RESUMO
OBJECTIVE: This meta-analysis investigates the efficacy of inpatient psychotherapy in major depressive disorders compared to control conditions. METHODS: In total, 14 studies were entered into the meta-analysis with a total of 1.080 patients. Primary outcome was the standardized mean differences in self-rated depression outcomes. A priori planned subgroup analyses included the influence of different control conditions: (a) no psychiatric inpatient treatment (e.g., waitlist control), (b) treatment as usual (TAU; e.g., non-manualized clinical management), (c) TAU determined by study design (manualized/'placebo' control condition), as well as number of sessions and influence of self- vs. clinician ratings. RESULTS: The meta-analysis of 19 available comparisons resulted in a moderate pooled effect size showing a small and statistically significant benefit of the psychotherapeutic intervention over control conditions (g = 0.24, P < 0.001, I2 = 0%). This corresponds to a number needed to treat of 7.4. The effects of the interventions were stable over 12-month follow-up (g = 0.21, P < 0.01, I2 = 30%). Comparisons with waitlist or non-standardized control treatments tended to be associated with larger effect sizes than standardized control treatments. CONCLUSIONS: Despite some limitations (small number of studies), this meta-analysis provides evidence for a small but sustained effect of inpatient psychotherapy in patients with major depressive disorders.