Mount Sinai has reported a jump in enrollment in a Phase III trial of a Bayer drug after it used natural language processing to sift through electronic health records (EHR) in search of eligible participants. The success has prompted Mount Sinai to use the technology across its clinical trials and precision medicine initiatives.
Researchers at Icahn School of Medicine at Mount Sinai committed to widespread use of the technology, Clinithink’s CLiX Enrich for Clinical Trials, after testing it on a study of a Bayer diabetic kidney disease drug, finerenone. The protocol features more than 30 inclusion and exclusion criteria, only seven of which can be identified from the structured data in Mount Sinai’s EHR. This made it hard to spot potential trial participants in the EHR data, contributing to sluggish early enrollment in the study.
“In the four months prior to CLiX Enrich, seven [potential participants] had been found, only one was enrolled. An estimated 50 hours had been spent,” Mount Sinai’s Dr. Steve Coca said at the Clinical Trial Innovation Summit.
Faced with this situation, Mount Sinai tried identifying eligible participants using CLiX Enrich, a tool designed to probe unstructured data that had shown promise when applied retrospectively to one of its studies. Once a user has selected their queries, the Clinithink tool searches through notes and other pieces of unstructured data that make up health records in search of patients who may meet the inclusion/exclusion criteria.
This had a dramatic effect on the enrollment rate. “There’s 97 eligible right now,” Coca said. “There’s even more that we have not even had time to screen ... that may be eligible. Thirteen are going to be enrolled now in this next couple of weeks. It’s actually just become an issue of man or woman power, of getting these patients in, scheduling the screening visits, etcetera.”
Dr. Girish Nadkarni, the lead investigator for the CLiX Enrich evaluation, thinks the tool enabled the Mount Sinai team to do in one week what would have taken four months if performed using earlier, more manual processes. These time savings have convinced Mount Sinai to partner with Clinithink to use the technology more widely. In doing so, Coca thinks the organization can start to extract insights from its wealth of unstructured data.
“There’s so much more rich data that are written in the charts, in progress notes, in the discharge summary, in nurses notes, etcetera. Those don’t always come out as structured data,” Coca said. “We should leverage it because so much of it will apply to those inclusion/exclusion criteria.”
FDA seeks feedback on use of EHR data in clinical research
Yale creates Hugo to make it easier for patients to open EHRs up to researchers
Mount Sinai, Yale roll out updates to Apple ResearchKit programs