Compliance With Antidepressant Medication in the Treatment of Major Depressive Disorder in Primary Care: A Randomized Comparison of Fluoxetine and a Tricyclic Antidepressant

OBJECTIVE: Many claims have been made for superior compliance with selective serotonin reuptake inhibitors (SSRIs) compared with tricyclic antidepressants, but to date meta-analyses have not confirmed reduced dropouts in randomized controlled trials. The authors used a randomized study design to evaluate differential compliance with antidepressant medications in a primary care setting. METHOD: A total of 152 patients treated in 10 primary care practices in the United Kingdom were included in a randomized, open-label, parallel-group study of fluoxetine and dothiepin at therapeutic doses for 12 weeks. Compliance was assessed by using pill count, patient questionnaires, and the Medication Event Monitoring System. RESULTS: The level of compliance with fluoxetine was numerically higher than the level of compliance with dothiepin on all three primary outcome measures, although the differences were not significant. In a secondary analysis using data from the Medication Event Monitoring System, both a survival analysis for length of time without a gap in medicine taking and a derived compliance index showed a significant advantage to fluoxetine. Patients in the fluoxetine group reported superior response on the health transition scale of the 36-item Short-Form Health Survey Questionnaire and numerically greater improvement on the Hamilton Depression Rating Scale. In both treatment arms patients with a superior compliance index were more likely to have improved in Hamilton depression scale scores by the last study visit. CONCLUSIONS: This study supports recent meta-analyses of SSRIs versus tricyclic antidepressants in finding no significant differences in crude indices of compliance between fluoxetine and dothiepin, despite marked differences in side effect profile and dose regimen. However, both a survival analysis and a new measure that takes account of prolonged periods of noncompliance distinguished between the treatments and was associated with improvement in both groups.