Time to remission in late‐life depression: Analysis of effects of demographic, treatment, and life‐events measures

This study examined elderly patients being treated with a combination of nortriptyline and interpersonal psychotherapy to determine what demographic, clinical, illness‐history, and treatment‐parameter variables might predict rate of remission of the index episode. The 115 patients, aged 60 to 80 years, were suffering from at least their second lifetime episode of major depression (unipolar, nonpsychotic). Proportional hazards modeling (univariate and multivariate) showed that 4 of the 28 variables examined were significant in predicting a slower rate of response to treatment: 1) the presence of a stress‐provoking agent, such as a severely threatening life event or an ongoing major difficulty during the 6 months before onset of the index episode, 2) eccentric cluster personality features (i.e., paranoid, schizoid, and schizotypalpersonality pathologies), 3) a low level of tangible social support, and 4) earlier age at first episode. Median time to remission was increased by 2.5 weeks (from 6.5 to 9 weeks) in the presence of a stress‐provoking agent. The final model retained two variables (the presence of a stress‐provoking agent and perception of diminished social support) as significantly predictive of prolonged time to remission. Thus, these findings suggest a somewhat longer expected time to response among patients with negative psychosocial circumstances. We speculate that the use of psychotherapy and/or other adjunctive pharmacotherapy may be critical to earlier successful treatment outcome in such patients. Depression 1:250–256 (1993). © 1993 Wiley‐Liss, Inc.