Longitudinal Associations between Substance Use and Mental Health In Young Adulthood

Background and Purpose:

Although adolescent substance use has declined over the past few decades, it remains a public health issue as adolescence is a period of identity exploration, development, and choices that can impact adult health outcomes (Boer et al., 2024). While research supports an association between mood and anxiety disorders in adolescents who use substances, there is an increased need to understand the relationship of anxiety and depression symptoms in young people as it relates to substance use and substance use behavior of peers (Lubman et al., 2007). Research has supported the link between peer substance use and individual substance use, however, less is known about the impact of peer substance use as a moderator for mental health symptoms (Rodriguez-Ruiz & Espejo-Stiles, 2024; Marschall-Levesque et al., 2014; Siennick et al., 2016).

This study sought to answer the following questions: Does substance use predict depressive and anxiety symptoms across time for young adults at ages 21, 23, and 26? Does family bonding moderate the association between substance use and mental health symptoms (depressive and anxiety symptoms) over time? Does peer substance use moderate the association between young adults’ own substance use and their depressive and anxiety symptoms? Does the association between substance use and mental health symptoms differ for individuals with and without a history of incarceration by age 21?

Methods:

This study used data from the Community Youth Development Study (CYDS), a community-randomized trial of the Communities That Care (CTC) program. The baseline sample included 4,407 students. Web surveys were administered in 2014, 2016, and 2019 corresponding to ages 21, 23, and 26 (Kuklinski et al., 2021; Pandika et al., 2023).

Fixed-effects multilevel models were estimated to evaluate the associations between individual substance use, peer substance use, and mental health symptoms over time. Mental health symptoms were measured using the PHQ-9 and GAD-7. Interaction models tested moderation by family bonding, and multiple-group models assessed whether associations differed based on incarceration history. Gender and ethnicity were included as covariates.

Results:

Substance use predicted higher depressive (b = 0.106, p < .001) and anxiety symptoms (b = 0.143, p < .001) across all models. Family bonding functioned as a protective factor (b = –0.109, p < .001) through significant main effects but did not moderate the association between substance use and mental health outcomes. Peer substance use predicted higher symptom levels, although interaction effects were not significant. Incarceration history did not predict depressive or anxiety symptoms and did not impact the substance use mental health relationship.

Conclusions and Implications:

Findings highlight substance use and peer substance use as contributors to mental health symptoms in young adulthood. These findings underscored the protective role of family relationships. Although moderation effects were not supported, results point to the value of continued prevention and intervention efforts focused on substance use and relational supports. This underscores the need for social work research centered on the wellbeing of young people in relation to family and positive relationships to mitigate substance use and negative mental health outcomes.