Skip to main content

Table 4 MAR and efficacy rate 98 % versus 95 % (RD 0.030): estimated efficacy differences, coverage and bias for 5 %, 15 % and 30 % averages of number of simulated data sets that converged of the 5000 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

No. of data sets*

RD (RMSE)

Coverage

Bias

All outcomes recorded:

5,000

0.030 (0.026)

0.939

0.000

5 % of outcomes missing

    

CC

5,000

0.030 (0.026)

0.940

0.000

MI: wt, hb, age, para

4,982

0.027 (0.027)

0.951

-0.003

MI: hb, age, para

4,988

0.027 (0.027)

0.955

-0.003

MI: hb, age, para, group

4,989

0.029 (0.027)

0.946

-0.001

MI: wt, hb, age, para, group

4,986

0.029 (0.027)

0.947

-0.001

15 % of outcomes missing

    

CC

5000

0.030 (0.028)

0.942

0.000

MI: wt, hb, age, para

4969

0.025 (0.029)

0.970

-0.005

MI: hb, age, para

4983

0.025 (0.029)

0.963

-0.005

MI: hb, age, para, group

4981

0.030 (0.030)

0.965

0.000

MI: wt, hb, age, para, group

4982

0.030 (0.030)

0.965

0.000

30 % of outcomes missing

    

CC

5,000

0.030 (0.030)

0.937

0.000

MI: wt, hb, age, para

4,893

0.020 (0.033)

0.970

-0.010

MI: hb, age, para

4,949

0.021 (0.033)

0.98

-0.009

MI: hb, age, para, group

4,945

0.030 (0.037)

0.967

0.000

MI: wt, hb, age, para, group

4,938

0.031 (0.037)

0.971

+0.001

  1. *Number of data sets for which convergent analysis was achieved