International Psychogeriatrics

Research Article

Time to response for duloxetine 60 mg once daily versus placebo in elderly patients with major depressive disorder

Joel Raskina1 c1, Jimmy Y. Xua2 and Daniel K. Kajdasza2

a1 Lilly Research Laboratories, Eli Lilly Canada, Toronto, Canada

a2 Lilly Research Laboratories, Eli Lilly and Company, Indianapolis, U.S.A.


Background: Rapid response to antidepressant therapy is desirable and may be particularly critical in elderly patients with major depressive disorder (MDD).

Methods: Findings are based on post-hoc analyses from a double-blind trial of elderly patients with MDD ≥ 65 years, randomly assigned 2:1 to duloxetine 60 mg QD (N = 207) or placebo (N = 104) for 8 weeks. Depression and pain measures included the Geriatric Depression Scale (GDS), 17-item Hamilton Depression Scale (HAMD17), CGI-Severity, and Visual Analog Scale (VAS) for overall pain. The time to response and remission for duloxetine compared with placebo was evaluated using Cox proportional hazards (PH) modeling, Kaplan-Meier estimation, and categorical repeated measures analysis.

Results: Significant improvements of estimated HAMD17 response and remission rates for duloxetine started at week 2 (P = 0.022 and P = 0.033, respectively). Time to HAMD17 response and remission were significantly shorter for duloxetine versus placebo (P < 0.001 and P = 0.004, respectively). Placebo-referenced duloxetine hazard ratios (HR) for HAMD17 response and remission were 2.03 (P = 0.002) and 2.01 (P = 0.006), respectively. Results for GDS-based response (HR = 1.54, P = 0.023) and remission (HR = 1.54, P = 0.104) rates were consistent with the HAMD17 findings. Patients ≥75 years (n = 93) responded with similar rapidity to duloxetine compared with those <75 years (n = 199, P > 0.10 for all PH treatment-by-age interactions). The placebo-referenced duloxetine HR for time to 50% reduction in overall pain was 1.75 (P = 0.024) for patients with moderate to severe pain.

Conclusion: Duloxetine demonstrated a faster time to antidepressant response and improvement in self-reported pain as compared with placebo.

Clinical trial registry number for this study: NCT00062673, at

(Received December 21 2006)

(Online publication February 06 2007)

(Revised March 05 2007)

(Accepted March 08 2007)

(Online publication June 22 2007)


c1 Correspondence should be addressed to: Dr. Joel Raskin, Lilly Research Laboratories, Eli Lilly Canada, 3650 Danforth Ave., Toronto, Ontario, Canada MIN 2E8. Phone: +1 416 699 7260; Fax: +1 416 699 7352. Email: