Bridging Disciplinary Boundaries (January 11 - 14, 2007)

Purpose: Research supports the increased risk of Problem Gambling (PG) and Pathological Gambling Disorder (PGD) among individuals with varying socio-demographic characteristics, comorbid substance abuse/dependence and psychiatric disorders. Yet, there is a dearth of research considering the role of Axis II personality disorders (PD) with PG/PGD. Much of the literature has relied on treatment samples, leading to concerns of inaccurate estimates of PD comorbidity. Additionally, the literature in this area has included only certain PD types (e.g., antisocial personality disorder) and has also failed to address PG/PGD specific to certain games (i.e., game-specific disorders). Therefore, this study aims to identify associations between PD and PG/PGD when controlling for socio-demographic factors and co-occurring substance abuse/dependence.

Methods: Community-recruited gamblers (i.e., gambled >5 times lifetime) with complete data (n=146; mean age=47 yrs.; SD=18; range=18-79yrs; 55% female; 32% minority; 69% employed; 30% married; 36% <$5K annually; 58% HS graduate or less) were selected for a clinician-administered telephone interview (randomly occurring one-week before or after two non-clinician structured interviews; n=315) as a part of two NIDA-funded psychometric studies of the Computerized-Gambling Assessment Module (C-GAM)©. Moreover, we used the computerized SCID-II to assess for PD. Chi-square analyses determined significant associations of DSM-IV Personality Disorders (No PD n=80%; PD=20%) by DSM-IV Gambling Status (i.e., Recreational n=44%; Problem n=41%; Pathological n=15%). Logistic regression estimated the probability of PG/PGD in the presence of PD after controlling for socio-demographics and substance abuse/dependence. We further explored PD subtypes and game-specific disorders in this sample.

Results: Gambling status was associated with PD when controlling for socio-demographic variables and substance use disorders (Wald χ2=4.789, df=1, p=0.01, OR=2.78). There were also significant associations between PG/PGD and substance abuse/dependence (Wald χ2=4.046, df=1, p=0.044, OR=2.17), minority status (Wald χ2=8.435, df=1, p=0.003, OR=3.04) and education level (Wald χ2=6.56, df=7, p=0.01, OR=.64). Descriptively, in terms of PD subtypes, those with PGD comorbidity, were commonly diagnosed with Schizoid, Borderline and Antisocial Personality disorders. In terms of game-specific disorders, those with comorbid PD, met PGD criteria for slot machines, lotteries, card games and casino games.

Implications: The results of these analyses support existing literature that PD is associated with PG/PGD. Additional associations included substance abuse/dependence, as well as minority status and education. This information is useful to clinicians in screening and treatment planning for those with PD, particularly for those at increased risk due to socio-demographic factors and substance use disorder. We will discuss implications of these findings for PG/PGD theory development in the context of co-occurring disorders.