Family Transmission and Heritabilityof Externalizing DisordersA Twin-Family StudyBrian M. Hicks, MA; Robert F. Krueger, PhD; William G. Iacono, PhD; Matt McGue, PhD; Christopher J. Patrick, PhDBackground: Antisocial behavior and substance de-pendence disorders exact a heavy financial and humancost on society. A better understanding of the mecha-nisms of familial transmission for these “externalizing”disorders is necessary to better understand their etiol-ogy and to help develop intervention strategies.Objectives: To determine the extent to which the fam-ily transmission of externalizing disorders is due to a gen-eral vs a disorder-specific vulnerability and, owing to thegenetically informative nature of our data, to estimate theheritable vs environmental nature of these transmissioneffects.Design: We used structural equation modeling to si-multaneously estimate the general and specific transmis-sion effects of 4 externalizing disorders: conduct disor-der, adult antisocial behavior, alcohol dependence, anddrug dependence.Setting: Participants were recruited from the commu-nity and were interviewed in a university laboratory.Participants: The sample consisted of 542 families par-ticipating in the Minnesota Twin Family Study. All fami-lies included 17-year-old twins and their biological motherand father.Main Outcome Measures: Symptom counts of con-duct disorder, the adult criteria for antisocial personal-ity disorder, alcohol dependence, and drug depen-dence.Results: Transmission of a general vulnerability to allthe externalizing disorders accounted for most familialresemblance. This general vulnerability was highly heri-table (h2=0.80). Disorder-specific vulnerabilities werealsodetected for conduct disorder, alcohol dependence, anddrug dependence.Conclusions: The mechanism underlying the familialtransmission of externalizing disorders is primarily ahighly heritable general vulnerability. This general vul-nerability or common risk factor should be the focus ofresearch regarding the etiology and treatment of exter-nalizing disorders.Arch Gen Psychiatry. 2004;61:922-928ANTISOCIAL BEHAVIOR ANDsubstance dependence dis-orders are most notablycharacterized by the ex-ceedingly high financialand human costs they exact on society. Al-though institutions such as the criminaljustice system and treatment programshave been developed to cope with the con-sequences of these disorders, effective in-tervention will also require a better un-derstanding of their etiology, particularlythe nature of familial resemblance. Suchan understanding requires answers to fun-damental questions, such as “Is intergen-erational transmission attributable to a gen-eral vulnerability to all disorders in thisspectrum or is familial transmission pri-marily due to disorder-specific vulner-abilities?” and “Are the origins of familialresemblance largely genetic or environ-mental in nature?”Family studies consistently show thatthe presence of a substance dependenceor antisocial behavior disorder in first-degree relatives greatly increases a per-son’s risk of eventually developing the dis-order.1Although there is evidence of ageneral familial risk factor, family stud-ies2-6tend to show that the familial aggre-gation of different disorders is relativelymore specific than general. Adoption stud-ies7-11are generally consistent with fam-ily studies in suggesting specific geneticand environmental risk factors rather thanageneraladdictivetendencyorpatternofdeviant behavior. In addition, the famil-ial aggregation of substance dependenceand antisocial behavior disorders tends tobe the same in men and women, al-ORIGINAL ARTICLEFrom the Department ofPsychology, University ofMinnesota, Minneapolis.(REPRINTED) ARCH GEN PSYCHIATRY/ VOL 61, SEP 2004 WWW.ARCHGENPSYCHIATRY.COM922©2004 American Medical Association. All rights reserved. at Univ of Colorado Hlth Sci Ctr, on July 11, 2005 www.archgenpsychiatry.comDownloaded fromthough women may require greater familial loading be-fore expressing the disorder.6,10-14Twin studies, however, tend to show substantial ge-netic overlap among disinhibitory syndromes, includ-ing conduct disorder and alcohol dependence,15-17childand adult antisocial behavior,15,16,18adult antisocial be-havior and alcohol and drug dependence,15,16,19and de-pendence among different classes of illicit drugs.20,21Twinstudies15,16also tend to show the same pattern of geneticand environmental effect for men and women. Al-though twin studies15-21consistently reveal genetic over-lap among these disorders, they also find evidence of ge-netic and environmental risk factors that are specific toeach disorder. Analyses22-24that examine the patterns ofcomorbidity in large epidemiologic samples consis-tently identify a broad dimension of risk underlying an-tisocial personality disorder (ASPD), conduct disorder,alcohol dependence, and drug dependence. This gen-eral vulnerability factor is typically referred to as exter-nalizing (EXT),15,16,22-24and findings from recent twin stud-ies15,16,25indicate that the source of the comorbidity inthese externalizing disorders can be largely attributed tocommon genetic factors.Rare are twin-family studies that integrate the 2 lit-eratures,26-28allowing for the simultaneous estimation ofgeneral and specific transmission and the genetic and en-vironmental contributions to these effects. Although fam-ily studies tend to show specificity of transmission andtwin studies tend to show generality, a recent critique29of family studies concluded that analyses typical of fam-ily studies do not provide valid tests of alternative co-morbidity models. As a remedy, the authors recom-mended a model-fitting approach, which we use in thepresent investigation. The present study is also uniquein that, to our knowledge, it is the first twin-family studythat incorporates the assessment of multiple externaliz-ing disorders in both biological parents and their twinoffspring. Such a study provides an opportunity to clarifywhether familial resemblance (both parental and sib-ling) is due to general vs disorder-specific vulnerabili-ties while also estimating the genetic and environmen-tal contributions to these general and specific risk factors.METHODSPARTICIPANTSParticipants were members of 542 families participating in theMinnesota Twin Family Study (MTFS), a longitudinal-epidemiologic study investigating the development of sub-stance abuse and related disorders. A comprehensive descrip-tion of the goals,