These results were contrasted to those of Gelenberg et al. (2), whose study sample consisted of patients with bipolar I disorder who had been euthymic for 2 months before intake so that relapse into mania or depression (not just mood improvement) could be determined. Moreover, those with four or more episodes were excluded from the study. In other words, the sample assessed by Nierenberg et al. would not have been in the Gelenberg et al. study, whose participants, granted, represented only a minority of mood-disordered patients (157 of 1,200). The comparison, therefore, is between apples and oranges.