Purpose: To investigate change across development in two smoking outcomes (smoking status and rate), describe demographic differences in smoking, and longitudinally examine the effects of psychosocial variables on smoking (psychological distress, victimization, and social support) in lesbian, gay, bisexual, and transgender (LGBT) youth.

Methods: Participants were 248 ethnically diverse LGBT youth (ages 16-20 years at baseline) from a longitudinal cohort study with six waves over 3.5 years. Baseline questionnaires included demographic variables and a measure of impulsivity, and longitudinal questionnaires included measures of cigarette smoking (status and average number of cigarettes smoked daily), LGBT-based victimization, psychological distress, and perceived social support. Analyses were conducted with hierarchical linear modeling.

Results: Males had higher odds of smoking and smoking rate than females, but females' smoking rate increased more rapidly over time. Psychological distress was associated with higher odds of smoking and smoking rate at the same wave, and it predicted smoking rate at the subsequent wave. LGBT victimization was associated with higher odds of smoking at the same wave and predicted smoking rate at the subsequent wave. Finally, significant other support predicted higher odds of smoking and smoking rate at the subsequent wave, but family support was negatively correlated with smoking rate at the same wave.

Conclusions: There are several viable avenues for the development of smoking prevention interventions for LGBT youth. To optimize the efficacy of prevention strategies, we must consider experiences with victimization, the impact of psychological distress, and optimizing support from families and romantic partners.