Clinical Trial Covariates Guidance
FDA has published a draft guidance, Adjusting for Covariates in Randomized Clinical Trials for Drugs and Biologics with Continuous Outcomes, to give the agency’s recommendations for adjusting for covariates in randomized clinical trials with continuous endpoints that are appropriate for analysis with normal-theory methods, such as the two-sample t-test. The document says nonparametric methods, categorical outcomes, and survival methods, among others, are outside its scope, although some of the same principles might apply to those methods as well.
“The target population for a new drug usually includes patients with diverse prognostic factors,” the guidance says, “and the population studied in clinical trials should reflect this diversity. However, potential baseline differences between treatment groups in prognostic factors increase the variability of estimates of treatment effects and reduce the power of significance tests. Incorporating prognostic factors in the primary statistical analysis of clinical trial data can result in a more efficient use of data to demonstrate and quantify the effects of treatment.”