Journal of Abnormal Psychology 1988, Vol. 97, No. 3,265-274 Copyright 1988 by the American Psychological Association, Inc. 0021-843X/SS/$00.75 Dysthymia in the Offspring of Parents With Primary Unipolar Affective Disorder Daniel N. Klein University of Illinois at Urbana-Champaign Laura Dansky Northwestern University David C. Clark Department of Psychiatry Rush-Presbyterian-St. Luke's Medical Center Eliezer T. Margolis University of Rhode Island This study examined whether there is a familial relation between primary early-onset dysthymia and major affective disorder. In addition, it explored the prevalence of other forms of psychopathol- ogy and social impairment in the adolescent and young adult offspring of patients with primary unipolar affective disorder. Subjects included 47 offspring of patients with primary unipolar depres- sion, 33 offspring of patients with chronic orthopedic and rheumatological conditions, and 38 off- spring of randomly selected community controls with no personal or family history of psychiatric disorder. All offspring received structured diagnostic interviews. Diagnoses were derived blind to parental group by using multiple sets of diagnostic criteria. The offspring of unipolar patients exhib- ited significantly higher rates of affective disorder, major depression, and dysthymia than did the offspring of medical and normal controls. The groups did not differ on rates of nonaffective disorders. Parental characteristics associated with dysthymia in offspring included chronic depression, age of onset of major depression, number of hospitalizations, and multiple family members with major affective illness. These results support the view that at least some forms of early-onset dysthymia are variants of major affective illness. The idea that certain types of personality or temperament might be attenuated, or "subsyndromal," forms of the major affective disorders dates to the classic German descriptive psy- chopathologists of the 19th century. For example, Kraepelin (1921) identified four forms of affective temperament—manic, depressive, cyclothymic, and irritable (a mixture of hypomanic and depressive characteristics akin to mixed states in full syn- dromal bipolar disorder)—that he believed were the fundamen- tal states underlying manic depressive illness. Other investiga- tors, such as Kurt Schneider (1958), however, argued that the affective personalities were simply extreme manifestations of normally distributed personality traits that were generally unre- lated to major affective disorder. Until recently, the United States (American Psychiatric Association, 1968) and interna- tional (World Health Organization, 1978) diagnostic systems adopted the latter view, classifying the affective personalities as This study was supported in part by National Institute of Mental Health Research Grant R03-MH-39782 to Daniel N. Klein and by the University of Illinois Research Board. Portions of this article were pre- sented at the first annual meeting of the Society for Research in Psycho- pathology, Cambridge, Massachusetts, October 1986. The assistance of Janice Syslo and Stephanie Lyons in conducting some of the diagnostic interviews, of Salvatore Nuzzarello and Ira Halper in obtaining patients, and of Cathy Clayton in data coding and analysis is gratefully acknowledged. Helpful comments on an earlier draft of this article were provided by Kathryn Harding, Cindy Neff, Ellen Taylor, and Beth Williams. Correspondence concerning this article should be addressed to Dan- iel N. Klein, who is now at the Department of Psychology, State Univer- sity of New York at Stony Brook, Stony Brook, New York 11794-2500. forms of personality disorder. In placing two of these condi- tions, cyclothymia and dysthymia, under the rubric of the affective disorders, the Diagnostic and Statistical Manual of Mental Disorders (DSM-HI; American Psychiatric Associa- tion, 1980) represented a significant departure from this tradi- tion. In recent years, a number of studies have presented evidence supporting the relation between cyclothymia and major affec- tive illness. These studies have indicated that cyclothymics re- semble patients with Bipolar I affective disorder in terms of phe- nomenology (Akiskal, Djenderedjian, Rosenthal, & Khani, 1977; Depue et al., 1981), family history (Akiskal et al., 1977; Depueetal., 1981;Gershonetal., 1975; Klein, Depue, & Slater, 1985, 1986), neuroendocrinology (Depue, Kleiman, Davis, Hutchinson, & Krauss, 1985), and pharmacological response (Akiskal et al, 1977). In addition, cyclothymics appear to be at increased risk for developing more severe forms of affective disorder (Akiskal etal., 1977, 1985; Klein & Depue, 1984). In contrast, the nature of the relation between dysthymia and the major affective disorders is less clear. In part, this may be due to the greater heterogeneity of the dysthymia category. As Akiskal (1983) has noted, the DSM-III criteria for dysthymia identify at least three distinct groups: (a) primary early onset, or characterological dysthymia, the group that is closest to the Kraepelinian concept of the depressive temperament; (b) mild chronic depressions that represent incomplete recoveries from prior major depressive episodes; and (c) chronic dysphoric con- ditions that are secondary to chronic nonaffective psychiatric disorders or incapacitating medical conditions. In addition, Akiskal (1983) has suggested that there may be considerable heterogeneity even within the early-onset subgroup, with some 265