To the Editor: In their randomized discontinuation trial, published in the April 2008 issue of the Journal, Graham J. Emslie, M.D., et al. (1) demonstrated that depressed children and adolescents who receive continuation treatment with fluoxetine will remain in remission longer than those whose treatment is discontinued after 12 weeks. Although the trial was rigorously conducted and reported, we have concerns regarding the choice of design, which is not applicable to all depressed children and adolescents, including those who recover spontaneously. Discontinuation trials apply only to patients who respond to medication, and they can be difficult to identify in practice (2).
To establish the practical benefits of a treatment in populations with high spontaneous response rates (e.g., children, adolescents), a different design is needed in which all patients are randomly assigned to active treatment or placebo and followed for the full study duration. This strategy, also known as the classic two-arm parallel randomized controlled trial, has been used to definitively establish the efficacy of selective serotonin reuptake inhibitors in depressed adults for up to 9 months (3). Until similar trials have been conducted in children and adolescents, long-term treatment decisions will be based on a select study population of patients who respond to drug treatment. Little guidance is provided for the identification and management of those who recover without drug therapy.
Dr. Deshauer has received funding from the Ontario Mental Health Foundation. Dr. Grimshaw serves as Chairperson of the Canadian Institute of Health Research in Health Knowledge Transfer and Uptake; he is also the Director of the Canadian Cochrane Network and Centre.
This letter (doi: 10.1176/appi.ajp.2008.08040531) was accepted for publication in May 2008.