AbstractBackgroundElectroconvulsive therapy (ECT) remains irreplaceable in the treatment of several psychiatric conditions. However, evidence derived using data from a national database to support its safety is limited. The aim of this study was to investigate in-hospital mortality among patients with psychiatric conditions treated with and without ECT.MethodsUsing data from the Taiwan National Health Insurance Research Database from 1997 to 2013, we identified 828,899 inpatients with psychiatric conditions, among whom 0.19% (n = 1571) were treated with ECT.ResultsWe found that ECT recipients were more frequently women, were younger and physically healthier, lived in more urbanized areas, were treated in medical centers, and had longer hospital stays. ECT recipients had lower odds of in-hospital mortality than did those who did not receive ECT. Moreover, no factor was identified as being able to predict mortality in patients who underwent ECT. Among all patients, ECT was not associated with in-hospital mortality after controlling for potential confounders.ConclusionECT was indicated to be safe and did not increase the odds of in-hospital mortality. However, ECT appeared to be administered only on physically healthy but psychiatrically compromised patients, a pattern that is in opposition with the scientific evidence supporting its safety. Moreover, our data suggest that ECT is still used as a treatment of last resort in the era of modern psychiatry.