Deploying radiologists as ‘human clinical decision support’ to quash inappropriate imaging
One East Coast provider has found success deploying radiologists as “human clinical decision support” to help cut inappropriate imaging. And the switch is helping to limit costs and patient radiation exposure, too.
Yale New Haven Hospital recently tested out this workflow change, adding a button in its electronic health record that allows radiologists to suggest changing ordering providers’ incorrect CT or MRI requests. The intervention has shown early promise when compared to typical electronic CDS tools, researchers wrote in the analysis, published Jan. 17 in the Journal of the American College of Radiology.
Previous investigations have found that clinicians rejected as many as 99% of such recommendations from computer-based tools, and still performed 63% of inappropriate tests, wrote Sarvenaz Pourjabbar, MD, a Yale radiology resident, and colleagues.
“In contrast, the radiologist-driven process yielded an 82% acceptance rate of suggested order changes, confirming our hypothesis that the program would be well-accepted. These data suggest that direct radiologist involvement in modifying orders has greater impact than CDS tools,” the team noted. “Radiologist review of imaging orders for appropriateness also addresses the need for patient-centric imaging initiatives,” they added.
Pourjabbar and colleagues tested the tool out on 79,000 outpatient CT and MRI scans at one tertiary care center, requested between April 2017 and January 2018. Radiologists made change-order requests for about 3% of the CT scans requested during the study period, and roughly 5% of the time out of MRI tests, the analysis found. All told, there were 2,865 change requests, which ordering providers accepted 82% of the time in both modalities.
Digging deeper into the data, the team found the most commonly switched studies were those conducted both with and without contrast agents—45% of the time in CT and 26% in MRI. Meanwhile, radiologists rarely made alterations to the anatomy imaged (8% for CT and 4% for MRI).
Along with benefiting the provider by reducing inappropriate imaging, the change also had spillover benefits to patients, the team found. About half of the changes in CT resulted in an imaging test that delivered less radiation, and 76% led to a less costly test. In MRI, Yale New Haven tallied a 26% decrease in gadolinium-based contrast agent use, and 48% less costly imaging. About 52% of switched MRIs led to costlier examinations, however, the uptick could “rationalized by medical necessity and appropriateness,” Pourjabbar noted.
All this was achieved at an added cost of about $50 per day for staff to run the program, and they believe the numbers would look even better when accounting for change orders that weren’t easily categorized.
“A large gray zone exists between clearly inappropriate examinations and clearly appropriate examinations. Bundling all ‘questionable’ studies into an inappropriate category will drive up percentage of examinations deemed inappropriate,” the team noted.
Marty Stempniak has covered healthcare since 2012, with his byline appearing in the American Hospital Association's member magazine, Modern Healthcare and McKnight's. Prior to that, he wrote about village government and local business for his hometown newspaper in Oak Park, Illinois. He won a Peter Lisagor and Gold EXCEL awards in 2017 for his coverage of the opioid epidemic.