MU’s push for CDS: How will it impact Radiology?

Twitter icon
Facebook icon
LinkedIn icon
e-mail icon
Google icon
 - Safwan Halabi, MD
Safwan Halabi, MD

As the federal meaningful use incentive program has progressed, its direct impact on the practice of radiology has intensified through the addition of provisions specific to the utilization and sharing of medical images. Meaningful use also stands to impact the field indirectly through changes in incentivizes across the healthcare continuum. Most significant among these may be the use of clinical decision support to encourage appropriate, outcomes-based ordering of imaging studies.

“Radiology is going to be affected by what meaningful use does to everyone else as well,” says Safwan Halabi, MD, radiologist at Henry Ford Hospital, Detroit, Mich. Along with Joshua S. Broder, MD, Halabi co-authored a March 2014 article in Journal of the American College of Radiology titled “Improving the Application of Imaging Clinical Decision Support Tools: Making the Complex Simple.” As he notes, “Radiology touches almost every single patient at some point or other during their lives. We have a lot of information that’s not really being used—yet.”

Weaknesses in CDS tools

Halabi and Broder point out that the promotion and incentivization of EHRs and CPOE on the part of CMS have created “a unique opportunity to catalyze the use of evidence-based guidelines with the inclusion of clinical decision support tools.” Halabi says, however, that these tools will require improvement if they are to become part of clinicians’ daily workflows. “The onus will be on future-looking physicians to help develop tools that will give ordering clinicians guidance while not hindering them in seeing as many patients as they can.”

Halabi also points out that while current CDS tools include guidance for thousands of indications, even that level of granularity comes with significant gaps. “If you look at the guidance, it doesn’t really cover the entire spectrum, and it’s not one-size-fits-all,” he says. “It’s rare to see a patient with just one problem. Further, a lot of the existing guidance doesn’t include prior imaging or prior treatments—it’s just looking at the here and now. Patients are very nuanced, and it’s hard to build an algorithm that covers every scenario.”

Finally, he offers that at this point, delivering an evidence-based justification for every one of the guidelines just isn’t possible. “A lot of different groups got together to decide on what they believe are the most appropriate tests for certain diseases based on what we know now,” Halabi says. “But we must do a better job at basing the guidelines on outcomes and clinical evidence. We have yet to see whether a lot of this will help patients in the long run.”

Halabi envisions a future in which CDS is robust enough to accommodate the nuances of a patient’s comorbidities and medical history to offer both short- and long-term recommendations regarding that patient’s care. “Thedata need to be utilized in CDS so that you don’t have to go through a whole process to get the guidance you need—it should be teed up based on the patient’s age, gender and problem lists,” he says. “It should be like on Amazon.com, when it tells you ‘people who ordered this also ordered this’—in this case, it would say ‘patients like this one saw their health improve from these steps.’”

MU and adoption

By incentivizing the use of CDS as part of CPOE, the meaningful use program has the potential to drive the widespread adoption of the technology necessary to create this robust feedback loop, Halabi says. “In this version of MU, CMS is laying the groundwork for healthcare organizations,” he observes. “With this whole push toward using health IT, more and more practices and health systems are collecting a lot of data on patients. Now what we need is feedback on outcomes to improve future practice.”

Halabi describes the continuous integration and utilization of that feedback as “the holy grail,” but acknowledges it raises philosophical questions about the role of the physician in general, as well as the role of the radiologist. “I have this discussion with my residents all the time. They feel like they’re artisans honing a skill—they don’t want to be pushed into just following checklists. And that’s the real struggle: How do you maintain physician and patient autonomy while still putting people into best practice sequences? The person who has examined thousands of patients could know better than a piece of software.”

With so much guidance automated, where will the radiologist fit in the care continuum? “Medical imaging will