Acute and Long-term Treatment of Late-Life Major Depressive Disorder: Duloxetine Versus Placebo

Objective To compare the efficacy of duloxetine with placebo on depression in elderly patients with major depressive disorder. Design Multicenter, 24-week (12-week short-term and 12-week continuation), randomized, placebo-controlled, double-blind trial. Setting United States, France, Mexico, Puerto Rico. Participants Age 65 years or more with major depressive disorder diagnosis (one or more previous episode); Mini-Mental State Examination score ≥20; Montgomery-Asberg Depression Rating Scale total score ≥20. Intervention Duloxetine 60 or 120 mg/day or placebo; placebo rescue possible. Measurements Primary -Maier subscale of the 17-item Hamilton Depression Rating Scale (HAMD-17) at week 12. Secondary -Geriatric Depression Scale, HAMD-17 total score, cognitive measures, Brief Pain Inventory (BPI), Numeric Rating Scales (NRS) for pain, Clinical Global Impression-Severity scale, Patient Global Impression of Improvement in acute phase and acute plus continuation phase of treatment. Results Compared with placebo, duloxetine did not show significantly greater improvement from baseline on Maier subscale at 12 weeks, but did show significantly greater improvement at weeks 4, 8, 16, and 20. Similar patterns for Geriatric Depression Scale and Clinical Global Impression-Severity scale emerged, with significance also seen at week 24. There was a significant treatment effect for all BPI items and 4 of 6 NRS pain measures in the acute phase, most BPI items and half of the NRS measures in the continuation phase. More duloxetine-treated patients completed the study (63% versus 55%). A significantly higher percentage of duloxetine-treated patients versus placebo discontinued due to adverse event (15.3% versus 5.8%). Conclusions Although the antidepressant efficacy of duloxetine was not confirmed by the primary outcome, several secondary measures at multiple time points suggested efficacy. Duloxetine had significant and meaningful beneficial effects on pain.