Skip to main content

Table 3 Mean changes from baseline to last observation in all studies and the subset of positive studies

From: Duloxetine in the treatment of Major Depressive Disorder: A comparison of efficacy in patients with and without melancholic features

  Melancholic Status Mean Change (SD) p-valuea p-valueb
   Duloxetine Placebo   
HAMD 17 Total Score
All Studies MEL (n = 1249) -8.97 (7.36) -6.57 (7.24) <.001  
  NON-MEL (n = 621) -8.25 (6.55) -6.20 (6.22) <.001 .651
Positive Studies MEL (n = 984) -9.83 (7.20) -7.03 (7.06) <.001  
  NON-MEL (n = 474) -9.15 (6.23) -6.47 (6.27) <.001 .781
CGI-Severity
All Studies MEL (n = 1251) -1.52 (1.25) -1.11 (1.21) <.001  
  NON-MEL (n = 621) -1.47 (1.23) -1.14 (1.19) .001 .511
Positive Studies MEL (n = 986) -1.62 (1.25) -1.16 (1.20) <.001  
  NON-MEL (n = 474) -1.54 (1.13) -1.15 (1.17) <.001 .679
PGI-Improvement Mean (SD)
All Studies MEL (n = 1249) 2.67 (1.27) 3.07 (1.27) <.001  
  NON-MEL (n = 621) 2.71 (1.31) 3.17 (1.27) <.001 .865
Positive Studies MEL (n = 984) 2.64 (1.26) 3.09 (1.27) <.001  
  NON-MEL (n = 474) 2.70 (1.28) 3.18 (1.32) <.001 .912
  1. a. p-value for duloxetine vs. placebo
  2. b. p-value for treatment-by-melancholia status interaction
  3. † Mean score (SD). Lower mean score indicates greater improvement.