Emil Kraepelin’s demarcation between dementia praecox and manic depressive illness defined affective disorder as a remitting and recurring disease. He considered that only long-term outcome was useful in assessing accuracy of diagnosis and treatment response in patients.1 The more recent interest in the outcome of single mood episodes probably indicates motives to register and market drugs rather than assisting clinical practice. This interest has resulted in many 4–8 week randomised trials but few well designed long-term studies in patients with depression.