Prospective EHR-Based Clinical Trials: The Challenge of Missing Data

Excerpt

This discussion focuses on the challenges of using prospectively collected electronic health record (EHR) data as outcomes in clinical trials, with a particular emphasis on the issue of missing data. Our discussion is motivated by the article in this issue: ‘Translating the Hemoglobin A1C with More Easily Understood Feedback: A Randomized Controlled Trial’ by Gopalan et al.1 In the spirit of open science, the authors generously shared their study protocol, statistical analysis plan and analysis data set. Using their data set, we conducted analyses to help emphasize important statistical issues. This editorial should not be considered a criticism of their paper; rather, their study is used as a reference to expand on the challenges of missing data in EHRs and to provide suggestions for future studies. …