Bipolar I affective disorder: predictors of outcome after 15 years

Background: Robust predictors of long-term outcome in bipolar affective dis order would have substantial importance to both clinicians and researchers. Such predictors are not available, however, perhaps because of the limitat ions of previous efforts to find them. Methods: In this study, 113 patients with bipolar affective disorder were followed semiannually for 5 years and annually for a subsequent 15 years. Of these, 23 (20.4%) had a poor long-t erm outcome indicated by the presence of mania or major depressive disorder throughout the 15th year. Results: Among the baseline demographic and clin ical variables tested, only active alcoholism and low levels of optimum fun ctioning in the preceding 5 years characterized poor outcome patients. The persistence of depressive symptoms in the first 2 years of follow-up predic ted depressive symptoms 15 years later but the early persistence of manic s ymptoms seemed to have no predictive value. A regression analysis eliminate d alcoholism as an independent predictor. Thus, only poor optimal functioni ng in the 5 years before baseline assessment, and the persistence of depres sive symptoms in the two subsequent years, were independently associated wi th poor, long-term prognosis. Limitations: Patients were recruited at terti ary care centers and sampling was therefore biased toward greater severity and chronicity. As is true of all naturalistic studies of course, treatment was not controlled. Conclusion: These findings suggest the: existence of a poor outcome, depression-prone subtype of bipolar affective disorder. (C) 1998 Elsevier Science BN. All rights reserved.