Relapse rates after psychotherapy for depression - stable long-term effects? A meta-analysis

- RCTs of psychotherapy for depression with long follow-ups (beyond 2 years) are very rare.
- Relapse rates of depression more than 4 years after psychotherapy are considerably high.
- Psychotherapy may have a more favorable long-term effect than comparison treatments.
- Results are based on a limited number of studies and can only be preliminary.
- Longer follow-ups are needed to develop enhanced strategies for relapse prevention.

BackgroundDepression is the most common mental disorder. Effective psychotherapeutic treatments for depression exist; however, data on their long-term effectiveness beyond a time span of two years is still scarce. Our aim was to perform a meta-analysis, investigating (a) overall rates of relapse more than two years after psychotherapy (meta-analysis 1), and (b) if psychotherapy has more enduring effects than non-psychotherapeutic comparison conditions (e.g. pharmacotherapy, treatment as usual), again beyond a time span of two years post-therapy (meta-analysis 2).MethodsWe searched electronic databases Medline, PsycINFO and the COCHRANE Library. Main selection criteria were (i) RCT of psychotherapy with follow-up interval of more than 2 years, (ii) primary diagnosis of depression, assessed by observer ratings, (iii) report of relapse at follow-up.ResultsWe identified 11 studies, 6 of which included a non-psychotherapeutic comparison condition. Together they comprised long-term data of 966 patients. Mean follow-up duration was 4.4 years. The overall relapse rate at long-term follow-up was 0.39 (95% CI 0.29, 0.50). Psychotherapy resulted in significantly less relapses (53.1% vs. 71.1%, OR 0.51; 95% CI 0.32, 0.82, p=0.005) than comparison treatments. This finding corresponded to a number needed to treat (NNT) of 5.55.LimitationsResults can only be preliminary as data was sparse and studies differed methodologically. Heterogeneity in the first meta-analysis was high (I2=82%). Results indicated publication bias.ConclusionsThe relapse rate more than two years after psychotherapy is relatively high, but significantly lower compared to non-psychotherapeutic treatments. Multiannual follow-ups should routinely be included in future psychotherapy RCTs.