To improve the precision of estimation and power of testing hypothesis for an unconditional treatment effect in randomized clinical trials with binary outcomes, researchers and regulatory agencies recommend using g-computation as a reliable method of covariate adjustment. However, the practical application of g-computation is hindered by the lack of an explicit robust variance formula that can be used for different unconditional treatment effects of interest. To fill this gap, we provide explicit and robust variance estimators for g-computation estimators and demonstrate through simulations that the variance estimators can be reliably applied in practice.

References

• FDA (2021). Adjusting for covariates in randomized clinical trials for drugs and biological products. Draft Guidance for Industry. Center for Drug Evaluation and Research and Center for Biologics Evaluation and Research, Food and Drug Administration (FDA), U.S. Department of Health and Human Services, May 2021. 
• ICH E9 (R1) (2019). Addendum on estimands and sensitivity analysis in clinical trials to the guideline on statistical principles for clinical trials. International Council for Harmonisation (ICH). 
• Chernozhukov, V., Chetverikov, D., Demirer, M., Duflo, E., Hansen, C., & Newey, W. (2017). Double/debiased/neyman machine learning of treatment effects. American Economic Review, 107(5), 261–265. https://doi.org/10.1257/aer.p20171038
• Freedman, D. A. (2008). Randomization does not justify logistic regression. Statistical Science, 23(2), 237–249. https://doi.org/10.1214/08-STS262 
• Ge, M., Durham, L. K., Meyer, R. D., Xie, W., & Thomas, N. (2011). Covariate-adjusted difference in proportions from clinical trials using logistic regression and weighted risk differences. Drug Information Journal: DIJ/Drug Information Association, 45(4), 481–493. https://doi.org/10.1177/009286151104500409 
• Haldane, S. (1956). The estimation and significant of the logarithm of a ratio of frequencies. Annals of Human Genetics, 20(4), 309–311. https://doi.org/10.1111/ahg.1956.20.issue-4 
• Kennedy, E. H. (2016). Semiparametric theory and empirical processes in causal inference. In Statistical causal inferences and their applications in public health research (pp. 141–167). 
• Lin, W. (2013). Agnostic notes on regression adjustments to experimental data: Reexamining freedman's critique. Annals of Applied Statistics, 7(1), 295–318. https://doi.org/10.1214/12-AOAS583 
• Moore, K. L., & van der Laan, M. J. (2009). Covariate adjustment in randomized trials with binary outcomes: Targeted maximum likelihood estimation. Statistics in Medicine, 28(1), 39–64. https://doi.org/10.1002/sim.v28:1 
• Tsiatis, A. A., Davidian, M., Zhang, M., & Lu, X. (2008). Covariate adjustment for two-sample treatment comparisons in randomized clinical trials: A principled yet flexible approach. Statistics in Medicine, 27(23), 4658–4677. https://doi.org/10.1002/sim.v27:23 
• Woolf, B. (1955). On estimating the relation between blood group and disease. Annals of Human Genetics, 19(4), 251–253. https://doi.org/10.1111/ahg.1955.19.issue-4 
• Yang, L., & Tsiatis, A. A. (2001). Efficiency study of estimators for a treatment effect in a pretest–posttest trial. The American Statistician, 55(4), 314–321. https://doi.org/10.1198/000313001753272466
• Ye, T., Shao, J., Yi, Y., & Zhao, Q. (2023). Toward better practice of covariate adjustment in analyzing randomized clinical trials. Journal of the American Statistical Association, 117, in press. https://doi.org/10.1080/01621459.2022.2049278 
• Ye, T., Yi, Y., & Shao, J. (2022). Inference on average treatment effect under minimization and other covariate-adaptive randomization methods. Biometrika, 109(1), 33–47. https://doi.org/10.1093/biomet/asab015 

To cite this article: Ting Ye, Marlena Bannick, Yanyao Yi & Jun Shao (2023) Robust variance estimation for covariate-adjusted unconditional treatment effect in randomized clinical trials with binary outcomes, Statistical Theory and Related Fields, 7:2, 159-163, DOI: 10.1080/24754269.2023.2205802

To link to this article: https://doi.org/10.1080/24754269.2023.2205802