Radiology reports and images contain mounds of insightful data, and much of it could contribute significantly to improving patient outcomes if easily accessible through the electronic health record.
But that isn’t the case for many imaging departments. Instead, free-text reports are often radiologists’ main contribution to the EHR. And these write-ups—largely unchanged since the early days of the profession—are not going to cut it in the team-oriented, data-driven healthcare delivery system of the future.
That’s according to a Friday morning presentation during SIIM’s virtual meeting, delivered, in part, by Tarik K. Alkasab, MD, PhD, service chief of informatics/IT & operations at Harvard Medical School. Radiology must close this “data gap,” he said, and include structured clinical exam information within systemwide data streams.
Alkasab pointed to measurement information such as lymph node size, nodule location and liver density, and clinical-categorization data like tumor and stroke volumes, liver stiffness and BI-RADS categories. Such insights would undoubtedly impact decision-making. The key, he added, is using “common data elements” to label values, so physicians across the enterprise can quickly access and understand the information.
And the ACR and RSNA already have such a resource in place with RadElement. The online site creates machine-readable values in radiology reports that are easily transferrable to the EHR, regardless of vendor.
In practice, imaging experts would enter these descriptors into a reporting module to illuminate their findings, which then would generate report text and, in some cases, appropriate recommendations.
This will only get easier as artificial intelligence continues to progress and will thrust radiologists into a more advisory role, one Alkasab describes as “data wranglers.”
“But I think one of the most valuable things for us as radiologists is, as the economic model of healthcare changes, it’s going to give us an opportunity to change, to demonstrate, exactly how we’re providing increased value to the care of our patients,” he said.
All of this would allow physicians across the system to interpret imaging exams and use the information to improve follow-up imaging adherence and avoid unnecessary downstream care.