As concerns mount about the overuse of powerful and costly diagnostic imaging tests, such as CT scans and MRIs, a new study from MIT suggests that software designed to help doctors make better decisions could decrease certain scans by about 6%. The results of the study are published in the journal PLOS ONE.
“There is a lot of debate about the health risks and high costs that stem from the overuse of potentially inappropriate tests,” says Joseph Doyle, the Erwin H. Schell Professor of Management and Applied Economics at the MIT Sloan School of Management, and one of the authors of the study. “Our research shows that technology can improve healthcare delivery by helping physicians make the right decisions about which diagnostic scans to use when.”
This is the first large-scale study where physicians and other healthcare providers were randomized to receive Clinical Decision Support (CDS) software to guide imaging decisions. The CDS provides information about whether a test they order for a given patient is appropriate based on guidelines from the American College of Radiology. Beginning next year, the Centers for Medicare and Medicaid Services (CMS) have put in place new regulations that require imaging orders to be accompanied by a CDS recommendation in order to be reimbursed by Medicare.
Doyle and his colleagues—Sarah Abraham, Laura Feeney, and Amy Finkelstein of MIT, and Sarah Reimer of Aurora Health Care—conducted a yearlong trial of CDS software at Aurora Health Care, the largest healthcare system in Wisconsin. The study involved 3,511 healthcare providers, half of whom were randomly assigned to receive the tool. The control group continued to order images as they had prior to when the trial began.
“Going in, we didn’t know whether doctors in the treatment group would be receptive to the technology,” says Doyle. “If the system recommended a different test from the one that was ordered, would the physician consider changing course? We also worried about the potential of alert fatigue, which happens when people are exposed to a large number of frequent alarms and consequently become desensitized to them.”
The researchers found that CDS helped reduce targeted scans by about 6% relative to the control group. CT scans—the most common high-cost imaging type, which also carries the greatest concerns about over-ordering—were responsible for four-fifths of the overall reduction in targeted scans. While the software changed the nature of image orders, it did not change the number of images ordered overall. The effects persisted over time, suggesting that this type of alert can continue to affect ordering even with concerns about alert fatigue more generally.
The study was supported by Arnold Ventures. The philanthropy funds randomized controlled trials in healthcare, justice, and education, to understand problems and identify policy solutions, and it was conducted in coordination with J-PAL North America[LF1] , which also supports such randomized evaluations. With the impending wave of digital-health tools designed to guide physician decision-making, randomizing the rollout of these tools provides a golden opportunity to test how doctors respond to the information in a rigorous way.
“Our study was meant to understand whether software alone has potential to help doctors improve their decision-making around ordering these expensive and often risky tests because such an intervention is easily scaled,” says Prof. Doyle. “This is especially the case for diagnostic testing given the imminent mandate that CDS be used for high-cost imaging to be eligible for Medicare reimbursement. 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.”
