Familial and developmental factors in characterological depressions.

Patients with chronic low-grade depressions (screened to exclude primary affective illness and those secondary to rigorously defined nonaffective disorders) were divided into subaffective dysthymic versus character-spectrum groups and compared to 40 primary unipolar controls. A prior report found the 30 character-spectrums different from the 20 dysthymics (and usually from the unipolars) on pharmacological, phenomenological, REM sleep, social and outcome criteria. The present study parsed family history and developmental differences: The character-spectrum group had significantly lower incidence of familial depressions, but higher frequencies of loss of a parent in childhood, familial alcoholism, and parental assortative mating than both other groups--which did not differ. Just 10% of our 90 patients had bipolar family histories; 7 were dysthymics and 6 of these had earlier shown brief, tricyclic-induced hypomania. The results support, at the subsyndromal level, Winokur's separation of disorders with +FH for alcoholism from those with +FH for affective illness. Furthermore, data suggest the DSM-III concept of 'dysthymia' is too broad and needs further distinctions among several subaffective and nonaffective chronic depressions.