> ## Content Index
> Fetch the complete content index at: https://www.fdaweb.com/llms.txt
> Use this file to discover other available public pages before exploring further.

# Comments on Clinical Trial Covariates Draft Guidance
- URL: https://www.fdaweb.com/comments-on-clinical-trial-covariates-draft-guidance/
- Published: 2019-07-16T12:00:00.000Z
- Updated: 2026-09-15T01:38:05.000Z
- Author: David McFarland
- Tags: Drugs, #legacy-id-D5144543

Amgen says an FDA draft guidance on adjusting for covariates in randomized clinical trials for drugs and biologics with continuous outcomes is easy to read and concise. In a comment [letter](https://www.regulations.gov/document?D=FDA-2019-D-0934-0017&ref=fdaweb.com), the company says it would be useful to add recommendations on how stratification variables should be handled, more detailed recommendations for based imbalances observed *post hoc*, and how missing baseline covariates should be handled. Amgen also includes line-by-line specific recommendations.

Regeneron [asks](https://www.regulations.gov/document?D=FDA-2019-D-0934-0011&ref=fdaweb.com) that FDA articulate its position on the way to stratify randomization and give recommendations or examples of scenarios when a continuous covariate could be used instead of a stratified randomization. It also asks that the guidance address scenarios when a continuous covariate could be used instead of a stratified randomization.

Janssen [says](https://www.regulations.gov/document?D=FDA-2019-D-0934-0010&ref=fdaweb.com) that when the draft is finalized, it “will help statisticians better understand regulatory expectations around the method of adjusting for covariates in the primary analysis in randomized clinical trials.”

Pfizer’s comment [letter](https://www.regulations.gov/document?D=FDA-2019-D-0934-0009&ref=fdaweb.com) says additional information or clarification is needed about use of covariate adjustment for longitudinal studies; other model deviations and acceptance approaches to reduce sensitivity, agency recommendations considering International Conference on Harmonization E9 guidance on limited existing data to assess the predictive value of a baseline variable, combination of design stratification and ANCOVA, and use of pre-specified variable-selection methods for baseline covariate selection.

Finally, Merck [says](https://www.regulations.gov/document?D=FDA-2019-D-0934-0015&ref=fdaweb.com) the draft is very concise and provides high level recommendations. It lists several topics for which it says additional details would be beneficial to industry.

GlaxoSmithKline submitted a [chart](https://www.regulations.gov/document?D=FDA-2019-D-0934-0016&ref=fdaweb.com) with specific recommendations.