Poster #51 - Sequence of Co-Occurring Social Anxiety and Antisocial Behavior during Early Adolescence: A Latent Transition Analysis

• In Event: 1-061 - Poster Session 02
  In Poster Session: PS 02 Section - Developmental Psychopathology

Integrative Statement

Studies have identified subgroups of youth with social anxiety who exhibit risk-prone, antisocial behaviors and/or who affiliate with antisocial peers (Cunningham et al., 2013; Kashdan et al., 2006; 2008; blinded). These youths are at greater risk for more severe psychopathology, compromised treatment, and impaired functioning than youth with social anxiety or who engage in antisocial behavior alone (Cunningham et al., 2013; Conner et al., 2008). Research that examines how constellations of co-occurring social anxiety, antisocial behaviors, and antisocial peer affiliation unfold over time is needed to inform preventative interventions in the progression and timing of co-occurring risk.

Using a subsample of 979 youth from the PROSPER project (Spoth et al., 2004), we used latent transition analysis (LTA; Collins & Lanza, 2010) to examine if and how youths’ constellations of social anxiety, antisocial behaviors, and antisocial peer affiliation change during early adolescence (6th- 9th grades). Adolescents reported on social anxiety symptoms (fear of negative evaluation, social avoidance), antisocial behaviors (conduct problems, lying), and antisocial peer affiliations (peer conduct problems, lying, substance use) during annual assessments (6th-9th grades; α’s .69 - .93).
To provide guidance during model selection with LTA, we estimated models with 1 to 7 latent profiles at each time point using latent profile analysis (LPA). Solutions selected at each time point consistently included 5 latent profiles with similar interpretations. A 5-profile solution was selected for the LTA based on model fit, parsimony, and consistency with previous research. Profiles of Low Risk Youth, Socially Anxious Youth, and Antisocial Youth emerged. A fourth group of Non-Anxious Deceiving Youth were socially confident and reported lying and affiliating with lying peers. A fifth group of Socially Anxious Youth with Antisocial Peers reported higher than average social anxiety symptoms and antisocial peer affiliation (Table 1).

Transition probabilities appear in Table 1. Across profiles, youth had the highest probability of being in the same profile at the subsequent time point (i.e., bolded values). Low Risk Youth who changed profiles were most, and increasingly, likely to be classified as Non-Anxious Deceiving Youth at the subsequent time point (19-36%). Socially Anxious Youth who changed profiles were most likely to be classified as Socially Anxious Youth with Antisocial Peers at the subsequent time point (22-29%). The highest probabilities of transitioning to Antisocial Youth were among Socially Anxious Youth with Antisocial Peers and Non-Anxious Deceiving Youth. The highest probabilities of transitioning to Low Risk Youth were for 6-7th and 7-8th grades among Socially Anxious Youth and Non-Anxious Deceiving Youth.

These results illustrate important sequencing in the progression of risk constellations during early adolescence. A noteworthy proportion of low risk and socially anxious youth go on to affiliate with antisocial peers and engage in deceptive behaviors one year later. In turn, these youths are most likely to engage in antisocial behavior during 7th to 9th grades. Findings suggest an important developmental window prior to 9th grade in which preventative interventions might reduce social anxiety and deceptive behavior and prevent progression towards co-occurring risk.

Authors

• Bridget Weymouth, The University of Alabama
  Presenting Author

• Greg Fosco, Pennsylvania State University
  Non-Presenting Author

• Bethany C Bray, The Pennsylvania State University
  Non-Presenting Author

• Mark Feinberg, The Pennsylvania State University
  Non-Presenting Author