Comparison of balanced and random allocation in clinical trials: A simulation study

Background: Before analysing the results of a randomised controlled clinica l trial in which 200 children were balanced over five prognostic factors, w e wanted to know the efficiency of balanced allocation compared to simple r andomisation as well as the efficiency of adjusted as compared to unadjuste d statistical analysis in small and large sample sizes. Methods: A simulati on study with 1000 replications of each assignment was performed for both r elatively large trials (n = 100) and for small trials (n = 20). Four option s for the design and analysis were studied: (1) simple randomisation with s imple univariate analysis, (2) simple randomisation with multivariate model ling, (3) balanced allocation with simple univariate analysis and (4) balan ced allocation with multivariate modelling. In addition, we also considered the effect of an unmeasured covariable, which was either uncorrelated or c orrelated with another covariate. Results/conclusion: The simulations show that a combination of balanced allocation and multivariate analysis as comp ared to simple randomisation and multivariate analysis leads to more valid and precise treatment effects as well as to smaller confidence intervals, e specially in small trials (n = 20). Multivariate analysis with all known pr ognostic factors produces on average smaller Type I errors and Type II erro rs in balanced allocation compared to simple randomisation. If an unmeasure d covariate is strongly correlated with another covariate the treatment eff ect is estimated more precisely as compared to an unmeasured covariate that is not correlate or less strongly correlated.