Objective: Current clinical knowledge holds that antidepressants have a delayed onset of efficacy. However, the delayed onset hypothesis has been questioned recently by survival analytical approaches. We aimed to test whether early improvement under antidepressant treatment is a clinically useful predictor of later stable response and remission.
Method: We analyzed data from a randomized double-blind controlled trial with mirtazapine and paroxetine in patients with major depression (DSM-IV). Improvement was defined as a 17-item Hamilton Rating Scale for Depression (HAM-D-17) score reduction of > or = 20%. Stable response was defined as > or = 50% HAM-D-17 score reduction at week 4 and week 6, and stable remission as a HAM-D-17 score of < or = 7 at week 4 and week 6. Sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were calculated.
Results: Improvement occurred in a majority of the analyzed patients within 2 weeks (mirtazapine: 72.7% of 109 patients; paroxetine: 64.9% of 103 patients). Early improvement was a highly sensitive predictor of later stable response or stable remission for both drugs. NPV approached maximum values as early as week 2 for mirtazapine and week 3 for paroxetine. After 2 weeks of treatment with mirtazapine and 3 weeks with paroxetine, almost none of the patients who had not yet improved became a stable responder or stable remitter in the later course.
Conclusion: Our results strongly suggest that early improvement predicts later stable response with high sensitivity. These empirically derived data question the delayed onset hypothesis for both antidepressants tested and provide important clinical clues for an individually tailored antidepressant treatment.