Study protocol of the Diabetes and Depression Study (DAD): a multi-center randomized controlled trial to compare the efficacy of a diabetes-specific cognitive behavioral group therapy versus sertraline in patients with major depression and poorly controlled diabetes mellitus

Background: Depression is common in diabetes and associated with hyperglycemia, diabetes related complications and mortality. No single intervention has been identified that consistently leads to simultaneous improvement of depression and glycemic control. Our aim is to analyze the efficacy of a diabetes-specific cognitive behavioral group therapy (CBT) compared to sertraline (SER) in adults with depression and poorly controlled diabetes. Methods/Design: This study is a multi-center parallel arm randomized controlled trial currently in its data analysis phase. We included 251 patients in 70 secondary care centers across Germany. Key inclusion criteria were: type 1 or 2 diabetes, major depression (diagnosed with the Structured Clinical Interview for DSM-IV, SCID) and hemoglobin A1(C) >7.5% despite current insulin therapy. During the initial phase, patients received either 50-200 mg/d sertraline or 10 CBT sessions aiming at the remission of depression and enhanced adherence to diabetes treatment and coping with diabetes. Both groups received diabetes treatment as usual. After 12 weeks of this initial open-label therapy, only the treatment-responders (50% depression symptoms reduction, Hamilton Depression Rating Scale, 17-item version [HAMD]) were included in the subsequent one year study phase and represented the primary analysis population. CBT-responders received no further treatment, while SER-responders obtained a continuous, flexible-dose SER regimen as relapse prevention. Adherence to treatment was analyzed using therapeutic drug monitoring (measurement of sertraline and N-desmethylsertraline concentrations in blood serum) and by counting the numbers of CBT sessions received. Outcome assessments were conducted by trained psychologists blinded to group assignment. Group differences in HbA1c (primary outcome) and depression (HAMD, secondary outcome) between 1-year follow-up and baseline will be analyzed by ANCOVA controlling for baseline values. As primary hypothesis we expect that CBT leads to significantly greater improvement of glycemic control in the one year follow-up in treatment responders of the short term phase. Discussion: The DAD study is the first randomized controlled trial comparing antidepressants to a psychological treatment in diabetes patients with depression.