Abstract Background Substance use by youth living with HIV (YLWH) is a concern, given potential interactions with virus-associated immune suppression and adverse effects on risk behaviors, neurocognition, and adherence. Self-report substance use measures provide efficient cost-effective assessments. Analyses describe self-reported substance use among YLWH and examine agreement with toxicology assays. Methods Seventy-eight youth age 18–24 years (87% male, 71% African–American) with behaviorally acquired HIV-1 infection and 55 uninfected youth completed the Alcohol, Smoking, and Substance Involvement Screening Test to assess drug use frequency, including tobacco, marijuana, cocaine, and alcohol, over the prior three months. Elisa-based toxicology assays were used to detect 27 substances in plasma. Chi-square tests compared substance use between YLWH and uninfected youth; Kappa statistics compared agreement between self-report and toxicology. Results YLWH reported marijuana (49%), tobacco (56%), and alcohol (87%) use, with 20%, 28% and 3% reporting daily use of each substance, respectively; other substance use was uncommon. Uninfected youth reported less tobacco use but otherwise similar substance use. All youth who reported daily use of marijuana or tobacco had positive plasma toxicology results, while concordance decreased with less frequent self-reported use. Among youth reporting no substance use, few tested positive (4% YLWH, 2% uninfected youth for cannabis; 8%YLWH for tobacco). Conclusions Youth report high rates of marijuana, tobacco, and alcohol use. Concordance between self-report and toxicology for marijuana and tobacco use, particularly for daily users, supports self-report as a valid indicator of substance use in research studies of youth with or without HIV-1 infection.