Clinical Trial Eligibility Optimization Could Aid Recruitment, Add Robustness To COVID-19 Trials

Clinical Trial Eligibility Optimization Could Aid Recruitment,
Add Robustness To COVID-19 Trials

An abundance of clinical trials has been initiated in response to the ongoing COVID-19 pandemic throughout the world, but questions have arisen about the robustness of trial designs, particularly in relation to insufficient power or sample sizes, suboptimal reporting, and unclear randomization, among other challenges.

Researchers at Columbia University attempted to answer some of those questions by measuring the impact of eligibility criteria on recruitment and observable clinical outcomes of COVID-19 clinical trials using electronic health record (EHR) data.

Their work, entitled “Towards Clinical Data-Driven Eligibility Criteria Optimization for Interventional COVID-19 Clinical Trials,” was recently published in JAMIA.

Jaehyun Kim provided this presentation on “Towards Clinical Data-Driven Eligibility Criteria Optimization for Interventional COVID-19 Clinical Trials” at the 2020 OHDSI Global Symposium.

Eligibility criteria impacts both patient accrual and the incidence of severe endpoints, and it can limit a large number of patients and the observation of outcome events throughout trials. This can hinder the robustness required to accurately assess the efficacy of COVID-19 treatments in these trials.

The research team studied 3,251 patients diagnosed with COVID-19 in the Columbia University Irving Medical Center (CUIMC) EHR and determined that just under 25 percent (690) would qualify for the most common eligibility criteria, while the total number of outcome events was 153 (22.2%).

The team used real-world data to modify criteria such as age, renal and liver function, and generated two sets of candidate eligibility criteria that allowed for more than 28% of qualifiers, as well as a higher total of overall outcomes, including one that nearly reached 27% (187/693). These changes also allowed for commonly excluded groups, such as pediatric and pregnant patients, to be included in these studies.

“A large number of clinical trials have been launched in a short time in response to the global COVID-19 health crisis, but it remains unknown how inclusive these trials are and how well they are powered by their sample sizes,” said lead author Jaehyun Kim, a postdoctoral research scientist in Dr. Chunhua Weng’s lab in the Columbia Department of Biomedical Informatics. “Our study uses electronic health records data of COVID-19 patients to assess the influence of COVID-19 trials’ eligibility criteria on recruitment and observed outcome measures. We contribute a novel method for optimizing eligibility criteria specification using EHR data.”

This data-driven eligibility criteria design can positively impact clinical trial potential by optimizing criteria specification (while providing rationale for modifications), minimize the number of patients needed to see a similar total of events, and improve the overall efficiency and cost-effectiveness of enrollment.

This study was sponsored by National Library of Medicine grant 5R01LM009886-11 and National Center for Advancing Clinical and Translational Science grant UL1TR001873 and 1OT2TR003434-01.

Chunhua Weng is a member of the Data Science Institute at Columbia University.