Recently published controlled studies comparing electroconvulsive therapy (ECT) with other treatments for depression offer the opportunity to perform a meta-analysis of ECT in depression. Fifteen studies were identified which fulfilled the inclusion criteria. From these controlled trials, 20 effect sizes of ECT were calculated. The speed of action during the course and the efficacy after a full course of ECT were explored. The efficacy of sine wave and brief pulse machines were compared. The comparison between ECT and four other comparative treatments was made. Predictive variables were explored using homogeneity tests. ECT was shown to be superior after a full course. The funnel plot showed the absence of publication bias. There was no exaggeration of effect size in the lower quality trials. No evidence was found for a superior speed of action of ECT or for a difference in efficacy between sine wave and brief pulse stimulation. ECT was shown to be superior to medication and simulated ECT. Some evidence was found that psychosis predicted better response to ECT.