Skip to main content

Table 4 Estimated effects for models M1-M5 in the per-protocol (PP) analysis

From: Multiple treatment comparisons in a series of anti-malarial trials with an ordinal primary outcome and repeated treatment evaluations

  M1 M2 M3   M4   M5  
  Est SD Est SD Est SD   Est SD   Est SD  
Study 3-1 2.47 1.01 −1.26 1.61 0.10 1.20   1.70 0.93   1.71 0.98  
Study 4-1 1.03 1.26 1.33 1.40 0.78 1.04   1.07 1.04   1.03 1.12  
Study 5-1 1.08 1.07 −1.88 1.23 −0.66 1.03   0.49 0.76   0.27 0.85  
AQ 0.68 1.23 1.35 1.39 0.39 0.92   0.35 0.86   0.45 0.97  
AMLM 2.04 1.30 1.50 1.20 1.35 1.07   1.65 1.15   1.71 1.23  
ASCD −1.76 1.37 −1.97 1.41 −1.25 1.24   −1.56 1.22   −1.69 1.26  
ASSP −1.77 1.10 2.31* 1.09 1.59 0.95   2.22* 1.07   2.00 1.20  
DHPP 3.09* 0.74 2.93* 0.77 2.88* 0.68   2.27* 0.94   2.04* 1.04  
γ D 21 (Day21-14) −4.17* 0.60 −3.14* 1.50 −2.99* 0.72   −3.28* 0.57   −3.46 0.61  
γ D 28 (Day28-14) −3.87* 0.98 −3.04* 1.53 −3.03* 1.10   −3.69* 0.60   −3.91* 0.63  
τ 6.57* 1.73 6.89* 1.75 7.52* 1.16   7.77* 1.00   7.40* 1.29  
η    0.80* 0.26        0.34 0.29  
DIC 636.69 371.24 302.32   536.82   532.07  
  1. Study # 2, excluded. Est=estimation; SD=standard deviation; *: significant effect at the 5% level. τ: standard error of the subject random effect in modeling the logit- cumulated categorical variable. η: standard error of the subject random effect in modeling the log- variance (model 2).