This study used prospective data from the Avon Longitudinal Study of Parents and Children (ALSPAC), a population-based UK cohort, to characterize longitudinal patterns of substance use during pregnancy. Alcohol consumption and cigarette smoking were assessed across all three trimesters and categorized into four longitudinal patterns: abstained, persistent, intermittent, and discontinued. Modeling patterns across trimesters allowed the investigators to move beyond a binary exposed/unexposed approach and to examine whether persistence or cessation of use during pregnancy relates differentially to maternal mental health.
Maternal depressive and anxiety symptoms were measured using the Crown-Crisp Experiential Index. Symptom assessments spanned from the second trimester of pregnancy through 21 months postpartum, enabling analysis of symptom trajectories across the perinatal period. The longitudinal design permitted evaluation of both mean symptom levels associated with different substance use patterns and the rate of symptom change over time.
The analytic sample derives from ALSPAC. Ethical approval for the original cohort was obtained from the ALSPAC Ethics and Law Committee and Local Research Ethics Committees; secondary analyses reported here were approved with oversight by the Mass General Brigham Institutional Review Boards (Protocol 2023P000011). Informed consent procedures for ALSPAC participants followed committee recommendations; questionnaire completion was considered written consent and data were anonymized. The authors note that ALSPAC data are not publicly available in a third-party repository but can be requested from the ALSPAC Executive Committee for researchers who meet access criteria.
When prenatal alcohol use was modeled by pattern, women with persistent alcohol use across pregnancy had higher depressive and anxiety symptom levels compared with abstainers. In addition to higher baseline symptom levels, persistent alcohol use was associated with slower declines in depressive and anxiety symptoms over time across the perinatal period. By contrast, women who exhibited intermittent alcohol use or those who discontinued alcohol during pregnancy had symptom trajectories that were similar to women who abstained, suggesting that it is the persistence of alcohol exposure through pregnancy that is most strongly associated with elevated and sustained maternal psychological distress in this cohort.
Patterns of prenatal cigarette smoking showed a different relationship with maternal mental health. Persistent, intermittent, and discontinued smoking patterns were each associated with higher overall depressive and anxiety symptom levels relative to abstainers. Unlike the alcohol findings, smoking patterns other than abstention did not converge with abstainers’ symptom profiles; instead, elevated symptoms were observed across persistent, intermittent, and quitting patterns. Notably, persistent smoking was additionally associated with steeper declines in symptoms over time, indicating that although persistent smokers had higher symptom levels initially, their symptoms decreased faster than in some other groups over the follow-up period.
The findings emphasize the value of considering persistence and cessation of substance use during pregnancy rather than treating prenatal exposure as a simple binary. For alcohol, persistence through pregnancy marked women with both higher symptom burden and slower recovery over the first two postpartum years, identifying a group who might benefit from targeted mental health screening and early intervention. For smoking, elevated symptom levels across persistent, intermittent, and quitting patterns suggest that any smoking during pregnancy may flag increased risk for perinatal psychological distress, though trajectories differed by pattern.
These differential associations suggest that perinatal screening protocols and prevention efforts could be refined by incorporating information on substance use patterns (persistent vs intermittent vs discontinued) to identify women at higher risk for sustained depressive or anxiety symptoms. The authors propose that persistent substance use during pregnancy may serve as an actionable risk marker to inform early support and referral.
This report is a preprint posted on medRxiv and has not undergone peer review; the authors explicitly note that the findings should not yet guide clinical practice. The article provides information on ethics approvals and funding sources, including NIH grants and the MQ Foundation. Data access is controlled by ALSPAC and available by request. The article does not provide trial registration identifiers because it reports secondary analyses of cohort data. Readers should interpret the associations described here in the context of observational cohort data and await peer-reviewed confirmation.
Overall, this analysis from ALSPAC supports nuanced consideration of alcohol and cigarette use patterns across pregnancy when assessing maternal mental health risk across the perinatal period. Persistent alcohol use appears particularly marked by higher and more persistent depressive and anxiety symptoms, while any smoking during pregnancy was associated with higher symptom levels, with varying symptom trajectories by pattern of use. These distinctions may inform future screening strategies and intervention priorities for perinatal mental health services.