- Ahead of a 2020 start for CMS' reimbursement requirement that doctors ordering advanced imaging tests for patients be assisted by clinical decision support (CDS) software, economics researchers at MIT found the technology can help reduce use of the most expensive tests, but does not change the overall amount of tests being ordered.
- Use of CDS software is meant to ensure providers avoid ordering an overly targeted test and comply with American College of Radiology guidelines when selecting which test a patient needs.
- The MIT Sloan School of Management flagged the research Thursday, published in PLOS One last month, touting it as the first large-scale, randomized study of how CDS software impacts imaging decisions.
The U.S. spends nearly twice as much on healthcare than some other high-income countries, partly because of heavy use of diagnostic testing. Imaging tests can range from plain X-rays and ultrasounds to expensive positron emission tomography (PET) and computerized tomography (CT) scans, which are considered advanced services.
A MedPAC study of CMS spending in 2012 found lab tests accounted for 9% of services and 4% of payments, while imaging accounted for 11% of services and 17% of payments. The panel estimated that incidental findings occurred in 31% cases of imaging, provoking costly follow-up studies that only sometimes confirmed the presence of an abnormality.
CMS thinks CDS software can save the agency money and cut down on unnecessary radiation exposure to patients by directing providers to, in some cases, choose less targeted, burdensome or expensive imaging options. The Medicare Physician Fee Schedule for 2019 confirmed the mandate will go into effect at the start of 2020 in line with its Appropriate Use Criteria provision in the Protecting Access to Medicare Act.
FDA issued draft guidance in late 2017 on CDS software, making regulatory distinctions between software helping a provider remind a patient to take his or her medicine, which would be exempted as a medical device, versus imaging decision-related software, which FDA said presents higher risk for patient harm.
The MIT researchers tracked 3,511 healthcare providers in Wisconsin's Aurora Health Care system, wherein half were assigned to use the CDS tool and half made imaging decisions without it. At one year, use of highly targeted scans were down 6% in the CDS group, the majority of which accounted for decreased CT scans, the most common high-cost imaging type, the authors noted.
The researchers also reported the effect of the CDS software continued over time, a positive indicator that healthcare providers did not experience significant "alert fatigue," which can desensitize a user to the software's messages.
"Going in, we didn't know whether doctors in the treatment group would be receptive to the technology," said MIT Sloan researcher Joseph Doyle, noting that use of the software had the potential to be easily scaled. "Further understanding of the most effective ways to employ the technology beyond simply showing the information about the guidelines remains an important area for future research."