Automated Eligibility Screening Improves Patient Selection For Trials

By Katie Wike, contributing writer

Using technology to match patients with clinical trials saved doctors a hefty workload in a study conducted at Cincinnati Children’s Hospital Medical Center.

A study recently published in the The Journal of the American Medical Informatics Association has found automated eligibility screening for clinical trials in more time efficient and has reduced physician workload.

The authors of the study write, “In most circumstances, ES (eligibility screening) is still conducted manually. Manual screening typically requires a lengthy review of patient records, a cumbersome process that creates a significant financial burden for an institution. In a busy clinical care center, the task of screening patients for clinical trials without bias is labor-intensive.”

From 2010 to 2012, researchers used automated screening to select candidates for 13 different trials. Fierce EMR explains that they used data fields from EHRs, including demographics, laboratory data, and clinical notes to assess more than 202,000 patients that had visited Cincinnati Children's Hospital Medical Center's emergency department.  iHealth Beat reports researchers used multiple methods for selection; information extraction, machine learning, natural language processing.

They compared the results of the automated system to the selections of two board-certified pediatric emergency medicine physicians who had reviewed charts.

Not only did the automated system result in a 92 percent workload reduction and have a mean of 62.9 percent precision, but researchers found it to be 450 percent more efficient. “We hypothesize that the automated ES approach, when rolled out for production, will have potential for significant impact in reduction of time and effort for executing clinical research, particularly as important new initiatives greatly expand the number of, and access to, potential clinical trials for patients,” write researchers.