Background: To determine the optimal duration of an antidepressant trial in elderly patients, the authors examined the probability of eventually responding to treatment based on early improvement.
Methods: Four hundred seventy-two elderly patients with major depression (nonpsychotic, nonbipolar) were treated under protocolized conditions for up to 12 weeks and assessed weekly with the Hamilton Rating Scale for Depression. The probability of full response after 12 weeks of treatment was calculated in patients who had not fully responded after periods of treatment that lasted for 4 to 10 weeks.
Results: Most of the patients who had shown a partial improvement after 4 weeks of treatment became full responders after 4 or more additional weeks of treatment. By contrast, only a few of those who were nonresponders became full responders even after up to 8 additional weeks of treatment.
Conclusions: After 4 weeks of treatment, it is possible to reliably identify a subgroup of elderly patients with depression who are more likely to benefit from a change in their treatment than from a few additional weeks of treatment with the same agent.