Abstract
Similar to many domains in the psychopathology literature, overlap and covariation between antisocial behavior (ASB) and alcohol dependence (AD) are oft documented but little understood. Although the relation between ASB and AD is reliably found and of substantial magnitude, it is not possible given the extant research to discriminate among alternative causal models that could give rise to this relation (e.g., ASB→AD, AD→ASB, reciprocal causation between ASB and AD, common causes of ASB and AD). In our opinion, true comorbidity among disorders can only be demonstrated and understood in the context of considerable knowledge regarding the disorders' underlying causes (viz., pathology and etiology). In this article, we present a number of behavior genetic models that may be useful for illuminating the causes of comorbidity among two or more disorders, as well as for understanding the etiology of each disorder individually. Using these behavior genetic approaches, psychopathology researchers can directly test alternative models for the comorbidity among disorders, as well as estimate the magnitude of different etiological factors (i.e., genetic and environmental influences) on comorbidity. Although not a panacea and somewhat demanding technically, behavior genetic approaches can shed new light on the comorbidity among disorders.
Introduction
THERE EXISTS A considerable literature documenting the so-called “comorbidity” between antisocial behavior and substance abuse, most notably alcoholism. Evidence for the covariation and overlap of these problems comes from studies using clinically referred and nonreferred populations, adolescents and adults, males and females, experimental and naturalistic approaches, and cross-sectional and longitudinal designs. Antisocial behavior and substance abuse appear to go hand-in-hand, as indicated by the substantial covariance of behavior problems that vary continuously among individuals in the general population and by the considerable overlap among relevant DSM diagnoses (e.g., conduct disorder, antisocial personality disorder, alcohol abuse and dependence, substance abuse and dependence). The extant literature in this domain suggests that the relation between antisocial behavior and alcohol and substance abuse is one of the best documented findings in the psychopathology literature as a whole.
Despite the reliability of findings documenting the relation between antisocial behavior and substance abuse, the meaning of these findings is somewhat obscure. It is clear that individuals who commit more rather than less antisocial acts use alcohol and other substances at higher rates and in less adaptive ways, just as individuals who are alcoholic or abuse drugs engage in antisocial behaviors at higher rates than non-substance-abusing individuals. Unfortunately, although many studies have documented such associations, the causal processes underlying them have remained unclear. Does engaging in numerous and varied antisocial behaviors lead a person to develop problems with alcohol and other substances? On the other hand, does problematic use of alcohol and other substances lead individuals to commit frequent and severe acts of aggression and antisocial behavior? Alternatively, do antisocial behavior and alcohol and substance abuse develop due to a shared predisposition or liability; that is, due to a common set of causes?
Although some studies focusing on the association between antisocial behavior and alcohol or substance abuse have attempted to discover the causes underlying this relation, our knowledge of these causes is at present limited. Hence, rather than once again simply document the association between antisocial behavior and substance abuse as many previous studies and reviews have done, we intend to focus on the causes of this association. As such, we will discuss research designs and illustrative substantive findings that are targeted at revealing the causes of comorbidity, both for psychopathological disorders in general and for antisocial behavior and substance abuse in particular. A distinguishing feature of this article will be a discussion of the unique contribution of behavior genetic methods to problems of comorbidity. Given this emphasis, our article is indebted to previous treatments of the contribution of behavior genetic methods to understanding comorbidity Carey & DiLalla 1994, Klein & Riso 1993 and Neale & Kendler 1995. Behavior genetic methods disentangle genetic and environmental influences on a trait or disorder, thus facilitating a stronger characterization of distinct underlying causes than that afforded by most other research designs. While behavior genetic methods possess many strengths of traditional experimental designs, they typically focus on the naturally occurring variation in a trait, thus providing greater generalizability than experimental methods. Although we will discuss other methods for inferring cause and their application to the study of comorbidity, behavior genetic approaches to comorbidity will form the cornerstone of this article. We will describe in detail a number of behavior genetic models that can test interesting hypotheses regarding the causes of comorbidity. While not a panacea, we feel that these behavior genetic models can reveal novel insights regarding the causes of comorbidity.
