated rates comparable to or exceeding those seen in the gay/lesbian group for several substances.

"Many people - especially younger individuals - identify as queer, pansexual or asexual, or they use other terms, and some are actually still figuring out their identity," Yang said. "These are populations that we have been missing in prior research."

Adolescent data adds a new dimension

Because the 2023 NSDUH was the first nationally representative survey to ask 12-to-17-year-olds about sexual identity, the study provides baseline data on substance use among sexual minority youth that did not previously exist at this scale. Adolescent substance use is particularly concerning given that earlier initiation is associated with higher rates of dependence and worse long-term outcomes.

The fact that identity-based disparities are detectable at adolescent ages - before most individuals have had extended exposure to the minority stressors that accumulate over adulthood - suggests that the drivers of elevated substance use include experiences specific to adolescence: family rejection, school-based discrimination, and the stress of identity navigation in unsupportive environments.

The minority stress framework and its limits

Researchers typically explain substance use disparities in sexual minority populations through minority stress theory: discrimination, stigma, concealment, and victimization produce chronic stress that some individuals manage through substance use. The consistent elevation across all non-heterosexual groups in this study, including those whose identities are less socially recognized, is consistent with that framework.

But the study is cross-sectional - it captures a single point in time for each respondent, so causal relationships cannot be established. Whether substance use precedes, follows, or co-develops with identity-related stress cannot be determined from survey data alone. Longitudinal studies that track individuals from adolescence into adulthood would be needed to parse those sequences.

Self-reported substance use also tends to be underreported in surveys, and the degree of underreporting may differ across identity groups, potentially obscuring true disparities or creating artifactual ones. Nonetheless, the consistency of the findings across substances, identity categories, and age groups makes random error an unlikely explanation for the overall pattern.