Factors associated with current, frequent, and multi-substance use among adult ever-users: a cross-sectional analysis of the 2024 National Survey on Drug Use and Health

Abstract Aim Substance use is complex, often involving multiple products, variable frequency, and different levels of intensity. Dichotomizing substance use into product-specific user and nonuser categories fails to adequately capture such complexity. This study investigates factors associated with current, frequent, and multi-substance use patterns. Subject and methods Analyses involved 24,830 adults from the 2024 National Survey of Drug Use and Health (NSDUH) who had tried at least one of seven selected illicit substances. Logistic regression models assessed the likelihood of current, frequent, and multi-substance use. A multinomial logistic regression model assessed the likelihood of five mutually exclusive levels of substance use combining current use, frequency, and multi-substance use. Results Current use of at least one of the seven illicit substances was reported by 31.7% of the participants, with 50.1% reporting frequent use and 6.1% multi-substance use. Although marijuana was the most commonly used substance (91.1% exclusive use), frequent use was most prevalent among heroin users (81.0%) versus marijuana (49.4%). Age at first use (OR ≤ 0.94, p ≤ 0.001) was associated with lower odds of current, frequent, and multi-substance use. Higher perceived risk of marijuana use was associated with lower odds of current and frequent use (e.g., OR ≤ 0.54, p < 0.001 for current use and OR ≤ 0.75, p ≤ 0.003 for frequent use). Current tobacco use was associated with higher odds of all three outcomes (OR ≥ 1.52; p < 0.001). Compared to younger users, older users were less likely to report infrequent use without multi-substance versus no current use (relative risk ratio [RRR] = 0.45, p < 0.001). Conclusion More than half of current users reported frequent use. Overall, multi-substance use was low but varied significantly across sociodemographic characteristics and cognitive factors.