At RSNA 2021, a Radiology Leader Pushes His Colleagues Forward into Value

Brink told the assembled audience that, if radiologists as a community are to address some of the policy, payment, and regulatory issues facing them as a specialty, they must essentially clean up their own specialist house. On the one hand, he noted, “We’re being crushed with an enormous work burden; we’ve seen a 25-to-30-percent increase in radiologic work volume at Massachusetts General Hospital in 2021 alone.” Some of that volume this year has doubtless been the result of backed-up demand unfulfilled during the early months of the pandemic in 2020. On the other hand, he quickly added, “Odds are high that some of that volume growth is related to exams that might be of low volume.”

In that context, he asserted, radiologists need to be involved in helping to clear some of their reading workload by participating in eliminating “low-value” diagnostic imaging volume, through patient education, encouragement of the referring-physician clinical criteria that are aimed at reducing unnecessary imaging volume, and self-education and participation in quality assurance activities inside the specialty itself, he said. Given involvement in all those efforts, he said, “It certainly behooves us to participate as vigorously and fully as we can” in the policy dialogue. “If we don’t participate, we could become pawns for others” in the constantly shifting, never-ending policy debate around diagnostic imaging.

“Let begin by asking how we eliminate low-value imaging. I was very intrigued by a JAMA Viewpoint article, published on April 8, 2021,” he Brink said. “The article’s authors implored us to find ways to measure it, and how to carve it out of the lexicon of any given healthcare organization. Whenever I get the slightest twinge of pain, particularly in my lower back, I want to rush to get imaging. Those knee-jerk sort of reflex actions lead us to low-value care, and we have to avoid those.”

The article, “Reducing Low-Value Care and Improving Health Care Value,” authored by Allison H. Oakes, Ph.D., and Thomas R. Radomski, M.D., starts out thus: “Low-value care, defined as the use of a health service for which the harms or costs outweigh the benefits, is a pervasive and enduring problem in the US. Enacting policies that limit reimbursement for low-value services is an important step in mitigating such care. For example, Powers et al1 proposed a framework to identify and prioritize policies to govern the de-adoption of low-value care focused on evidence, eminence, and economics. However, transitioning to a state of health care delivery that prioritizes value over volume will require balancing “top-down” policy prescriptions with a “bottom-up” approach geared toward affecting local cultural change. Such an approach involves implementing de-adoption strategies tailored to address the behavioral and organizational factors that drive the provision of low-value care within a local health care ecosystem, whether it is an individual practice or a tertiary referral center.”

Clinical decision support as a key tool in the arsenal

If there’s a single tool that can help to eliminate the production of low-value diagnostic imaging scans, Brink says, it’s clinical decision support (CDS), which comes in two forms, when it comes to diagnostic imaging: first, the form of CDS that helps referring physicians to make use of evidence-based criteria for ordering; and second the type that helps guide radiologists.

And he cited the American College of Radiology (ACR) and its set of appropriateness criteria, first created several years ago now. As the description of that set explains, “The ACR Appropriateness Criteria® are evidence-based guidelines to assist referring physicians and other providers in making the most appropriate imaging or treatment decision for a specific clinical condition. By employing these guidelines, providers enhance quality of care and contribute to the most efficacious use of radiology.” He noted that “The guidelines are developed and reviewed annually by expert panels in diagnostic imaging and interventional radiology. Each panel includes leaders in radiology and other specialties. In 2021, there are 216 Diagnostic Imaging and Interventional Radiology topics with over 1030 variants and for our Diagnostic Imaging topics 2,400 clinical scenarios.”

Meanwhile, Brink emphasized that his own organization, Massachusetts General Hospital, has been doing pioneering work in this area. “Back 17 years ago in August 2004,” he noted, he and colleagues including Daniel Rosenthal, M.D., “developed an approach to avoid preauthorization with commercial payers by creating a Clinical Decision Support with Risk Sharing program, led by Dr. Dan Rosenthal. They took this set and fed it into an electronic order entry system—novel for its time—in which a practitioner could feed the data into the system and receive an appropriateness score. This has become an important part of the culture of Massachusetts General.” Among other elements in the program was the creation of a color-based system (green, yellow, red), providing input for every ordering physician at the point of ordering; the use of that system helped provide feedback to ordering physicians, helping them to lower their “red rate”—the percentage of times in which the system told them that they were about to order inappropriately—an important step in helping ordering physicians to order more thoughtfully, thus reducing low-value diagnostic imaging orders.

Brink noted that one version of such a program was developed into a commercial software product, called ACRSelect, and that its effectiveness was documented in an article in the Journal of the American College of Radiology (JACR), published on May 25, 2018. The title of that article is “Impact of a Commercially Available Clinical Decision Support Program on Provider Ordering Habits,” and it was authored by Timothy C. Huber, M.D., Arun Krishnaraj, M.D., James Patrie, M.S., and Cree M. Gaskin, M.D. As the article’s authors noted, referring to low-value diagnostic imaging studies as “low utility studies,” “The commercially available CDS-generated scores for 34 percent and 20.4 percent of pre- and postintervention studies, respectively. After feedback, the relative frequency of low utility studies decreased to 5.4 percent from 11 percent, and the relative frequency of indicated studies increased to 82 percent from 64.5% percent.”

The article’s authors noted that “The shift from volume- to value-based payment models has prompted physicians to examine which aspects of the care they provide are warranted and which aspects are of low value. Radiology, as a referral specialty, can contribute to this effort by providing guidance to referring providers on the most appropriate imaging test, if any, given the clinical presentation. Locally developed or institution specific CDS software that has been implemented at several institutions across the United States has demonstrated that ordering patterns can change, with reductions in the number of low utility examinations ordered after implementing such systems. Our study demonstrates that similar results can be achieved with commercially available CDS and that the impact of these efforts can vary by provider type and imaging modality.”

Brink went on to discuss a number of other topics of importance, including the potential for enhancing patient engagement, expanding virtual care and ambulatory access in order to improve health equity for underserved and marginalized patient populations, and the strong potential of care coordination to improve crucial follow-up among care coordinators on behalf of patients who need to be seen by specialists following the performance of diagnostic imaging studies by radiologists.

Among the innovations that have taken place at Brigham & Women’s Hospital in the past year-and-a-half, for example, has been the development of a “virtual reading room,” giving patients the opportunity to speak virtually with radiologists and go over their study findings; that innovation in particular, Brink said, has meaningfully enhanced patient engagement.

On a practical level, one innovation that has taken place at Mass General has had nothing to do with clinical work at all, per se; and that has been the creation of a transportation program that ensures that patients show up for their imaging appointments, something that is far from guaranteed among patients from marginalized populaltions.

What’s more, Brink and his colleagues have been involved in a follow-up-to-study program involving three teams of professionals: a radiology care coordination team inside the central radiology scheduling team; a safety net team that assists with care coordination for patients seen from outside the hospital’s network or who have no known primary care physician; and a back-end operations team that makes sure every case is resolved.

What’s more, Brink said, it’s clear that diagnostic imaging services cost too much in the U.S. compared to in Europe. One element of a solution, he asserted, is to shift as much imaging care delivery from academic medical centers to community hospitals and ambulatory care settings, where the cost will be lower.

In the end, Brink said, radiologists need to become fully engaged in pushing quality within the specialty of radiology, focusing on accurate diagnoses and precise measurements, integrated with quality processes, and a strong focus on eliminating the delivery of low-value diagnostic imaging tests and reducing costs wherever possible.