Symposia
Research Methods and Statistics
Ki Eun (Kay) Shin, Ph.D.
Assistant Professor
Long Island University - Post
Brookville, New York
Michelle G. Newman, Ph.D.
Professor
Penn State University
University Park, Pennsylvania
Generalized anxiety disorder (GAD) has been linked to not only interpersonal dysfunction, but also interpersonal heterogeneity. GAD featured interpersonal subgroups, with individuals endorsing distinct types of interpersonal problems (e.g., Przeworski et al., 2012). Importantly, such distinct interpersonal problems predicted differential treatment outcomes in GAD. For instance, self-reported overly nurturant problems (e.g., putting others’ needs ahead of one’s own) predicted worse outcomes (Crits-Christoph et al., 2004) and higher anxiety at post and follow-up (Newman, Jacobson, Erickson, & Fisher, 2017). Self-reported intrusive (e.g., inserting oneself in other people’s business) and domineering problems (e.g., trying to change or control others) predicted better responses to behavior therapy than cognitive therapy or cognitive behavioral therapy for GAD (Newman et al., 2017). Despite the important clinical implications, there have been only two prior studies examining interpersonal subgroups in GAD (Przeworski et al., 2012; Salzer et al., 2008), and both studies had the limitations of a modest sample size and reliance on cluster analyses. The current study aimed to identify and compare interpersonal subgroups in GAD while improving upon prior studies (e.g., a larger sample, latent profile analysis vs. cluster analysis).
328 young adults with moderate to severe symptoms of GAD and up to three of their significant others rated the participants’ interpersonal problems on the Inventory of Interpersonal Problems-Short Circumplex (IIP-SC; Soldz et al., 1995). Scores from the IIP-SC were subjected to latent profile analysis, which led to six distinct interpersonal subgroups (domineering, exploitable, vindictive, nonassertive, overly nurturant, and undifferentiated). Based on multivariate analyses of variance (MANOVA) with Tukey’s post-hoc tests, exploitable and nonassertive groups showed greater social anxiety than domineering and undifferentiated groups and greater pathological worry than the undifferentiated group (ps < .05). Multilevel modeling will be used to test differences between the groups on significant others’ ratings to see whether the groups are distinguishable on informant-report as well. Findings have implications for understanding the nature of interpersonal heterogeneity in GAD and designing individualized treatments for GAD that address associated interpersonal problems.