Ziming Xuan

Background and Aims Analyzing long‐term trajectories of alcohol use has the potential to strengthen policy and intervention priorities and timing. We identified and described trajectories of alcohol use and binge drinking frequency from mid‐adolescence to early adulthood and measured the association of the role of early drinking initiation with trajectory membership. Design, Setting, Participants This was a longitudinal cohort study conducted in the United States. The National Longitudinal Survey of Youth 1997 is a nationally representative cohort of youth aged 12–16 years at baseline. The analytic sample included individuals who participated in two or more annual interviews between ages 15 and 30 years ( n = 8809). Measurements Participants self‐reported the number of days in the past 30 days they: (1) drank alcohol and (2) binge drank (five or more drinks on one occasion). We used group‐based trajectory modeling to identify distinct trajectories from ages 15–30 years of past 30‐day drinking and past 30‐day binge drinking. Using multinomial logistic regression, we evaluated associations between early drinking initiation (≤ 14 years) and key demographics with trajectory membership. Findings We identified five past 30‐day drinking groups: late‐escalating (16.0%), moderate frequency (19.0%), high frequency (11.2%), low frequency (35.4%) and no/infrequent (18.4%). Early drinking initiation (versus later) was associated with higher odds of membership in the moderate [adjusted multinomial odds ratio (aMOR) = 4.88; 95% confidence interval (CI) = 4.00, 5.94] and high‐frequency groups (aMOR = 4.68; 95% CI = 3.74, 5.86) than in the no/infrequent comparator trajectory. We identified five groups with distinct binge drinking frequency patterns: later escalating (9.9%), high frequency (3.9%), low frequency (28.7%), earlier onset (9.5%) and no/infrequent (48.0%). Early initiation was associated with increased odds of membership in earlier‐onset and high‐frequency groups compared with the no/infrequent group. For both outcomes, additional differences in probability of group membership were identified by gender, racial identity, parental factors (religiosity, high school completion) and household characteristics (household size, income, and region of residence). Conclusions Youth in the United States appear to follow heterogeneous drinking and binge drinking trajectories from adolescence into adulthood. These may include higher‐use trajectories as well as trajectories with different escalation timing (e.g. earlier versus later). Early initiation of drinking may increase risk of membership in higher‐ and earlier‐use trajectory groups.