Cannabis Surpasses Tobacco Among Pregnant Women: Trends and Predictors of Use in the U.S., 2015-2023

• In Event: Regulating Risk: State Policy and Health Outcomes Across Substance Use and Reproductive Care

Abstract

Background: The rapid expansion of cannabis legalization across the United States has intensified concerns about prenatal substance exposure, particularly as cannabis becomes more socially normalized and its risks during pregnancy are increasingly underestimated. While tobacco use in pregnancy has declined steadily over decades – the product of sustained public health investment and robust clinical guidance – emerging national evidence suggests that cannabis may be filling this void. The public health implications are significant: prenatal cannabis exposure has been associated with adverse neonatal and developmental outcomes, yet prevention infrastructure has not kept pace with the shifting landscape of use.

Purpose: This study is the first to comprehensively examine national trends in tobacco-only use, cannabis-only use, and co-use among pregnant individuals in the United States from 2015 to 2023, and to identify sociodemographic predictors of each pattern across policy-relevant periods.

Data: We use nationally representative data from the 2015-2023 National Survey on Drug Use and Health (NSDUH), restricted to pregnant individuals ages 15-44 (N=6,099).
Research Design and Methods: We estimate weighted prevalence of past-30-day tobacco-only use, cannabis-only use, and co-use overall and by sociodemographic characteristics across three periods: pre-pandemic (2015-2019), 2020, and post-pandemic recovery (2021-2023). Weighted multivariable logistic regression models for 2015-2019 and 2021-2023 identify sociodemographic predictors of each use pattern relative to non-use, with adjusted odds ratios and average marginal effects reported to quantify absolute percentage-point differences. Models for 2020 were not estimated due to sample size constraints (N=419).

Results: Tobacco-only use declined sharply across the study period, falling from 12.0% in 2015 to 3.9% in 2023. Cannabis-only use, by contrast, more than doubled from 2015 to 2019 (1.4% to 3.1%), spiked to 4.8% in 2020, and surged to 6.3% in 2022 – surpassing tobacco-only use for the first time – before declining modestly in 2023. Socioeconomic gradients were pronounced and persistent. During 2015-2019, lower educational attainment, unmarried status, and lower income were consistently associated with elevated tobacco-only use and co-use. Hispanic pregnant individuals were significantly less likely than non-Hispanic White individuals to engage in any form of use. Medicaid beneficiaries exhibited elevated tobacco-only risk, while in 2021-2023, pregnant individuals with a high school education or some college were significantly more likely than college-educated peers to use cannabis exclusively. Non-Hispanic Black individuals were less likely than non-Hispanic White individuals to exclusively use tobacco or co-use across periods.

Conclusion and Implications: Prenatal substance use patterns in the United States are undergoing a fundamental shift. As tobacco use continues its long decline, cannabis use has risen to historically high levels and, in recent years, has surpassed tobacco as the predominant substance used during pregnancy. These trends demand an urgent policy response: updated prenatal screening protocols, evidence-based clinical guidance, and public health messaging that accurately conveys the potential risks of prenatal cannabis exposure. Policymakers should integrate cannabis risk communication into existing tobacco prevention frameworks, attend carefully to the socioeconomic gradients that shape differential exposure, and reckon with the maternal and child health implications of continued cannabis retail expansion.

Author

• Joanne Constantin, State University of New York (SUNY) Downstate Health Sciences University
  Presenting Author