A Simulation-Based Comparison of Covariate Adjustment Methods for the Analysis of Randomized Controlled Trials.

Pierre Chaussé,Jin Liu,George Luta

doi:10.3390/ijerph13040414

Abstract

Covariate adjustment methods are frequently used when baseline covariate information is available for randomized controlled trials. Using a simulation study, we compared the analysis of covariance (ANCOVA) with three nonparametric covariate adjustment methods with respect to point and interval estimation for the difference between means. The three alternative methods were based on important members of the generalized empirical likelihood (GEL) family, specifically on the empirical likelihood (EL) method, the exponential tilting (ET) method, and the continuous updated estimator (CUE) method. Two criteria were considered for the comparison of the four statistical methods: the root mean squared error and the empirical coverage of the nominal 95% confidence intervals for the difference between means. Based on the results of the simulation study, for sensitivity analysis purposes, we recommend the use of ANCOVA (with robust standard errors when heteroscedasticity is present) together with the CUE-based covariate adjustment method.

Highlights

When baseline covariate information is available for randomized controlled trials in the areas of environmental research and public health, statistical methods that perform covariate adjustment are usually employed
There are two main reasons to use covariate adjustment methods for the statistical analysis of randomized experiments: one is variance reduction for the estimators for the parameters of interest, which will lead to narrower confidence intervals and more powerful statistical tests; the other is to achieve the equivalence of the treatment groups that is expected as a consequence of randomization [1]
We evaluated the usefulness of three important members of the generalized empirical likelihood (GEL) family, including the EL, exponential tilting (ET), and continuous updated estimator (CUE) methods, with respect to performing covariate adjustment for randomized studies in environmental research and public health

Summary

Introduction

When baseline covariate information is available for randomized controlled trials in the areas of environmental research and public health, statistical methods that perform covariate adjustment are usually employed. We note that under Neyman’s causal model for randomization inference, the use of ordinary least squares regression covariate adjustment may increase the asymptotic variance in some cases [2]. This issue can be addressed by the inclusion of treatment by covariate interactions, or by the use of robust standard error estimators [3]. The intervention was terminated when the children were 24-months of age. The outcome for this experimental study was

Methods

Results

Conclusion