The review addressed a well-defined question in terms of participants, interventions, study design and relevant outcomes. Only one relevant database was searched for published studies in English and it appeared that unpublished studies were not considered; therefore, some relevant studies may have been missed. The authors may have identified potentially relevant studies without using the search strategy and included one of these in the review. Study quality was assessed using suitable criteria, but few relevant details were reported. Validity assessment was carried out with efforts to reduce error and bias; it was unclear whether this applied to other aspects of the review process.
Some relevant study details were reported, but no details of patient age and gender were provided. Statistical heterogeneity was assessed. The statistical method used for the meta-analysis seemed appropriate, but there were reservations about the appropriateness of combining the results considering heterogeneity in symptoms and lack of patient data. No subgroup analyses were performed. The authors noted that patients with psychotic depression were known to respond better to ECT and so the proportion of those patients in a study would affect the overall results.
Potential limitations in the review process, uncertainties about the quality of included studies (all the studies were observational studies and these have their own biases) and the limited number of studies identified mean the authors’ conclusions should be treated with caution.