Symposia
Addictive Behaviors
Emily E. Levitt, PhD
McMaster University
Guelph, Ontario, Canada
Assaf Oshri, PhD
Associate Professor
Department of Human Development and Family Science, University of Georgia
Athens, Georgia
Sandra S. Sanchez-Roige, PhD
Assistant Adjunct Professor
Department of Psychology, University of California
La Jolla, California
Abraham A. Palmer, PhD
Professor
Department of Psychiatry, University of California San Diego
La Jolla, California
James MacKillop, PhD
Professor
Peter Boris Centre for Addictions Research, McMaster University & St. Joseph's Healthcare Hamilton
Hamilton, Ontario, Canada
Background: Delay discounting (DD), a behavioural economic index of impulsivity, has been consistently associated with numerous psychiatric conditions and health behaviours. As a result, DD has been hypothesized to be a transdiagnostic mechanism. Few studies, however, have explored the relationships between DD and multiple conditions concurrently or done so in population-level datasets.
Methods: In study 1, data came from 1388 community adults who completed measures of DD, alcohol (AU), tobacco (TU), cannabis (CU), and other drug (DU) use, depression, anxiety, PTSD and ADHD. In study 2, data came from the Healthy Nevada Project (N = 11,061), a community population health study, including measures of DD, AU and TU, self-reported diagnoses for substance use disorder (SUD), behavioural addiction (BA), ADHD, anxiety, depression, bipolar disorder (BP), obsessive-compulsive disorder, PTSD, eating disorders (ED), life-time substances used (LTSU), and BMI. A 3-step data analytic strategy was used across studies exploring DD 1) in relation to individual health behaviours; 2) in relation to indicators of latent overlap amongst conditions using principal component analysis (PCA); 3) using structural equation models (SEM) examining all indicators concurrently.
Results: In study 1 individual analyses revealed elevations in DD in those screening positive for TU, CU, and DU disorders, ADHD, depression, and anxiety. PCA produced two latent factors (substance use vs. other psychiatric indicators) and DD was significantly related to both. SEM revealed unique significant positive associations between DD and TU, CU, and depression. In study 2 individual models revealed significant elevations in DD based on the presence of all health behaviours except ADHD, ED, and PTSD. PCA yielded two latent factors (addictive behaviours vs. all other psychiatric disorders and obesity) and DD was significantly related to both. SEM revealed unique significant positive associations between DD and AU, TU, LTSU, SUD, BA, anxiety, and BMI.
Conclusion: Across samples, DD was associated with an array of psychiatric conditions when considered individually and as latent aggregations. However, SEM revealed attenuated associations with some indicators, suggesting that specific links are not present when considered concurrently. While the extent of the transdiagnosticity of DD may not be as robust as initially hypothesized, these findings do suggest that DD is a transdiagnostic indicator, that is particularly prevalent across externalizing behaviours.