Because most published family studies of depression ascertained subjec
ts from treatment settings, the reported familial aggregation of depre
ssion could be an artifact if a family history of depression increased
the likelihood of seeking treatment. To investigate this possibility,
we compared the family history of depression in three groups of proba
nds aged 18-44: 54 women randomly selected from the community with dep
ression in the prior year, 41 women who entered a clinical trial for d
epression and 37 women randomly selected from the community who had no
t been depressed in the prior year. The presence of depression in the
parents and siblings of the probands was assessed by the family histor
y method and quantified via family history scores which took the age,
gender and number of relatives into account. Depressed probands ascert
ained from clinical sources had markedly higher family history scores
of depression than other two groups (P < 0.00005 in each instance). in
the absence of direct interviews with relatives, we cannot exclude th
e impact of differential reporting. A family history of depression mig
ht be associated with an increased probability of treatment or the dif
ferential reporting of family history, It is thus possible that the fa
milial aggregation of depression observed in probands from treatment s
ettings is an artifact.