A controlled family history study of prepubertal major depressive disorder.

First-degree (N = 195) and second-degree (N = 785) adult relatives of prepubertal children with major depression (N = 48), children with nonaffective psychiatric disorders (N = 20), and normal children (N = 27) were assessed by the Family History-Research Diagnostic Criteria method (FH-RDC), except for the adult informant (usually the mother), who was directly interviewed. Compared with normal controls, prepubertal children with major depressive disorder (MDD) had significantly higher familial rates of psychiatric disorders in both first- and second-degree relatives, especially MDD, alcoholism, and "other" (mostly anxiety) diagnoses. Relatives of children in the nonaffective psychiatric control (PC) group had low rates of alcoholism, high rates of other (anxiety) disorder diagnoses, and intermediate rates of MDD (accounted for by those children with separation anxiety). This suggests that prepubertal onset of major depression may be especially likely in families with a high aggregation of affective disorders when these families also have a high prevalence of alcoholism, and that a proportion of children without affective disorder but with separation anxiety disorder in this study were at high risk for the development of affective illness later in life. These results support the validity of prepubertal-onset depressive illness as a diagnostic category, and are consistent with high familial rates of MDD and other psychiatric disorders found in family studies of adolescent and early-onset adult probands with major affective disorders, and with studies of the offspring of parents with major affective disorders. Within the child MDD group substantial heterogeneity was found. Low familial rates of MDD were associated with suicidality and comorbid conduct disorder in the child probands. The highest familial rates of MDD, approximately threefold those in the normal controls, and all the bipolar relatives, were found in the families of prepubertal probands with MDD who never had a concrete suicidal plan or act and who were without comorbid conduct disorder. A useful nosological continuum in which to classify prepubertal MDD may be to place at one end those patients with comorbid conduct disorder and at the other end those patients with manifestations related to bipolarity, including hypomania, mania, and psychotic subtype.