Semi-parametric approaches based on generalized estimating equations (GEE) are widely used to analyze correlated outcomes in longitudinal settings. In this paper, we present a package CRTgeeDR developed for cluster randomized trials with missing data (CRTs). For use of inverse probability weighting to adjust for missing data in cluster randomized trials, we show that other software lead to biased estimation for non-independence working correlation structure. CRTgeeDR solves this problem. We also extend the ability of existing packages to allow augmented Doubly Robust GEE estimation (DR). Simulation studies demonstrate the consistency of estimators implemented in CRTgeeDR compared to packages such as geepack and the gains associated with the use of the DR for analyzing a binary outcome using a logistic regression. Finally, we illustrate the method on data from a sanitation CRT in developing countries.
CRTgeeDR, gee, geepack, geeM, ipw, drgee, CausalGAM, tmle, tmlenet, numDeriv, geesmv
Text and figures are licensed under Creative Commons Attribution CC BY 4.0. The figures that have been reused from other sources don't fall under this license and can be recognized by a note in their caption: "Figure from ...".
For attribution, please cite this work as
Prague, et al., "CRTgeeDR: an R Package for Doubly Robust Generalized Estimating Equations Estimations in Cluster Randomized Trials with Missing Data", The R Journal, 2017
BibTeX citation
@article{RJ-2017-041,
author = {Prague, Melanie and Wang, Rui and Gruttola, Victor De},
title = {CRTgeeDR: an R Package for Doubly Robust Generalized Estimating Equations Estimations in Cluster Randomized Trials with Missing Data},
journal = {The R Journal},
year = {2017},
note = {https://doi.org/10.32614/RJ-2017-041},
doi = {10.32614/RJ-2017-041},
volume = {9},
issue = {2},
issn = {2073-4859},
pages = {105-115}
}