Minimizing Missing Data in Clinical Trials: Design, Operation, and Regulatory Considerations

Abstract

The ultimate question to be answered in a clinical trial is “What benefit/risk has the drug caused to the patient?” To answer this question, it is critical to avoid missing data on outcome that measures benefit and risk from the beginning. This article emphasizes that when a patient discontinues the study treatment, instead of trying to impute the missing outcome values, effort should be made to follow the patient until the endpoint or the end of the trial. The information in the follow-up period can be used to describe the benefit/risk to those patients who would discontinue treatment in the actual clinical practice, while imputing values that are not observed, yet may never exist, can only answer the unrealistic question,“ What would have happened had the subject not discontinued?” Minimizing missing data is possible through the joint efforts of drug companies and regulatory agencies. Drug companies can improve their efforts in study design and operation, including off-treatment follow-up, prioritization of information collection, endpoint selection, and detailed documentation of reasons for missing data. Regulatory agencies can provide general guidance and convey specific comments for improvement through protocol review regarding missing data, and can ensure proper labeling language that reflects the extent of missing data and the reliability of conclusions.