Instrumental and Partner Support As Predictors of Illicit Drug Use Among Urban Mothers: A Longitudinal Analysis

Background and Purpose:

Mothers in low-income urban families face disproportionate exposure to economic instability, relationship instability, and social marginalization, conditions that heighten vulnerability to substance use. Social support has been theorized as a key protective factor under chronic stress, yet longitudinal evidence on how distinct dimensions of support shape drug use risk remains limited, particularly among low-income, racially diverse urban mothers. We examined whether instrumental support and partner support predict maternal illicit drug use over time in a large urban birth cohort, drawing on data from the Future Families and Child Wellbeing Study.

Methods:

Our analytic sample included mothers who participated across four waves of the Future Families and Child Wellbeing Study (at childbirth and when the child was ages 3, 5, and 9), a longitudinal panel study oversampling nonmarital births in large U.S. cities. Illicit drug use is a binary indicator of any past-12-month use of marijuana, cocaine, opioids, or other substances. Instrumental support was measured as a summary index of access to material and practical assistance. Partner support was assessed using a mean scale of relationship quality items capturing affection, encouragement, and fairness. We estimated Bayesian random-effects logistic regression models via the brms package to account for the non-independence of repeated observations across waves. We preferred the Bayesian approach for its ability to yield probabilistic credible intervals and greater stability with sparse binary outcome data. Adjusted models included time-varying covariates (maternal depression, economic hardship, partner drug use, relationship status, poverty ratio) and time-invariant sociodemographic characteristics (age, race/ethnicity, education, nativity, parity).

Results:

Our sample included 4,897 mothers with a mean age of 25.3 years at baseline (SD = 6.0); nearly half identified as Black non-Hispanic (47.6%), and approximately 62% lived below 200% of the federal poverty level at baseline. Prevalence of maternal illicit drug use was 5.5% at childbirth, 7.2% at child age 3, 5.4% at child age 5, and 9.1% at child age 9. In adjusted models, we found no significant association between instrumental support and illicit drug use (OR = 1.08, 95% CrI [0.93, 1.25]). Partner support was independently and negatively associated with drug use net of instrumental support and all covariates (OR = 0.63, 95% CrI [0.47, 0.85]), such that mothers reporting higher partner support had significantly lower odds of drug use. Maternal depression (OR = 4.03), partner drug use (OR = 3.28), and economic hardship (OR = 1.85) were the strongest predictors of illicit drug use across models.

Conclusions and Implications:

Our findings suggest that relational quality, rather than material support availability, is the operative protective dimension against illicit drug use among urban mothers. Notably, partner support and partner drug use operated as simultaneous and opposing influences, suggesting that the same relationship can be both a source of protection and risk. Intervention efforts may benefit from couple-based approaches that strengthen relational quality while directly addressing partner substance use, alongside mental health treatment and economic support within home visiting and maternal health programs serving low-income families.