Poor Sleep at Baseline Predicts Worse Mood Outcomes in Patients With Co-Occurring Bipolar Disorder and Substance Dependence

Abstract

Sleep problems are common in patients with bipolar disorder and have been shown to predict subsequent mood symptoms. Sleep problems have also been shown to lead to worse substance use outcomes in individuals with substance use disorder. However, the relationship between sleep and clinical outcomes in a population with co-occurring bipolar disorder and substance use disorder is unclear. This secondary analysis included 60 outpatients (mean age = 38.1 years; recruited via advertisements, fliers, clinician referrals, and hospital treatment programs) who met DSM-IV criteria for both bipolar disorder and substance use disorder (assessed with the Structured Clinical Interview for DSM-IV Axis I Disorders) and who participated in a randomized clinical trial comparing integrated group therapy for bipolar disorder and substance use disorder to group drug counseling for substance use disorder alone. A 12-week treatment period preceded a 24-week follow-up. Poor sleep was assessed with the Pittsburgh Sleep Quality Index, which provides 7 component subscores and an overall sleep score. Data were collected from August 2003 through April 2007. When analyses were controlled for baseline mood, substance use, and treatment condition, baseline sleep score predicted mood over the course of the 12-week treatment (β = 0.28; P < .05) and 24-week follow-up (β = 0.46; P < .01): worse sleep was associated with worse mood outcomes. Sleep was not associated with substance use outcomes. Impaired sleep is a prognostic factor for mood outcomes in patients with co-occurring bipolar and substance use disorders. Further investigation is warranted into the long-term clinical outcomes of poor sleep in this population with co-occurring bipolar disorder and substance use disorder so that appropriate interventions can be developed.