BackgroundRelapse of major depressive disorder (MDD) is a common clinical problem. This study was designed to determine whether residual sleep disturbance (insomnia and hypersomnia) predict risk of relapse during the continuation and maintenance treatment of MDD.MethodsA total of 570 patients with MDD were treated with open-label, flexible dose fluoxetine (range 20 to 60 mg; mean dose = 45.8 mg/day; SD = 15.1) for 12 weeks. Under double blind conditions, 262 patients who achieved clinical response were randomly assigned to continue fluoxetine or to switch to placebo for 52 weeks or until relapse. Residual sleep disturbance during the baseline visit of the double-blind phase was assessed using items 4, 5, 6 (insomnia) and 22, 23, 24 (hypersomnia) of the Hamilton Depression Rating Scale (HDRS). Survival analysis was utilized to determine the effect of residual sleep disturbance on risk of relapse.ResultsThe severities of early (P > 0.05), middle (P > 0.05), late (P > 0.05), or total (P > 0.05) residual insomnia were not found to significantly predict risk of relapse during continuation and maintenance-phase treatment. Similarly, the severities of early bedtime (P > 0.05), oversleeping (P > 0.05), napping (P > 0.05), or total (P > 0.05) residual hypersomnia were not found to significantly predict risk of relapse during continuation and maintenance-phase treatment.ConclusionThe present study did not identify the severity of residual sleep disturbance among fluoxetine responders to predict risk of MDD relapse. The size of our sample may have precluded us from identifying more modest effects of residual sleep disturbance on the risk of relapse in MDD patients. Future studies are needed to further explore the relationship between residual sleep disturbance and relapse in MDD.Trial RegistrationClinicalTrials.gov Identifier: NCT00427128