BACKGROUND This study examined the relationship between age and outcome of electroconvulsive therapy (ECT). METHODS This was a naturalistic, prospective follow-up of 81 consecutive in-patients with primary major depression. ECT outcome was compared for three age groups - under 65, 65-74 and 75 years and over - on the Hamilton Rating Scale for Depression (HRSD), Global Assessment of Functioning scale (GAF) and clinical outcome rating scale. Assessments were performed pre-ECT, immediately post-ECT, 1-3 years later and, for patients suspected of having dementia, 5 years later. RESULTS At post-ECT and follow-up, improvement on HRSD and clinical outcome ratings were comparable for patients in the three age groups. Improvements on GAF were also comparable post-ECT, but not between post-ECT and follow-up. At follow-up, 35.7% of the oldest group had dementia. Importantly, patients who did and did not develop dementia were clinically indistinguishable prior to ECT. The number and severity of common adverse events were similar pre- and post-ECT and were not associated with age. CONCLUSIONS Depressive outcome and adverse effects of ECT are largely independent of age. Older patients receiving ECT appear to have a higher risk of developing dementia, possibly underpinned by cerebrovascular disease.