Skip to main content

Table 2 Full data estimate of cure at day 28 follow-up using the Kaplan-Meier method and cured proportion

From: Dealing with indeterminate outcomes in antimalarial drug efficacy trials: a comparison between complete case analysis, multiple imputation and inverse probability weighting

TreatmentEstimate95% Confidence IntervalSESE†
Full data K-M a
 AL0.9600.948—0.9720.00610.1559
 ASAQ0.9790.969—0.9890.00490.2367
 DP0.9830.977—0.9900.00350.2082
Full data cured proportion b
 AL0.9640.951—0.9730.00560.1562
 ASAQ0.9800.968—0.9870.00470.2357
 DP0.9840.975—0.9890.00340.2132
  1. AL artemether-lumefantrine , ASAQ artesunate-amodiaquine , DP dihydroartemisinin-piperaquine , SE standard error
  2. † Standard error after complementary log-log transformation. The cloglog transformation was applied as the MI estimates were computed on complementary log-log scale for the application of Rubin’s combination rules to be valid.
  3. a For the derivation of the K-M estimates, new infections were considered as censored on the day of recurrence
  4. bThe estimates of cured proportion (total cured/total number of patients treated) was computed by considering those with new infections as cured. The variance for cured proportion (\( \hat{p} \)) for a total number of patients (n) was calculated as \( \hat{p}\left(1-\hat{p}\right)/n \). The variance was converted to the cloglog scale using the delta method presented in Additional file 1, Section C. The 95% confidence interval was derived using Wilson’s method using binom.confint routine in binom package R.