OBJECTIVES: To examine the relationship between average daily dose (ADD) of duloxetine, medication adherence, and healthcare costs among patients with major depressive disorder (MDD). METHODS: Using a large US administrative claims database, commercially-insured patients aged 18-64 who initiated duloxetine in 2007 were identified. Initiation was defined as no duxolextine treatment over the prior 90 days, with the corresponding duloxetine dispense date as the index date. Patients were required to have 12-month continuous enrollment preceding and following duloxetine initiation, and at least 1 MDD diagnosis during the 12 months prior to, or on the index date. Five cohorts were constructed based on ADD: 20-29, 30, 31-59, 60, and >60 mg/day. Healthcare costs and duloxetine adherence (measured via medication possession ratio), were assessed over the 12-month post-index period. Multivariate regression models adjusting for differences in demographic and clinical characteristics were used to examine the association between ADD, adherence and healthcare costs. RESULTS: The study sample included 8572 MDD patients with 60.7% initiating duloxetine at 60 mg/day. During the 12-month post-index period, 3.4% of patients had an ADD of 20-29, 9.0% of 30, 21.3% of 31-59, 37.2% of 60, and 29.1% of >60 mg/day. Controlling for demographics and clinical characteristics, patients in the 31-59 or >60 mg/day cohorts (≤30 mg/day) were more (less) likely to adhere to duloxetine than those with 60 mg/day (all p<0.05). Compared with patients with an ADD of 60 mg/day, those with an ADD >60 mg/day had significantly higher total healthcare costs ($3,947, p<0.05), while those with an ADD of 30 mg/day had significantly lower costs (-$1,109, p<0.05). CONCLUSIONS: The findings suggest that having an ADD of 60 mg/day for duloxetine among commercially insured MDD patients is associated with improved medication adherence relative to low duloxetine doses and lower healthcare costs relative to high duloxetine doses.