Nationwide trends show imaging utilization has flattened, and yet scrutiny over the appropriate use of imaging is just gearing up. With pressure on physicians growing, two chief mechanisms for driving appropriate use are coming to the fore—clinical decision support (CDS) and radiology benefits managers (RBMs).
Clinicians have increasingly come to embrace some of these changes, but emphasize a subtle, but critical, objective. “Our goal is not to reduce utilization; it’s to reduce inappropriate utilization. No doubt there is overutilization, but there is some underutilization as well,” argues R. Parker Ward, MD, an associate professor of medicine who specializes in cardiac imaging at the University of Chicago Medical Center. The question now facing the medical community is, ‘How should providers and payors cooperatively implement evidence-based imaging guidelines?’
Defining and applying appropriateness
“What we’ve done is define the criteria for appropriate imaging, that was the first step,” Ward explains, referring to the American College of Radiology’s (ACR) Appropriateness Criteria (available at www.acr.org/ac). Ward adds that the next step is applying them, and he contends that clinical decision support is the ideal vehicle for doing so: “quickly, accurately and effectively.”
The ACR’s criteria are becoming the gold standard for imaging orders, comprising 175 topics across nearly two-dozen subspecialty imaging areas.
Increasingly, the ACR Appropriateness Criteria serve as the starting point for homegrown and commercial CDS systems. Minnesota practices and payors hungrily adopted a CDS tool (Nuance) after the non-profit Institute for Clinical Systems Improvement (ICSI) reported an $84 million savings following the software’s subdual of imaging growth.
One reason the ICSI approach may be catching on is because it takes an educational, as opposed to hard-stop, approach to orders. Physicians click through a series of indication-oriented questions aimed at assessing the value of the study ordered by the clinician. If the software grades a physician’s order as lacking in evidence (an inappropriate order), the CDS offers evidence-based alternatives; however, the CDS does not prevent the physician from following through with his or her orders. “It’s a stroke of genius to do it this way,” comments Chip Truwit, MD, chief of radiology at Hennepin County Medical Center (HCMC) in Minneapolis, which adopted the CDS tool in early 2011.
In fact, lacking data from any meta-analysis, the education-based approach to CDS appears to cut growth in equal measure to CDS systems that actually prevent inappropriate orders, with many studies showing reductions on the order of 10 to 15 percent. It makes physicians stop and think, and provides them evidence-based guidance. “It still puts physicians in the decision-making seat, which is where we want them. It just provides them with the right information at just the right time,” affirms Kevin Larsen, MD, CMIO and a primary care physician at HCMC.
Another model is the hard-stop approach, which requires radiologists to authorize any exams deemed inappropriate by CDS. Showing initial reductions of 20 to 25 percent in utilization for select high-tech imaging procedures, including lumbar and brain MRI and sinus CT, Virginia Mason Medical Center in Seattle has expanded its hard-stop program to cover a large cross-section of CT and MRI exams.
CDS vs. RBMs
Some say universal adoption of CDS is premature. “RBMs have been proven to work because they’ve been functioning for years and are now used for something like 90 million patients in the U.S. The experience with CDS, at least what has been published thus far, is much smaller,” contends David C. Levin, MD, professor emeritus of radiology at Jefferson Medical College in Philadelphia.
Levin is cautious about attributing the recent cost curve bends to CDS, wondering whether the technology is not in fact piggybacking onto broader RBM-induced utilization cuts. Pat Corneya, MD, a family practice physician with Health Partners and medical director for Health Partners Health Plan (both headquartered in Minneapolis), offers the rejoinder that, when the Health Partners medical groups implemented either RBMs or CDS, the reductions in inappropriate utilization between them were indistinguishable. But, Corneya notes, “CDS achieved greater cost-effectiveness and was a lot more palatable to physicians, as well as patients.”
Levin agrees that whether practices