Background: Patients with depression require treatment continuity when discharged from inpatient care. Interventions aimed at optimizing transition into outpatient care may be effective in preventing symptom deterioration and readmission. We aimed to evaluate the effectiveness of care transition interventions for patients with depression after psychiatric hospitalization.
Methods: Systematic review and random-effects meta-analysis of controlled trials. Primary outcomes were readmissions and symptoms of depression. The control condition was treatment as usual.
Results: We included 16 publications reporting the results of 13 different studies. Studies were heterogeneous concerning patient selection and interventional approach. Effects on readmissions and depression symptoms were non-significant in meta-analysis of 8 studies/710 patients and 7 studies/592 patients, respectively. Overall risk ratio for readmission during follow-up was 0.65 (95% CI [0.42;1.01], p=0.06), standardized mean difference for depression symptoms was -0.09 (95% CI [-0.37;0.19], p=0.53). Subgroup analyses indicated no preference for a specific interventional strategy. Data point to considerable risk for selection and publication bias.
Limitations: Included studies are heterogeneous; subgroups are often small and may not attain the power to detect effects. Reasonable classification of interventions into groups of comparable approaches was a challenge and may be arbitrary in some cases.
Conclusions: This systematic review and meta-analysis could not identify any convincingly effective interventional transition approach for patients with depression after psychiatric hospitalization. Current evidence regarding discharge management for depression is limited, heterogeneous and potentially prone to bias. Interventions might be more appropriate for patients with other diagnoses than depression. Further high-quality randomized studies are required.
Keywords: Depression; Discharge management; Meta-analysis; Systematic review; Transitional care.
Copyright © 2017 Elsevier B.V. All rights reserved.