ABSTRACT
BACKGROUND: Cigarette smoking escalates most in early to middle young adulthood. However, little research has examined a range of multilevel factors in relation to smoking trajectories during this time. METHODS: We examined: 1) trajectories of cigarette smoking among 2967 US college students (aged 18-25) in a two-year, six-wave longitudinal study (using growth mixture modeling); and 2) intrapersonal- (i.e., other substance use, depressive symptoms, ADHD symptoms,); interpersonal- (i.e., adverse childhood events, social support, parental tobacco and marijuana use), and community-level (i.e., type of college, rural vs. urban setting) predictors of differing trajectories (using multinomial logistic regression). RESULTS: We identified three trajectory classes: 1) Dabblers, who used cigarettes at one point in their life or not at all (85.6%); 2) College Onset Smokers, who began smoking regularly during the college years (6.2%); and 3) Later Onset Smokers, who began smoking during the mid- to late-20â¯s (8.2%). Multinomial regression (with Dabblers as the reference group) showed that predictors of being College Onset Smokers included being male (pâ¯=â¯.031); Asian (pâ¯=â¯.001) but not Black (pâ¯=â¯.008; Ref: White); early onset smokers (i.e., initiation before age 15; pâ¯=â¯.006); past 30-day users of little cigars/cigarillos (pâ¯=â¯.024), alcohol (pâ¯<â¯.001), and marijuana (pâ¯=â¯.008); children of tobacco users (pâ¯=â¯.050); and public (pâ¯=â¯.031) or a technical college students (pâ¯<â¯.001; Ref: private college); predictors of being Later Onset Smokers were being male (pâ¯=â¯.019) and technical college students (pâ¯=â¯.005). CONCLUSIONS: Despite some young adults' smoking initiating/escalating in middle young adulthood, few risk factors were documented. This understudied period warrants greater examination to inform intervention.