ORIGINAL ARTICLES Parsing the Association between Bipolar, Conduct, and Substance Use Disorders: A Familial Risk Analysis Joseph Biederman, Stephen V. Faraone, Janet Wozniak, and Michael C. Monuteaux Background: Bipolar disorder has emerged as a risk factor for substance use disorders (alcohol or drug abuse or dependence) in youth; however, the association be- tween bipolar disorder and substance use disorders is complicated by comorbidity with conduct disorder. We used familial risk analysis to disentangle the association between the three disorders. Methods: We compared relatives of four proband groups: 1) conduct disorder + bipolar disorder, 2) bipolar disor- der without conduct disorder, 3) conduct disorder without bipolar disorder, and 4) control subjects without bipolar disorder or conduct disorder. All subjects were evaluated with structured diagnostic interviews. For the analysis of substance use disorders, Cox proportional hazard survival models were utilized to compare age-at-onset distributions. Results: Bipolar disorder in probands was a risk factor for both drug and alcohol addiction in relatives, indepen- dent of conduct disorder in probands, which was a risk factor for alcohol dependence in relatives independent of bipolar disorder in probands, but not for drug depen- dence. The effects of bipolar disorder and conduct disor- der in probands combined additively to predict the risk for substance use disorders in relatives. Conclusions: The combination of conduct disorder + bipolar disorder in youth predicts especially high rates of substance use disorders in relatives. These findings sup- port previous results documenting that when bipolar disorder and conduct disorder occur comorbidly, both are validly diagnosed disorders. Biol Psychiatry 2000;48: 1037–1044 © 2000 Society of Biological Psychiatry Key Words: Bipolar disorder, conduct disorder, substance use, familial risk Introduction I n recent years, a focus on bipolar disorder (BPD) as a risk factor for substance use disorders (SUDs; alcohol or drug abuse or dependence) in youth has emerged as a clinical and public health concern. A prospective study of children and adolescents with and without attention-deficit/hyperactivity disorder (ADHD) found that early-onset BPD predicted subsequent SUD indepen- dently of ADHD (Biederman et al 1997). Similarly, an excess of SUDs has been reported in studies of adolescents with BPD or prominent mood lability and dyscontrol (Biederman et al 1997; West et al 1996; Wilens et al 1997a; Wills et al 1995; Young et al 1995). West et al (1996) reported that 40% of inpatient adolescents with BPD suffered from SUDs. Likewise, we reported that psychiatrically referred adolescent outpatients with SUDs were more likely than those without SUDs to have comorbid BPD (Wilens et al 1997a). But understanding the association between BPD and SUDs is complicated by the fact that BPD is frequently comorbid with conduct disorder (CD; Biederman et al 1998b; Faraone et al 1997b; Geller et al 1994; Kovacs and Pollock 1995; Kutcher et al 1989; Wozniak et al 1995a) and CD is a well-documented risk factor for SUDs in youth (Bukstein et al 1989, 1992; DeMilio 1989; Hovens et al 1994; Kaminer 1991; McKay et al 1991, 1992; West et al 1996; Wilens et al 1997a). One approach to addressing this issue is the use of data from families (Faraone et al 1999). Since BPD, CD, and SUD are known to be familial conditions, examining their familial patterns of aggregation and coaggregation can disentangle the associations among them. Although sev- eral studies have shown a familial association between BPD and SUD and between CD and SUD (Dunner et al 1979; Maier and Merikangas 1996; Morrison 1975; Penick et al 1978; Raskin and Miller 1993), no studies examined the three-way associations between BPD, CD, and SUD. A better understanding of the links among SUDs, CD, and BPD is of high scientific, clinical, and public health relevance. Clinically, the identification of BPD in SUD youth may permit the use of appropriate treatments target- ing the underlying mood disorder. Scientifically, the de- lineation of a subtype of SUDs linked to mood disorders in the young may lead to the identification of a more From the Pediatric Psychopharmacology Unit of the Child Psychiatry Service, Massachusetts General Hospital (JB, SVF, JW, MCM), Harvard Institute of Psychiatry, Epidemiology and Genetics, Department of Psychiatry, Harvard Medical School at the Massachusetts Mental Health Center, and the Common- wealth Research Center (SVF), and the Department of Epidemiology, Harvard School of Public Health (MCM), Boston, Massachusetts. Address reprint requests to Joseph Biederman, M.D., Massachusetts General Hospital, Pediatric Psychopharmacology Unit, ACC 725, 15 Parkman Street, Boston MA 02114-3139. Received October 14, 1999; revised January 31, 2000; accepted April 25, 2000. © 2000 Society of Biological Psychiatry 0006-3223/00/$20.00 PII S0006-3223(00)00906-9