Prevention and prediction of adolescent problem behavior

Abstract

Background: The adolescent period is characterized by a huge leap in social, emotional, and cognitive development. For most people, it passes without major disruptions, but for a substantial number of adolescents, it does not. There is a range of social and psychological theories that aim to explain the development of problem behaviors, where some have a holistic approach and some focuses on individual features. Several strategies for preventing or decreasing problem behaviors have been developed, often by addressing identified risk factors that increase the likelihood of a negative development. Such prevention include for instance parent training programs, and a change in parenting skills has been shown to be a strong predictor for child problem behavior. Two parent training programs; ParentSteps (FöräldraStegen) and Comet for parents of adolescents aged 12-18 (ungdomsKomet) that address parents of adolescents with problem behaviors have been run since year 2004 within the regular preventive work of Swedish social services. Within this thesis, the two programs are evaluated for the first time. The aims of the three studies’ within the present thesis were to1) study the effectiveness of the two parent training programs that address parents of adolescents with at-risk behaviors with regard to the programs’ ability to decrease the adolescents’ parent- and self-rated antisocial behavior, delinquency, substance use, and psychosocial distress; 2) explore both self-rated and parent-rated short-term predictors of hazardous alcohol use (externalizing behavior, delinquency, internalizing behavior, psychosocial distress, perception of peers’ deviancy, perception of peers’ drinking) among alcohol using adolescents; 3) explore the ability of self-rated psychological risk factors to predict externalizing behavior or delinquency among female adolescents, and if perception of peers’ deviancy moderated any such effect.

Methods: A naturalistic inclusion sample comprised parents and adolescents aged 12-18 years, who were assessed in a randomized controlled trial, at baseline and at a 6-month follow-up. Study 1 consisted of 243 parents and 237 adolescents, who were randomized to ParentSteps, Comet or to a wait-list control group. The parent group training was carried out in real-world settings within the regular preventive work of social services. Study 2 comprised 167 alcohol-using adolescents and their parents, and Study 3 comprised 112 female adolescents. Initially, prior to study inclusion, parents were briefly screened for eligibility with questions about their adolescent’s behaviors, and participation required informed and written consent from both parent and adolescent. Participants also had to live in on of the five collaborating municipalities in Stockholm county. Data collection was carried out between fall 2008 and spring 2010.

Results: We found no significant effects from the programs over time with regard to reduction of parent- and adolescent-rated antisocial behaviors, delinquency, substance use or psychosocial distress. Of the tested predictors only perception of peers’ drinking had an effect on both females’ and males’ heavy episodic drinking and risk use of alcohol, and externalizing behaviors predicted females’ risk use. There were no significant differences in predictor slopes between females and males. Parent-rated externalizing behavior predicted males’ risk use. Among the psychological risk factors, cognitive problems emerged as a significant predictor for the females’ externalizing behaviors and delinquency. Perception of peers’ deviancy did not moderate the effect.

Conclusion: With a focus on clinical applicability, it can be concluded that ParentSteps and Comet were not suitable for this population of at-risk adolescents. The included adolescents were found to have elevated, sometimes clinical, levels of several co-existing problem behaviors, suggesting that adolescents who together with their parent seek parent training should be thoroughly examined for co-existing problems so they can be assigned to proper intervention or treatment. Future research should explore both females’ and males’ ADHD symptoms as risk factors for externalizing behaviors, delinquency, illicit drug use, and hazardous alcohol use in longitudinal studies with longer follow-up periods than 6 months, preferably in larger samples.