Study: Automated Screening Improves Clinical Trial Patient Matching
July 18, 2014 in News
Using automated eligibility screening to match patients with clinical trials greatly reduced physician workload and improved efficiency compared with a manual process, according to a study published in the Journal of the American Medical Informatics Association, FierceEMR reports.
For the study, researchers selected structured and unstructured data fields — including demographics, laboratory data and clinical notes — from the electronic health records of 202,795 patients who had visited Cincinnati Children’s Hospital Medical Center’s emergency department (Hall, FierceEMR, 7/17).
The researchers then used automated screening to select eligible candidates for 13 disease-specific clinical trials between January 2010 and August 2012.
Their objective was to test the effectiveness of three techniques for using real-world data to match patients with a clinical trial.
- Information extraction;
- Machine learning; and
- Natural language processing (JAMIA study, 7/16).
Selections made using the automated system were then compared with decisions made by two board-certified pediatric emergency medicine physicians.
The researchers found that the automated system reduced workload by 92%, while increasing efficiency by 450%.
Of the three methods tested, the researchers noted that the clinical notes in the unstructured data fields were particularly important in identifying potential trial participants (FierceEMR, 7/17).
The researchers predicted that as new initiatives expand the number of clinical trials accessible to patients, the automated eligibility screening approach “will have potential for significant impact in reduction of time and effort for executing clinical research” (JAMIA study, 7/16).