Impulsivity and negative emotionality associated with substance use problems and Cluster B personality in college students

Abstract

The co-occurrence of personality disorders (PDs) and substance use disorders (SUDs) can be partially attributed to shared underlying personality traits. This study examined the role of negative emotionality (NEM) and impulsivity in 617 university students with self-reported substance use problems and Cluster B PD symptoms. Results indicated that NEM was significantly associated with drug and alcohol use problems, antisocial PD, borderline PD, and narcissistic PD. Impulsivity was significantly associated with drug use problems, antisocial PD, and histrionic PD. Only NEM mediated the relationship between alcohol use problems and symptoms of each of the Cluster B PDs while impulsivity mediated only the relationship between drug use problems and histrionic PD. These results suggest that NEM may be more relevant than impulsivity to our understanding of the co-occurrence between substance use problems and Cluster B PD features.

Introduction

It has been well established that substance use disorders (SUDs) and personality disorders (PDs) frequently co-occur (Skodol, Oldham, & Gallaher, 1999). This relationship seems to be especially pertinent to Cluster B PDs, which are characterized by erratic behavior and emotions and include antisocial PD, borderline PD, histrionic PD, and narcissistic PD. A recent review of the literature suggested that SUDs are found in nearly 60% of individuals diagnosed with borderline PD (Trull, Sher, Minks-Bown, Durbin, & Burr, 2000) and similar rates of comorbidity are found between antisocial PD and SUDs (Koenigsberg, Kaplan, Gilmore, & Cooper, 1985). Both SUDs (Single, 2003) and PDs (Grant et al., 2004, Pilkonis et al., 1999) represent a significant cost to society, and their co-occurrence likely presents an even greater burden as PDs have a negative effect on treatment prognosis for many disorders including SUDs (Reich and Green, 1991, Reich and Vasile, 1993). Considerable research aimed at understanding SUDs and Cluster B PDs has focused on personality traits. The goal of the current study was to examine personality traits as potential mediators of the comorbidity between features of SUDs and Cluster B PDs.

Within the past decade, the role of personality traits in psychopathology has received increased attention in research. Proponents argue that underlying personality traits help explain the myriad intercorrelations among symptoms of psychological disorders and are therefore useful in understanding individual disorders as well as the comorbidity found among them (e.g., Krueger et al., 2001, Watson et al., 1994). If comorbidity among disorders reflects commonalities in underlying personality traits, then the descriptive clinical features of the disorders can reveal which traits are likely contributors (Watson et al.).

Substance use disorders have been conceptualized as disorders of impulse, undercontrol, or disinhibition (Fowles, 1980, Fowles, 1988, Gorenstein and Newman, 1980, Sher and Trull, 1994), and research on the association of SUDs to trait impulsivity (see Moeller & Dougherty, 2002, for review) and related personality traits like Constraint (e.g., McGue et al., 1999, McGue et al., 1997) and Conscientiousness (Trull, Waudby, & Sher, 1994) support this idea. However, a rich body of research on tension-reduction motives for substance use suggests that SUDs–perhaps at least in some cases – reflect a maladaptive coping response to high negative affect or high stress reactivity, and the positive relationship found between SUDs and personality traits like Neuroticism (see Ball, 2002 for a review; Brooner et al., 2002, Flory et al., 2002; Trull et al.) and Negative Emotionality (NEM; McGue et al., 1999, McGue et al., 1997) support this idea. Antisocial (Cloninger, 1987a, Gray, 1975, Gray, 1987) and borderline PD (Trull et al., 2000) have been similarly labeled as disorders of impulsivity or disinhibition, and that idea is supported by research showing that self-reported impulsivity was at least moderately related to all Cluster B PDs (Casillas & Clark, 2002). Like SUDs, Cluster B PDs are positively associated with Neuroticism (e.g., Ball, 2002, Flory et al., 2002, Morey and Zanarini, 2000) and antisocial PD has been linked to high NEM (Krueger et al., 2001, Verona et al., 2001). Thus, impulsivity and Neuroticism or NEM are traits that have a similar association to SUDs and Cluster B PDs and might serve to explain some of their comorbidity (Sher & Trull, 1994), and a few empirical tests provide some support for this idea.

Casillas and Clark (2002) demonstrated that trait impulsivity underlies part of the comorbidity among SUDs and Cluster B PDs. Similarly, Neuroticism (Ball et al., 1997, Brooner et al., 2002, Rounsaville et al., 1998) accounts for part of the relationship between SUDs and Cluster B PDs. Although a promising start, this literature does not paint a complete picture of how impulsivity and negative affectivity account for comorbidity between SUDs and Cluster B PDs. Krueger, McGue et al. (2001) pointed out that past research failed to deal effectively with the issue of comorbidity because of the tendency to focus on single disorders and single traits. Although more recent research has begun to examine comorbidity directly (i.e., focus on multiple disorders), the literature is still hampered in its available conclusions by a continued data analytic focus on single traits (even when multiple personality traits are examined within a study) and a lop-sided focus on comorbidity of SUDs with only some Cluster B PDs (antisocial and borderline PD, in particular).

The present study will further our understanding of the contribution of impulsivity and NEM to the comorbidity of SUDs and Cluster B PDs by examining the personality traits jointly, which will provide a means to test for an association of each trait to each disorder while controlling for the other trait to help clarify whether NEM and impulsivity are associated with unique variance in each disorder or whether their contributions largely overlap or interact. The present study will examine NEM and impulsivity as they relate separately to (a) alcohol use problems, controlling for symptoms of drug use disorders, (b) drug use problems, controlling for symptoms of alcohol use disorders, and (c) to all Cluster B PDs (not just the more commonly studied antisocial and borderline conditions), controlling for symptoms of other Cluster B PDs. The present study will improve upon the methodology of earlier investigations by controlling for symptoms of other Cluster B PD when analyzing each PD – an important control lacking in previous work given the high co-occurrence of PDs in the general population (Grant, Stinson, Dawson, Chou, & Ruan, 2005). Therefore, results of the present study will provide novel information about not only the comorbidity between alcohol and drug use problems to understudied Cluster B PDs (histrionic and narcissistic) but also about the association of NEM and impulsivity to histrionic and narcissistic PD that will help flesh out theories that attempt to explain the covariation of SUDs and Cluster B PDs. The current study is based on the premise that basic personality traits predispose individuals to the later development of PDs and SUDs; however, this is a cross-sectional study which prevents us from making causal attributions. Nonetheless, cross-sectional designs provide a useful first approach to novel research questions.

In line with previous research, we expected Cluster B PDs (particularly antisocial and borderline PD) to show independent positive associations with alcohol and drug problems. Notably, prior research showing a positive association between histrionic PD and SUDs (e.g. Oldham et al., 1995, Trull et al., 2004) failed to control for other Cluster B PDs and, therefore, our study will help clarify the relationship of histrionic PD to SUDs. Additionally, consistent with previous research, we expected both impulsivity and NEM to show independent positive associations with alcohol and drug problems and with symptoms of borderline and antisocial PD. Finally, we expected NEM and impulsivity to each mediate the association between substance use problems and Cluster B PD symptoms thereby demonstrating that each of those personality traits contributes to the comorbidity of SUDs and Cluster B PDs.