Abstract: Conjoint Trajectory Patterns of Tobacco and Alcohol Use throughout Emerging Adulthood (Society for Prevention Research 22nd Annual Meeting)

106 Conjoint Trajectory Patterns of Tobacco and Alcohol Use throughout Emerging Adulthood

Schedule:
Wednesday, May 28, 2014
Congressional D (Hyatt Regency Washington)
* noted as presenting author
Jessica Duncan Cance, MPH, PhD, Assistant Professor, The University of Texas at Austin, Austin, TX
Anna E. Talley, BA, Graduate Research Assistant, University of Texas at Austin, Austin, TX
Kim Fromme, PhD, Professor, University of Texas at Austin, Austin, TX
The concurrent use of tobacco and alcohol has a multiplicative effect on social and physical consequences for both society and the individual. While several studies have demonstrated that tobacco and alcohol use are strongly correlated in emerging adulthood, we know little about the prospective co-occurrence of patterns of use within this age group. The current study explores these conjoint trajectories, as well as the associated sociodemographic factors.

Results are from emerging adults entering a large university (N=2,247; 60% female; 55% White; aged 17-19 at baseline) and surveyed over a period of six years (10 waves total). Alcohol use was assessed with a factor score based on number of drinks per day, frequency of drinking days, binge drinking, and intoxication during a typical week within the past three months. Tobacco use was assessed by the frequency of using any tobacco product in the past three months (frequency, 0=”never” to 5=”daily”). Sociodemographics included gender, race/ethnicity, fraternity/sorority (i.e., Greek) membership, residence during the first semester (i.e., living on- or off-campus), parent education, and family income.

Latent curve growth analysis in Mplus 7.11 supported four trajectory classes of alcohol use: Normative low use (47%), Mid-use, gradually increasing (30%), Mid-use, late declining (17%), and Heavy use (6%). Five classes for tobacco frequency also emerged: Non-use (72%), Developmentally limited use (12%), Rapidly desisting use (6%), Steady medium use (7%), and Steady frequent use (3%). Estimation of conjoint trajectories via sequential process LCGA classified the largest percentage of participants into the non-smoking, low alcohol use group (42%), while less than 1% were classified in the low alcohol/heavy smoker joint class. There was more variation in smoking trajectories among alcohol users than variation in alcohol trajectories among smokers.

Multinomial logistic regression in SPSS 21.0 indicated that White race and Greek affiliation were significantly associated with being in riskier smoking and alcohol classes but  male gender, family income, and parent education were only associated with alcohol class membership. While higher family income increased the risk of belonging to either mid use class (alcohol classes 2 and 3) compared with the normative class, higher parent education decreased the odds of belonging to these classes.

Findings highlight the diversity of tobacco and alcohol use behaviors across emerging adulthood and the clear association between alcohol and tobacco use. Efforts to prevent or curb concurrent use of these substances might focus more on the high-risk drinking behaviors of smokers  than on the smoking behaviors of high-risk drinkers.