Newswise — Rochester, Minn. — Artificial intelligence can improve diagnosis and treatment for patients, but first AI-enabled clinical tools must be easily accessible and usable.
According to new research from the Mayo Clinic, physicians who are high adopters of an AI-enabled clinical decision support tool are twice as likely to diagnose a low left ventricular ejection fraction than low adopters of the tool. The study, published in Mayo Clinic Proceedings, found wide variation in the adoption rate of AI recommendations. Physicians with higher adoption were less experienced in dealing with patients with complex health problems, but age, gender, years of experience, and number of patients cared for were not significant factors.
“It was surprising to see such a significant difference in diagnosis rates between high adopters and low adopters,” says David Rushlow, MD, a Mayo Clinic physician and chair of family medicine for the Mayo Clinic in the Midwest. “The tool was very helpful, but we didn’t expect to fully double the diagnosis rate of low ejection fraction compared to low adopters.”
Ejection fraction measures the percentage of blood ejected each time the heart contracts. A low ejection fraction can be caused by heart muscle weakness such as cardiomyopathy, as well as heart valve problems, uncontrolled high blood pressure, or damage from a heart attack.
Early diagnosis and treatment in patients with low ejection fraction is critical to reduce the risk of symptomatic heart failure, hospitalization, and mortality. “AI decision support tools can be very effective in aiding the diagnosis of serious health conditions before the onset of common clinical symptoms and can potentially outperform traditional diagnostic methods,” Dr. Rushlow said.
Physicians at 48 Mayo Clinic primary care practices in Minnesota and Wisconsin participated in a randomized controlled trial involving 358 physicians, nurse practitioners, and physician assistants, of whom 165 physicians were randomized to the AI arm and were included in the current adoption analysis. The AI algorithm was run on 22,641 patients who had an electrocardiogram (ECG) between August 5, 2019 and March 31, 2020. Physicians randomized to the intervention group had access to a screening report demonstrating AI. ECG screening positive or negative; Randomized physicians had no access to usual care.
When the report is negative, no further testing is recommended, but when it is positive, “Consider ordering an echocardiogram” is recommended. Physicians also received an email alert when the AI-ECG screening was positive, indicating a high probability of previously undiagnosed low ejection fraction for patients.
“Physicians who mostly follow the recommendations of an AI decision aid have less experience dealing with complex patients,” says Dr. Rushlow. “This demonstrates the importance of AI systems that seamlessly integrate into the clinician’s workflow. Given the technological nature of AI in health care, it is often initiated and developed in academic specialty practices. To maximize the benefits of AI, more collaboration between specialty practices and primary care is needed.”
Mayo Clinic owns the patent for the AI technology and may derive financial benefits from it, but may not benefit financially from its use in patient care at Mayo Clinic. Co-authors Itzhak Attia, Ph.D., Paul Friedman, MD, and Francisco López-Jimenez, MD may also receive financial benefits from this agreement. The remaining co-authors report no competing interests.
The study was funded by the Mayo Clinic’s Robert D. and Patricia E. Supported in part by the Kern Center for the Science of Health Care Delivery.
About Mayo Clinic Proceedings
Mayo Clinic Proceedings is a monthly peer-reviewed journal that publishes original articles and reviews on clinical and laboratory medicine, clinical research, basic science research, and clinical epidemiology. Sponsored by the Mayo Foundation for Medical Education and Research as part of its commitment to medical education, the journal has been published for 95 years and has a circulation of 127,000.
About the Mayo Clinic
Mayo Clinic is a not-for-profit organization committed to innovation in clinical practice, education and research and providing compassion, expertise and answers to everyone in medical need. For additional Mayo Clinic news visit the Mayo Clinic News Network.