Why Radiologists Should be Using Decision Support Tools

GRAPEVINE, Texas - Imagine a scenario where no little clinical decision support was utilized: A 53-year-old female presents to the emergency department with a headache and VI nerve palsy.

The ordering physician requests a head CT to rule out a mass. The radiologist receives the order and has plenty of questions. He calls the referring doc, and a lengthy discussion results in the decision to instead order an MRI.

During acquisition, it becomes clear the patient has issues with her renal function, so there are concerns about the contrast agent. Another lengthy discussion, this time with the technologies, ensues. Later, once the images are interpreted, there’s confusion about what to do with the results. Are they critical? What’s the hospital’s policy for communication the findings?

And then even more obstacles present when the billing department finds issue with the report because a supervising physician didn’t sign off on a trainee’s report.

“We did a lot of work here basically for nothing,” said Luciano Prevedello, MD, MPH, of Brigham and Women’s Hospital and Harvard Medical School. He presented this scenario and all its pitfalls Thursday during a presentation at the SIIM annual meeting.

“CDS is the way to go,” he declared, because pressure is only increasing to find efficiencies in health care. A  lengthy, convoluted process like the one he first described isn’t ideal. Instead clinical decision support can provide the right information to the right person at the right time, he said.

Take for example that same case, but with CDS. When the ordering physician tried to request a head CT, the system would have let him know MRI is the preferred test, based on the patient data he entered. A series of rules running in the background of the electronic system based on ACR Appropriateness Criteria would guide the decision.

The proper protocol could then be selected from given options, and the radiologist can ensure he has the right clinical data for the patient. He can also do external searches to aid in interpretation. When it’s time to report the findings, the radiologist can select the level of notification, following the hospital’s policies.

“CDS can be in a lot of places in the process,” noted Katherine Andriole, PhD, FSIIM, also of Brigham and Women’s Hospital and Harvard Medical School, who chaired the session. Decision support can assist in ordering, optimizing protocols, interpretation, reporting, communication, and follow up, she said.

Decision support’s potential doesn’t stop there. That massive amount of data can be mined and utilized to improve efficiencies, according to Woojin Kim, MD, of the University of Pennsylvania School of Medicine.

For example, if an ordering physician regularly doesn’t follow the recommendations and bypasses the decision support, mining that data can help improve compliance. A handy bar chart showing how often this one doc ignores the guidelines, and how much he is an outlier in his organization, can effectively encourage compliance.

“That wouldn’t be possible unless you mine the [computerized physician order entry] data,” said Kim.

Similarly, providing a physician with a patient’s prior studies might change his ordering behavior. Data on imaging recommendations follow-up and radiation dose can also be researched and mined to benefit the organization, he said.