Reviewed by Emily Henderson, B.Sc.Nov 2 2022
Artificial intelligence (AI) can enhance patient diagnosis and treatment, but first, AI-assisted clinical tools must be readily available and used. A new study from Mayo Clinic shows that clinicians who were highly favorable for using an AI-powered clinical decision support tool were two times as likely to detect low left ventricular ejection fraction than those who were less favorable of the tool.
Image Credit: Skorzewiak/Shutterstock
The study has been published in Mayo Clinic Proceedings and has found broad variation in the acceptance rate of AI recommendations. Medical personnel who were high adopters were inclined to be less skilled in handling patients with complicated health issues; however, age, years of experience, gender, and the number of patients cared for were not major factors.
It was surprising to see the significant difference in the rate of diagnosis between high adopters and low adopters. The tool is extremely helpful, but we did not expect to see a full doubling of the diagnosis rate of low ejection fraction as compared to low adopters.
David Rushlow, M.D., Physician and Chair of Family Medicine, Mayo Clinic (Midwest)
The ejection fraction calculates the percentage of blood that leaves the heart every time it contracts. Low ejection fraction could occur due to heart muscle weakness, such as cardiomyopathy, as well as heart valve issues, damage caused by a heart attack, or uncontrolled high blood pressure.
Early diagnosis and care of patients with low ejection fraction is important to reducing the danger of symptomatic heart failure, hospitalization, and death.
AI decision support tools have the potential to be very effective in aiding the diagnosis of serious health conditions before the onset of usual clinical symptoms, and may outperform traditional diagnostic approaches.
David Rushlow, M.D., Physician and Chair of Family Medicine, Mayo Clinic (Midwest)
Clinicians at 48 Mayo Clinic primary care practices in Wisconsin and Minnesota took part in the randomized controlled trial, which comprised 358 nurse practitioners, physicians, and physician assistants, of which 165 clinicians were randomized to the AI arm and were included in the present adoption study.
The AI algorithm was tested on 22,641 patients who had an electrocardiogram (ECG) done between August 5th, 2019, and March 31st, 2020. The clinicians who were randomized to the intervention group could access the screening report, which showed the AI-ECG screening as negative or positive; the clinicians who were randomized to standard care were not given access.
When the report showed a negative outcome, no additional testing was proposed, but when it showed positive, the suggestion was “to consider ordering an echocardiogram.” An email alert was also received by the clinicians when the AI-ECG screening was positive, signifying patients had a high probability of formerly undetected low ejection fraction.
Clinicians who were most likely to follow through with the recommendations of the AI decision aid tended to be less experienced in dealing with complex patients.
David Rushlow, M.D., Physician and Chair of Family Medicine, Mayo Clinic (Midwest)
“This demonstrates the importance of AI systems that integrate seamlessly into the workflows of clinicians. Given the technical nature of AI in health care, it often is initiated and developed in academic specialty practices. To maximize AI's benefits, more collaboration is needed between specialty practices and primary care,” Dr Rushlow added.
Mayo Clinic has a patent for the AI technology and may obtain financial benefits from it, but it will not profit financially from its use in treating patients at Mayo Clinic. The study’s co-authors, Itzhak Attia, Ph.D., Paul Friedman, M.D., and Francisco Lopez-Jimenez, M.D., also may be offered financial benefits from this agreement. The other study co-authors report no competing interests.
The study received partial support from the Mayo Clinic Robert D. and Patricia E. Kern Center for the Science of Health Care Delivery.
Journal Reference:
Rushlow, D. R., et al. (2022) Clinician Adoption of an Artificial Intelligence Algorithm to Detect Left Ventricular Systolic Dysfunction in Primary Care. Mayo Clinic Proceedings. doi.org/10.1016/j.mayocp.2022.04.008.