Monetary delay discounting in gambling and cocaine dependence with personality comorbidities

Abstract

Background

Cocaine addiction and pathological gambling are commonly associated with steeper (impulsive) discounting of delayed rewards, which promotes ongoing drug and gambling behaviors. However, it is yet unclear whether impulsive delay discounting is a stable trait in cocaine and gambling disorders during abstinence, and whether it is significantly impacted by dysfunctional personality beliefs.

Methods

The aim of this study was to compare the delay discounting rates of four groups: 47 cocaine users with comorbid personality disorders, 41 cocaine users without psychiatric comorbidities, 28 pathological gamblers without psychiatric comorbidities, and 36 healthy comparison individuals. We also examined the association between dysfunctional personality beliefs and delay discounting rates. Participants completed the Kirby Delay Discounting Questionnaire and the Beck Personality Belief Questionnaire as part of a larger battery.

Results

We used non-parametric tests to compare discounting rates between the groups, and bivariate correlation analyses to examine the association between beliefs and discounting rates within each of the groups. We found that discounting rates were significantly higher in individuals with disordered gambling compared to controls. Specifically in cocaine users with Cluster B personality disorders, higher discounting rates were associated with the intensity of “dependent” dysfunctional beliefs (e.g., “I am needy and weak”).

Conclusion

We conclude that impulsive delay discounting is increased in gambling relative to controls and linked to personality beliefs in cocaine users with Cluster B personality disorders.

Introduction

The fifth edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-5) groups substance use and gambling disorders within the substance-related and addictive disorder category (American Psychiatric Association, 2013). The re-labeling of gambling (previously within the impulse control disorder category) as a behavioral addiction has built on empirical evidence showing that (i) substance use and gambling disorders have overlapping impulsivity features (Robbins et al., 2012, Verdejo-Garcia et al., 2008), and (ii) impulsivity is essential to characterize core aspects of addictive disorders, such as hypersensitivity to immediate rewards and subsequent neglect of long-term negative consequences (Bechara, 2005).

These aspects have been elegantly modeled by delay discounting monetary choice questionnaires (e.g., “Would you prefer $15 today or $35 in 13 days?”), which index the reduction in the current value of a future reward as the delay to that reward increases (Kirby, Petry, & Bickel, 1999). Steeper delay discounting, reflective of higher impulsivity, has been consistently demonstrated in gambling populations compared to healthy controls (Brevers et al., 2012, Dixon et al., 2003, Petry, 2001). However, few studies have contrasted delay discounting rates in gambling vs. other addictive disorders (Albein-Urios, Martinez-Gonzalez, Lozano, Clark, & Verdejo-Garcia, 2012). In the case of substance using populations, studies have demonstrated steeper delay discounting in cocaine users compared to heroin and alcohol users (Bornovalova et al., 2005, Kirby and Petry, 2004). Moreover, at difference with other substance using populations, steeper discounting rates are stable in cocaine users during drug abstinence (Kirby & Petry, 2004); hence, cocaine use related delay discounting impulsivity cannot be exclusively attributed to the pharmacological effects of the drug.

Collectively, the available evidence supports the notion that steeper delay discounting is a shared behavioral marker of cocaine use and gambling disorders (Bickel, Koffarnus, Moody, & Wilson, 2014). Notwithstanding this overlap, there is ongoing interest in contrasting delay discounting impulsivity between different addictive groups, mainly because a more precise understanding of how this marker relates to specific characteristics of each addiction may ultimately facilitate treatment development (Leeman & Potenza, 2012). For example, in the case of gambling, there is direct convergence of the task reinforcer (money) and the addictive cue (Limbrick-Oldfield, Clark, & van Holst, 2013). This convergence may render individuals with gambling disorder more sensitive to the appeal of immediate monetary choices. Conversely, in the case of cocaine users, delay discounting choices seem to be more influenced by stable characteristics than by specific addictive cues or drug effects (Bornovalova et al., 2005). In fitting with this notion, previous studies have demonstrated that substance using groups (including cocaine users) with comorbid personality disorders (i.e., antisocial, borderline) have steeper discounting rates than comparison groups without dual diagnoses (Coffey et al., 2011, Petry, 2002). Furthermore, dysfunctional beliefs associated with personality disorders (e.g., “I am needy and weak”) may boost the value of immediate rewards via utility-maximizing or negative reinforcement mechanisms (Callan et al., 2011, McGuire and Kable, 2013).

The main aim of this study was to compare monetary delay discounting rates between a healthy control group and three groups of individuals with addictive disorders: a group of individuals with gambling disorders, a group of individuals with cocaine use and personality disorders (diagnoses pertaining to Cluster B and Cluster C), and a group of individuals with cocaine use disorder (without other comorbidities). We also examined the association between a continuous measure of dysfunctional personality beliefs and delay discounting rates within each group. We predicted that (i) delay discounting rates would be steeper in individuals with gambling disorders and in individuals with cocaine use and personality disorders compared to controls and cocaine users without personality comorbidities, and (ii) higher degree of dysfunctional beliefs, especially those associated with perceived need (e.g., borderline or dependent personality beliefs), would be positively correlated with discounting rates.