Abstract
BACKGROUND: Polysubstance use is associated with adverse health outcomes, yet little research has measured changes in polysubstance use. We aimed to 1) estimate trends in marijuana and heavy alcohol use by cigarette smoking and demographic subgroups, and 2) examine patient factors associated with concurrent use among adults who were smoking. METHODS: We conducted a repeated cross-sectional analysis of 687,225 non-institutionalized US adults ≥18 years from the 2002-2019 National Survey on Drug Use and Health. Participants were stratified into current, former, and never smoking groups. Main outcomes were prevalence of heavy alcohol use, marijuana use, and concurrent use of both substances. RESULTS: From 2002-2019, heavy alcohol use declined from 7.8 % to 6.4 %, marijuana use rose from 6.0 % to 11.8 %, and concurrent use of alcohol and marijuana remained stable. Among adults who were smoking from 2005 to 2019, higher education was associated with higher odds of heavy alcohol use, while older ages, female gender, non-White race/ethnicity, and government-provided health insurance were associated with lower odds. The odds of marijuana use decreased in females, older ages, and higher incomes while increasing in people with poorer health status, higher education, government-provided or no health insurance, and serious mental illness. Compared to White adults who were smoking, Black counterparts had higher odds of marijuana use (OR=1.23; 95 %CI: 1.15-1.29), while Hispanic (OR=0.68; 95 %CI: 0.63-0.72) and other racial/ethnic identities (OR=0.83; 95 %CI: 0.77-0.90) had lower odds. CONCLUSIONS: Our study suggests marijuana use might not be sensitive to changes in the use of tobacco and alcohol.