While ECT is widely used for the management of severe and refractory depression, its utility in bipolar disorder is not extensively studied. The aim of this study was to examine the reported effectiveness of ECT in patients with unipolar and bipolar depression as reported by psychiatrists, nurses and patients (i.e. using objective and subjective measures). The records of 787 consecutive inpatient admissions to the Geelong Clinic, a private psychiatric centre based outside Melbourne, Victoria were reviewed in this file audit. Routine assessment measures were completed at admission and discharge, and included patient rated measures (Medical Outcomes Short Form SF-14 and Depression Anxiety and Stress Scale, DASS), nurse rated measures, (The Health of the Nation Outcome Scale, HoNOS) and a psychiatrist rated measure, the Clinical Global impression scale (CGI). In contrast to individuals with unipolar depression, where improvement was seen on all measures, in bipolar disorder, while improvement in clinician rated measures was seen (CGI, HoNOS), there was an absence of improvement in subjective measures of mood (DASS, SF14). This study suggests that in bipolar disorder, there is a poorer subjective response to ECT than in unipolar disorder.