Attachment, Self-Efficacy, and Mental Health As Organizing Dimensions of Cannabis Use Among College Students

Background and Objectives Emerging adulthood is a high-risk period for problematic cannabis use, with 36.5% of young adults aged 18–25 reporting past-year use. While existing research documents cannabis-related academic and health disruptions, few studies examine the relational and psychosocial contexts of use. This study draws from a framework based on attachment and self-efficacy theories to identify psychosocial profiles among college students who use cannabis. The primary objective was to characterize psychosocial heterogeneity and determine how these profiles relate to socioeconomic status (SES), race, sexual orientation, and Cannabis Use Disorder (CUD) symptoms.

Methods Data were drawn from a longitudinal cohort of undergraduate students (N = 513) at two large Midwestern universities. Inclusion criteria required past-month cannabis use (at least four occasions). Latent profile analysis (LPA) was conducted using nine continuous indicators: experiential avoidance (AAQ-II), emotion dysregulation (DERS), depressive symptoms (PHQ-8), generalized anxiety (GAD-7), life satisfaction (SWLS), self-compassion (SCS), attachment avoidance/anxiety (ECR-RS), and agency/hope (ASHS). A 3-step (R3STEP) multinomial logistic regression was applied to link predictors (SES, race, sexual orientation, and CUD symptoms) to class membership while accounting for classification uncertainty.

Results A three-class solution provided the optimal fit to the data (entropy = 0.878; BLRT p <.01), identifying distinct psychosocial profiles among the participants. The largest group, Class 1 (47%), reflected a Moderate Distress profile, with mid-range functioning across all domains. Class 2 (26%), the High Distress profile, was characterized by elevated experiential avoidance, emotion dysregulation, depressive symptoms, and attachment insecurity, alongside the lowest levels of protective resources in the sample. In contrast, Class 3 (27%), the Low Distress profile, represented an adaptive profile defined by high life satisfaction and self-compassion coupled with minimal psychological distress. Multinomial logistic regression revealed that lower SES (OR ≈ .89), non-White racial identity (OR ≈ .96), and minoritized sexual orientation (OR ≈ 0.98) significantly increased the likelihood of belonging to the High Distress class (p < .001). Total mental health diagnoses and CUD symptoms also reliably predicted membership in the High Distress and Moderate classes relative to the Low Distress group.

Conclusions and Implications Findings highlight that college-aged cannabis users are not a monolithic group but instead represent distinct psychosocial profiles. The strong association between High Distress group membership and minoritized identities (racial and sexual orientation) underscores the need for cannabis use interventions in university settings to center racial equity and social justice. By moving beyond stigmatizing language and focusing on attachment and self-efficacy, social work practitioners can develop tailored, identity-affirming prevention and harm-reduction strategies that address the specific needs of high-distress students.