A CONTROLLED FAMILY STUDY OF ANOREXIA-NERVOSA AND BULIMIA-NERVOSA - PSYCHIATRIC-DISORDERS IN FIRST-DEGREE RELATIVES AND EFFECTS OF PROBAND COMORBIDITY

Background: We used contemporary family-epidemiological methods to exa mine patterns of comorbidity and familial aggregation of psychiatric d isorders for anorexia and bulimia nervosa. Methods: Direct interviews and blind best-estimate diagnostic procedures were used with diagnosti cally ''pure'' groups of probands with eating disorders and a matched control group. Lifetime prevalence rates of eating disorders, mood dis orders, substance use disorders, anxiety disorders, and selected perso nality disorders were determined in female probands with restricting a norexia nervosa (n=26) or bulimia nervosa (n=47), control women (n=44) , and first-degree biological relatives (n=460). Results: Relatives of anorexic and bulimic probands had increased risk of clinically subthr eshold forms of an eating disorder, major depressive disorder, and obs essive-compulsive disorder. Familial aggregation of major depressive d isorder and obsessive-compulsive disorder was independent of that of a norexia nervosa and bulimia nervosa. These relatives also had increase d risk of other anxiety disorders, bur, the mode of familial transmiss ion was not clear-cut. The risk of substance dependence was elevated a mong relatives of bulimic probands compared with relatives of anorexic probands, and familial aggregation was independent of that of bulimia nervosa. The risk of obsessive-compulsive personality disorder was el evated only among relatives of anorexic probands, and there was eviden ce that these 2 disorders may have shared familial risk factors. Concl usions: There may be a common familial vulnerability for anorexia nerv osa and bulimia nervosa. Major depressive disorder, obsessive-compulsi ve disorder, and substance dependence are not likely to share a common cause with eating disorders. However, obsessional personality traits may be a specific familial risk factor for anorexia nervosa.