OBJECTIVES: This study compared self-reported pain scores among patients with major depressive disorder (MDD) in the U.S. Veterans Affairs (VA) health system treated with duloxetine versus other antidepressants. METHODS: The electronic medical records between October 1, 2004 and October 31, 2008 were obtained from the VA Veterans Integrated Service Network 16 data warehouse. All patients treated with either duloxetine monotherapy or other antidepressants (non-duloxetine) were selected. The first dispense date of the index agent was defined as the index date. All patients must have: 1) 1+ prior MDD diagnosis (ICD-9-CM: 296.2 or 296.3); 2) no prior diabetes (ICD-9-CM: 250.xx) or bipolar (ICD-9-CM: 296.4x-296.8x) diagnosis; and 3) self-reported pain score measured within 60 days both before the index date (baseline pain score) and after the last dispense date of the index antidepressants during the 12-month post-index period. The non-duloxetine-treated patients were matched to the duloxetine-treated patients via propensity scoring (1:1 ratio), controlling for demographics, comorbidities, prior opioid use, prior healthcare utilization, and baseline pain scores. Opioid utilization and pain scores over the 12-month post-index period were examined between cohorts. RESULTS: The study sample included 210 duloxetine- and 210 non-duloxetine-treated patients. Significantly less duloxetine-treated patients than non-duloxetine-treated patients used opioids (20.1%; vs. 34.3%, p=.002) over the 12-month post-index period. Both cohorts had similar morphine-equivalent opioid daily use and pain scores during the follow-up period. Controlling for baseline pain scores and medication duration, duloxetine-treated patients had 1.58 points (p=0.007) lower pain score than non-duloxetine-treated patients among individuals who had any opioid use during the 12-month post-index period. CONCLUSIONS: Among similar VA patients with MDD, patients treated with duloxetine were associated with lower opioid use than those treated with other antidepressants. Among individuals with any opioid use over the 12-month post-index period, duloxetine-treated patients had better pain scores than non-duloxetine-treated patients.