ADHD Presentation and Alcohol Use Among Juvenile Offenders: A Group-Based Trajectory Modeling Approach

ABSTRACT

While research has consistently identified attention deficit hyperactivity disorder (ADHD) as a risk factor for alcohol use in adolescence, there has yet to be an examination of how the multiple presentations of the disorder may differentially predict alcohol use. Some have posited that the individual symptom clusters of attention deficit hyperactivity disorder may actually represent distinct and unrelated disorders, providing impetus for examining whether or not they differentially predict alcohol use. Low self-control has consistently been found to be a predictor of alcohol use; therefore it is expected to be the driving force behind ADHD's predictive power for understanding alcohol use. Using the Pathways to Desistance data, this research sought to examine the relevance of this symptom cluster for understanding alcohol use among juvenile offenders. Utilizing group-based trajectory modeling, six alcohol use trajectories across adolescence were identified among the sample of juvenile offenders. Multinomial logistic regression indicated that Predominantly Inattentive Presentation predicted membership to the high chronic drinking pattern, while Predominantly Hyperactive-Impulsive Presentation did not predict membership to any trajectory. The Combined Presentation was found to be the best predictor of membership to the high chronic drinking group. These findings provide impetus for further exploration of the inattentive symptom cluster so that mechanisms by which it affects alcohol use may be better understood.

KEYWORDS:

• ADHD
• alcohol
• group-based trajectory modeling
• juvenile offenders

Note

Notes

1. Preliminary analyses indicated that outlier values were leading to the extraction of groups which were problematic for analyses. Because of this, a ceiling value of 100 drinks consumed in the previous observation period was coded for this variable. The full methodology employed for choosing 100 as the top code for the outcome variable can be found in Wojciechowski's (forthcoming 2017) research.