Predictors of 1-year outcome in depression

Objective: We compared several different methods for assessing depression ' recovery' over a 1-year review interval, to determine the utility of the co ntrasting approaches. Second, we assessed baseline predictors of 1-year out come and recovery status. Third, we examined the extent to which predictors showed consistency across the variable definitions of outcome and recovery . Methods: Twelve-month outcome was assessed in a sample of 182 subjects who at baseline assessment met DSM criteria for a major depressive episode. The contrasting methods involved a defined percentage reduction in Beck Depres sion Inventory self-rating scores, formalised change point definitions, no longer meeting DSM-IV major depression criteria, and clinical global improv ement (CGI) ratings. Results: Sixty-one per cent reached formalised change point criteria for fu ll remission or recovery when trajectories across the 12-month interval wer e examined. Other measures quantified recovery rates ranging from 43% to 70 %. Those with a psychotic or melancholic depression were more likely to hav e achieved recovery status in some analyses. Non-recovery at 12 months was predicted most consistently by higher baseline levels of anxiety and depres sion; high trait anxiety and a lifetime anxiety disorder; disordered person ality function; and having reported exposure to acute and enduring stressor s at baseline assessment. Conclusions: While the CGI was the superior system in terms of number of si gnificant discriminating predictors of outcome, the change point definition al approach provides much greater information across the follow-up interval , arguing for their complementary utility. As several currently identified baseline predictors of outcome (i.e. anxiety, disordered personality functi on) also predicted onset of depression, their relevance as both depression- inducing and depression-propagating variables is suggested.