Age of onset (AO) has been proposed as a promising criterion by which to select homogeneous subgroups for the genetic analysis of bipolar disorder. This is the first study to investigate the effect of the interaction between gender and family history (FH)-type on AO in bipolar disorder. In accordance with the literature, no difference in AO was observed between females and males in our sample of 264 Romanian bipolar I probands. Cox regression, however, showed a strong influence of FH-type on AO (P = 0.006). This was due to a significant variation in AO according to the type of FH in females (P = 0.002) but not in males (P = 0.64). Female bipolar disorder patients with a negative FH (FH(-)) had a later AO than females with either a FH of bipolar and/or schizoaffective disorder (P = 0.001) or a FH of recurrent unipolar major depression only (P = 0.04). Females with FH(-) had a later AO than males with FH(-) (P = 0.03). No sex difference was observed for AO in the group with a FH of recurrent unipolar depression. In the group with a FH of bipolar and/or schizoaffective disorder, females had an earlier AO than males (P = 0.01). A trend for support was observed in an independent sample of 217 German bipolar I patients for an influence of FH-type on AO in females (P = 0.09) but not in males (P = 0.15). Female bipolar disorder patients with FH(-) had a later AO than females with either a FH of bipolar and/or schizoaffective disorder (P = 0.04) or a FH of recurrent unipolar major depression only (P = 0.05). Females with FH(-) had a later AO than males with FH(-) (P = 0.05). Other comparisons were statistically not significant, which may be due to limited sample size. Our findings emphasize that the interaction between gender and FH-type is a source of heterogeneity for AO in bipolar disorder.
(c) 2004 Wiley-Liss, Inc.