Trends in Marijuana Use Among Pregnant and Nonpregnant Reproductive-Aged Women, 2002-2014

Qiana L. Brown, Aaron L. Sarvet, Dvora Shmulewitz

The Columbia University Medical Center institutional review board waived review of this study. Informed oral consent was obtained from each participant. Data from women aged 18 through 44 years from the annual National Survey on Drug Use and Health (NSDUH) from 2002 through 2014 were analyzed. The surveys used in-person audio computer-assisted self-interviews (ACASI) about substance use and other behaviors in nationally representative samples of the noninstitutionalized US population; average response rates since 2002 were 75%.⁵ Among participants reporting lifetime use of marijuana or hashish, recency of use was assessed with the question: “How long has it been since you last used marijuana or hashish?” Responses included “within the past 30 days,” “more than 30 days ago but within the past 12 months,” and “more than 12 months ago.”⁵ Among pregnant and nonpregnant women, log-Poisson regression (SUDAAN [RTI International], version 11.0.1) was used to estimate and test trends in the adjusted prevalences of past-month and past-year marijuana use over time, controlling for complex survey design, age, race/ethnicity, family income, and education. Differences in trends over time were examined by pregnancy status and age (18-25 years and 26-44 years). Results were considered statistically significant at a P value of less than .05 (2-sided).

Of the 200 510 women analyzed, 29.5% were aged 18 through 25 years and 70.5% were aged 26 through 44 years; 61.0% were white, 13.7% black, 17.2% Hispanic, and 8.1% other race/ethnicity; 59.2% had some college education; 55.9% had annual family incomes less than $50 000; and 5.3% (n = 10 587) were pregnant.

Among all pregnant women, the adjusted prevalence of past-month marijuana use increased from 2.37% (95% CI, 1.85%-3.04%) in 2002 to 3.85% (95% CI, 2.87%-5.18%) in 2014 (prevalence ratio [PR], 1.62 [95% CI, 1.09-2.43]). The adjusted prevalence of past-month marijuana use was highest among those aged 18 to 25 years, reaching 7.47% (95% CI, 4.67%-11.93%) in 2014, significantly higher (P = .02) than among those aged 26 to 44 years (2.12% [95% CI, 0.74%-6.09%]). However, increases over time did not differ by age (P = .76). Past-year use was higher overall, reaching 11.63% (95% CI, 9.78%-13.82%) in 2014, with similar trends over time.

In nonpregnant women, prevalences of past-month use (2014: 9.27% [95% CI, 8.90%-9.65%]) and past-year use (2014: 15.93% [95% CI, 15.48%-16.40%]) were higher overall, with similar trends over time. Increases over time in past-month marijuana use did not differ by pregnancy status (P = .64).

Among pregnant women, the prevalence of past-month marijuana use increased 62% from 2002 through 2014. Prevalence was highest among women aged 18 to 25 years, indicating that young women are at greater risk for prenatal marijuana use. Study limitations are noted. Self-reported marijuana use may lead to underreporting due to social desirability and recall biases. However, use of ACASI helps reduce such biases,⁵ and the increases over time observed in this study are consistent with increases over time in marijuana-related outcomes shown in other studies that did not rely on self-reports, supporting the validity of the findings.⁶ Additionally, future studies should address dose, frequency of use, and clinical outcomes.

These results offer an important step toward understanding trends in marijuana use among women of reproductive age. Although the prevalence of past-month use among pregnant women (3.85%) is not high, the increases over time and potential adverse consequences of prenatal marijuana exposure³ suggest further monitoring and research are warranted. To ensure optimal maternal and child health, practitioners should screen and counsel pregnant women and women contemplating pregnancy about prenatal marijuana use.

Corresponding Author: Deborah S. Hasin, PhD, Department of Psychiatry, Columbia University Medical Center, 1051 Riverside Dr, No. 123, New York, NY 10032 (deborah.hasin@gmail.com).

In addition to sociodemographic and mental and physical health characteristics, NSDUH collected data on substance use and use disorders, age of onset for each specific substance used, perceived risk of harm from marijuana use, perceived legalization of medical marijuana use in residing state, and perceived marijuana availability. To classify medical marijuana use, those reporting past-year marijuana use were asked if any marijuana use was recommended by health care professionals and, if yes, whether all marijuana use was recommended.

We estimated the 12-month prevalence of medical marijuana use only, nonmedical marijuana use only, and combined medical and nonmedical use (combined use; 2-sided t test with a significance level of .05). We used multinomial logistic regressions to examine characteristics distinguishing the 3 groups. Our analyses used SUDAAN software (RTI International), version 11.0.1, to account for the complex sample design and sampling weights of NSDUH data.

Based on 96 100 respondents, 12.9% (95% CI, 12.6%-13.2%) of US adults had past-year marijuana use (nonmedical use only, 11.6% [95% CI, 11.3%-11.8%], medical use only, 0.8% [95% CI, 0.7%-0.9%], combined use, 0.5% [95% CI, 0.4%-0.5%]). Among past-year adult marijuana users, 90.2% (95% CI, 89.5%-91.0%) used nonmedically only, 6.2% (95% CI, 5.6%-6.9%) used medically only, and 3.6% (95% CI, 3.1%-4.0%) used medically and nonmedically. Of medical marijuana users, 78.8% (95% CI, 75.7%-81.9%) resided in states where medical marijuana was legal, and 21.2% (95% CI, 18.1%-24.3%) resided in other states.

Prevalence patterns among adults were similar across medical use only, nonmedical use only, and combined user groups with few exceptions  compared with the West region, medical use only and combined use was less common in other regions, and nonmedical use only was more common in the Northeast. Medical use only was more common among those reporting fair or poor health than better health and among those with stroke; the opposite was found for nonmedical use only. Compared with full-time employed adults, nonmedical use only was less common and medical use only was more common among disabled adults.

Compared with nonmedical use only, medical use only was directly associated with older age, older marijuana initiation age, disability, Medicaid status, stroke diagnosis, poor self-rated health, anxiety disorder, daily or near daily marijuana use, residing in a medical marijuana legalization state, and perceived state legalization of medical marijuana, but was inversely associated with heavy alcohol use and nonmedical use of prescription stimulants and analgesics.

Using nationally representative data, 9.8% of adult marijuana users in the United States reported use for medical purposes. Although the prevalence of medical use was higher in states that had legalized medical marijuana, 21.2% of medical marijuana users resided in states that had not, suggesting physicians might recommend medical marijuana use regardless of legalization.³

Similarities in correlates of medical and nonmedical users, especially co-occurrence with psychiatric conditions and other substance use, suggest that some marijuana users may access medical marijuana without medical need.⁵However, medical-only marijuana users differed from nonmedical-only users in ways that are consistent with use to address medical problems.⁶ Limitations of this study include lower response rates compared with prior years, which increases the potential for nonresponse bias, and limited questions about medical marijuana use.

Corresponding Author: Wilson M. Compton, MD, MPE, National Institute on Drug Abuse, 6001 Executive Blvd, MSC 9589, Bethesda, MD 20892-9589 (wcompton@nida.nih.gov).