An examination of the genetic relationship between bipolar and unipolar illness in an epidemiological sample

In an epidemiologic sample of female-female twin pairs, we previously reported analyses of lifetime major depression. Because lifetime mania was not assessed, we could not differentiate unipolar from bipolar illness. Having completed such an evaluation in this sample, we now examine three questions: (i) does removing bipolar cases from our cohort substantially alter estimates for the heritability of major depression?; (ii) does our epidemiologic data support a familial relationship between major depression and mania?; and (iii) do our results for major depression and mania suggest that the two disorders are caused by the same underlying liability? We find that (i) the heritability of major depression declines only trivially if cases with a history of mania are removed; (ii) mania in one twin predicts major depression in her cotwin suggesting a familial/genetic relationship between major depression and mania; and (iii) a multiple threshold model fits our data well, consistent with the hypothesis that unipolar and bipolar disorders are points on a continuum of a single liability of illness. The validity of these results are tempered by the small number of bipolar cases detected, as expected from the low base rate of mania in general population samples.