Impulsivity and compulsivity in pathological gambling and obsessive-compulsive disorder
Impulsividade e compulsividade no jogo patológico e no transtorno obsessivo-compulsivo
The article by Tavares and Gentil represents an important contribution towards our understanding of the relationship between impulse control disorders (specifically here, pathological gambling - PG) and obsessive compulsive disorder (OCD).1 The authors investigated the clinical characteristics and personality features of three groups, those with: 1) PG; 2) OCD; and 3) neither. Their findings that personality features of novelty seeking and impulsivity were related and elevated in PG, and harm avoidance was elevated in OCD are largely consistent with prior studies, as discussed in the manuscript. The examination of patterns of clinical courses of the disorders and patterns of co-occurring disorders are consistent with the personality findings and the literature on PG and OCD. For example, the data presented on the temporal progression of PG and OCD further suggest a similarity between PG and substance use disorders (i.e., the presence of a "telescoping" phenomenon that is present in women as compared to men2-3) that is absent in OCD. The data presented on the episodic versus continuous nature of symptoms in PG and OCD indicate a more continuous pattern in OCD as compared with PG. The intermittent nature of symptom experience in PG is consistent with the "chronic, relapsing" pattern of behaviors typically ascribed to drug addiction.2 The high rate of co-occurrence between PG and substance use disorders is consistent with findings from prior studies of clinical and community samples and consistent with considering PG as a non-substance-related addiction.2 The high rate of co-occurrence between OCD and somatoform disorders and a substantially lower rate of co-occurrence between OCD and substance use disorders suggests that OCD clusters separately from substance use disorders in the structure of psychiatric disorders, although direct investigation of this notion is needed. These data are both useful and timely as considerations for how to classify PG and other impulse control disorders with respect to OCD and substance use disorders in psychiatric nomenclature systems are being discussed.2-3 Moreover, as the conceptualizations and categorizations of impulse control disorders have important implications for research, prevention and treatment efforts,4 the findings have substantial clinical relevance.
As often occurs in research, the present study raises a number of important questions. For example, what is the nature of the difference in the shorter symptom consolidation period observed in PG as compared with OCD, and how do processes like denial or insight influence the consolidation periods? What is the nature of compulsivity? Specifically, the authors found that one measure of temperament [Harm Avoidance 2 (HA2) Fear of Uncertainty vs. Confidence] distinguished individuals with OCD from healthy control subjects without the disorder. The authors then used this feature as a measure of compulsivity. In contrast, the authors used a scale to assess impulsivity and examine its relationship to the temperament features found to distinguish PG from control subjects. Although the research on impulsivity is more established than that on compulsivity, scales designed to assess compulsivity (e.g., the Padua Inventory) have been validated and explored in studies of OCD and PG to assess compulsivity.5 The inclusion of such a scale would have lent greater support to the notion that HA2 scores are related to compulsivity. Additionally, impulsivity and compulsivity represent complex, multi-faceted constructs, and factor analyses have identified multiple components contributing to each. Analysis of the relationship between the temperament features and the core components of impulsivity and compulsivity in relation to PG and OCD would be important to better understand specific contributions to the disorders. A recently formed research society, the International Society for Research on Impulsivity and Impulse Control Disorders (ISRI; www.impulsivity.org), provides a scientific forum for addressing these and related topics.
An important finding from the present study is tabulated in the graph for figure one. Specifically, using the measures of impulsivity and compulsivity forwarded by the authors, PG, OCD and control subject scores are displayed in a scatter plot. As expected from the definitions of impulsivity and compulsivity and their derivation from the PG and OCD sample data, the majority of PG subjects are in the impulsive half of the graph and the majority of OCD subjects, in the compulsive half. However, the upper right hand quadrant (high impulsivity, high compulsivity) contains the majority of both PG and OCD subjects. The findings from this graph are notable for several reasons. First, they suggest that aspects of impulsivity and compulsivity may be found in both PG and OCD. Second, they suggest that impulsivity and compulsivity as related to PG and OCD may not be entirely orthogonal and that a more complex relationship between the two might exist. Third, the scatter plot suggests some individuals with OCD score low on compulsivity and some with PG score low on impulsivity. Thus, as suggested by the authors, heterogeneities likely exist in PG and OCD and further examinations (e.g., of individual differences and along dimensional lines) are needed. Thus, although it is theoretically attractive, it is likely premature to adopt the quadrant model proposed in figure one, although such models are very useful for hypotheses testing in future research efforts.
Although caution is warranted in generalizing the findings from this study due to various limitations (including small sample size, limited geographical variability, clinical treatment-seeking samples, single measure of compulsivity, and lack of longitudinal assessment), the study is important in many ways and moves the field forward in terms of understanding the relationships between impulsivity, compulsivity, PG and OCD.
Marc N Potenza
Problem Gambling Clinic, Yale University School of Medicine,
New Haven, Connecticut, USA
Women and Addictive Disorders Core, Women's Health Research, Yale University School of Medicine, New Haven, Connecticut, USA
Neuroimaging, MIRECC, West Haven Veteran's Affairs Hospital, Connecticut, USA
Yale University School of Medicine,
New Haven, Connecticut, USA
This work was supported in part by funding from the National Institute on Drug Abuse, the US Department of Veteran's Affairs, and Women's Health Research at Yale.
1. Tavares H, Gentil V. Pathological gambling and obsessive compulsive disorder: towards a spectrum of disorders of volition. Rev Bras Psiquiatr. 2007;29(2):107-17.
2. Potenza MN. Should addictive disorders include non-substance-related conditions? Addiction. 2006;101(Suppl 1):142-51.
3. Potenza MN, Koran LM, Pallanti S. The relationship between obsessive-compulsive and impulse control disorders: a current understanding and future research directions. Psychiatry Res. (invited manuscript, under review).
4. Tamminga CA, Nestler EJ. Pathological gambling: focusing on the addiction, not the activity. Am J Psychiatry. 2006;163(2):180-1.
5. Grant JE, Potenza MN. Compulsive aspects in impulse control disorders. Psychiatr Clin N Am. 2006;29(2):539-51.
Financing: Dr. Potenza receives grant/research/clinical support from the Connecticut Department of Mental Health and Addiction Services, Mohegan Sun, the National Institute on Alcohol Abuse and Alcoholism, the National Institute on Drug Abuse, the United States Department of Veteran Affairs (VA), and Women's Health Research at Yale University School of Medicine
Conflict of interests: Dr. Potenza is a consultant to Boehringer Ingelheim; is on the advisory board of Boehringer Ingelheim; and has financial interest in Somaxon
Publication in this collection
06 July 2007
Date of issue