OBJECTIVES: Duloxetine, a serotonin and norepinephrine reuptake inhibitor, has been approved for the treatment of both major depressive disorder (MDD) and certain chronic pain-related diseases (CPD). This study examined the association between adherence to duloxetine and hospital utilization among MDD patients with CPD. METHODS: This is a retrospective cohort study analyzing data from the MarketScan commercial databases. Patients were included in the analyses if they used duloxetine during the index period of July 1, 2006-June 30, 2007 and had no record of duloxetine use during a 3-month washout period. Inclusion criteria required MDD diagnosis and at least one CPD of interest (fibromyalgia, diabetes with neurological manifestations, low back pain, headache, and osteoarthritis). Patients were followed up 12 months after index date. Adherence was defined as medication possession ratio (MPR) of 80% or higher. Hospital utilization included emergency room visits and hospitalizations. Psychiatric hospitalizations were based on principal diagnosis codes for admissions. Logistic regression and negative binomial regression models were used to adjust for patient characteristics. RESULTS: Compared to those with MPR<80% (n=2,988), patients adherent to duloxetine (n=2,589) had fewer emergency room visits (1.51 vs. 2.07; p<0.0001), lower likelihood of emergency room visit (36.85% vs. 41.57%; p=0.0003), fewer all-cause hospitalizations (0.34 vs. 0.46; p<0.0001), fewer psychiatric hospitalizations (0.10 vs. 0.15; p<0.0001), fewer all-cause hospitalization days (1.46 vs. 2.43; p<0.0001), fewer psychiatric hospitalization days (0.53 vs. 1.04; p=0.0001), lower likelihood of hospitalization for any cause (20.97% vs. 25.07%; p=0.0003) and lower likelihood of psychiatric hospitalization (6.80% vs. 8.97%; p=0.0028). These differences were essentially the same after adjusting for patient characteristics. CONCLUSIONS: Adherence to duloxetine was associated with lower hospital utilization among MDD patients with CPD, suggesting the importance of improving patient adherence to duloxetine. Future studies should examine whether the lower hospital utilization associated with duloxetine adherence translates to lower costs.