Skip to main content

Table 3 MNAR and efficacy rate 85 % versus 60 % (RD 0.250): estimated efficacy differences, coverage and bias for 5 %, 15 % and 30 % averages of 5000 simulated data sets, 50 imputations

From: Is using multiple imputation better than complete case analysis for estimating a prevalence (risk) difference in randomized controlled trials when binary outcome observations are missing?

Model RD (RMSE) Coverage Bias
All outcomes recorded: 0.250 (0.061) 0.950 0.000
5 % of outcomes missing    
CC 0.258 (0.063) 0.942 +0.008
MI: wt, hb, age, para 0.244 (0.064) 0.962 -0.006
MI: hb, age, para 0.245 (0.064) 0.958 -0.005
MI: hb, age, para, group 0.257 (0.063) 0.946 +0.007
MI: wt, hb, age, para, group 0.256 (0.063) 0.946 +0.006
15 % of outcomes missing    
CC 0.274 (0.068) 0.932 +0.024
MI: wt, hb, age, para 0.231 (0.069) 0.975 -0.019
MI: hb, age, para 0.232 (0.069) 0.975 -0.018
MI: hb, age, para, group 0.272 (0.068) 0.931 +0.028
MI: wt, hb, age, para, group 0.273 (0.068) 0.928 +0.023
30 % of outcomes missing    
CC 0.298 (0.078) 0.895 +0.048
MI: wt, hb, age, para 0.206 (0.077) 0.974 -0.044
MI: hb, age, para 0.206 (0.077) 0.974 -0.044
MI: hb, age, para, group 0.295 (0.078) 0.901 +0.045
MI: wt, hb, age, para, group 0.296 (0.077) 0.898 +0.046