TL24: Missing Data Analysis Planning in Late-State Clinical Trials: A Check-up on Current Practices
*Davis Gates, Merck 

Keywords: missing data, sensitivity analysis, multiple imputation, informative missing

There has been an extensive amount of discussion between FDA and Industry on the requirement to address missing data in late-stage clinical trials. The FDA has recommended methods of study conduct to minimize the quantity of missing data in the reporting database upon trial completion. However, methods of missing data sensitivity analyses are still required to complete the discussion on the impact of missing data. In response, the use of SAS multiple imputation procedures, coupled with additional programming designed to address informative missing data, i.e. missing not at random observations, have been provided in statistical analysis plans across therapy areas. This round table would discuss the progress industry has made to address the FDA’s request to handle missing data issues, as well as the impact missing data has had on the approvals of new drugs and vaccines.