Clinical decision support tools can help standardize best practices for radiologists and patients.
The evolving landscape of the health care system in the United States, including the focus on outcomes-based patient care, provides an ideal opportunity for the inclusion of imaging clinical decision support tools based on evidence-based guidelines, according to an article published in a recent special issue of The Journal of the American College of Radiology.
The article by Joshua Broder, MD, FACEP, associate professor in the division of emergency medicine at Duke University Medical Center, and Safwan S. Halabi, MD, of the department of radiology at Wayne State University School of Medicine, Detroit, focused on the incorporation of these decision tools to help make imaging more transparent and effective.
“Clinical decision support tools augment, but do not replace, physician judgment; they help to standardize care toward best practices, so that patients receive the best care,” Broder told Diagnostic Imaging. “In the best circumstances, when based on outstanding research methods, clinical decision support tools reduce both over and under-utilization of diagnostic imaging and overturn physician misconceptions.”
In order to be effective, clinical decision rules must be both highly sensitive and specificity sufficient to reduce imaging utilization. Ideally, these rules would reduce costs, radiation exposure and patient evaluation times without compromising outcomes or quality of life.
As an example, Broder discussed the Canadian CT Head Rule (see sidebar), which overturned the long-held belief that loss of consciousness alone is a sufficient risk factor to require a head CT. The rule was prospectively derived, validated and studied in 10,000 patients. However, according to the article, it has not been widely adopted in the United States because people believe it to be complex and difficult to remember.
Canadian CT Head Rule
This clinical decision rule states that head CT is only required for minor head injuries in patients with any one of the following findings:
High Risk (for neurological intervention)
• GCS score <15 at two hours post injury
• Suspected open or depressed skull fracture
• Any sign of basal skull fracture
• Vomiting of two or more episodes
• Age 65 years or older
Medium Risk (for brain injury on CT)
• Retrograde amnesia to the event for 30 minutes or more
• Dangerous mechanism (pedestrian struck by motor vehicle, occupant ejected from motor vehicle, fall from elevation)
Reference
Stiell IG, Wells GA, Vandemheen K, et al. The Canadian CT Head Rule for patients with minor head injury. Lancet. 2001;357:1391-1396.
In general, some clinicians have been hesitant to employ clinical decision rules for imaging.
“Some practitioners may be unfamiliar with clinical decision support tools, or uncomfortable with risk,” Broder said. “Most clinical decision support instruments help us to ‘risk stratify’ patients – to identify patients with a low risk of serious conditions. However, some residual risk often remains.”
According to Broder, practitioners may be hesitant to shoulder that risk. For example, the Pulmonary Embolism Rule Out Criteria, when negative, suggest a risk of pulmonary embolism risk of less than 2 percent-but not zero. Medical professional groups could help by endorsing decision support tools and defining an acceptable risk threshold.
“If we continue to follow a zero risk tolerance, we will find it hard to reduce medical imaging use,” Broder said.
Furthermore, changes to the medical landscape are beginning to make the use of clinical decision tools easier. Specifically, the growth of electronic health records means that these rules could be incorporated into an entry for the medical record. When a patient’s data and history is imported into the record, decision support rules would be embedded, allowing physicians to terminate orders at the moment of entry if the rule indicates that imaging is not required.
“The evolving healthcare landscape supports comparative effectiveness research – trying to achieve the same or better results at lower cost,” Broder said. “In many cases, clinical decision support systems can yield benefits to all parties – shorter visits for patients who don’t require imaging, lower radiation exposures to patients when imaging is avoided, more efficient clinical practices with greater throughput when clinicians apply rules and avoid imaging, and cost savings.”
Study Assesses Lung CT-Based AI Models for Predicting Interstitial Lung Abnormality
September 6th 2024A machine-learning-based model demonstrated an 87 percent area under the curve and a 90 percent specificity rate for predicting interstitial lung abnormality on CT scans, according to new research.
What a Prospective CT Study Reveals About Adjunctive AI for Triage of Intracranial Hemorrhages
September 4th 2024Adjunctive AI showed no difference in accuracy than unassisted radiologists for intracranial hemorrhage (ICH) detection and had a slightly longer mean report turnaround time for ICH-positive cases, according to newly published prospective research.
The Reading Room: Racial and Ethnic Minorities, Cancer Screenings, and COVID-19
November 3rd 2020In this podcast episode, Dr. Shalom Kalnicki, from Montefiore and Albert Einstein College of Medicine, discusses the disparities minority patients face with cancer screenings and what can be done to increase access during the pandemic.
New Study Identifies Key Computed Tomography Findings for Post-Op Recurrence of Pancreatic Cancer
August 22nd 2024For patients who had post-op recurrence after a Whipple procedure for pancreatic ductal adenocarcinoma (PDAC), between 80 to 86 percent of follow-up CT exams revealed new or increased soft tissue.