AI-powered ECG screening a cost-effective way to ID heart failure patients
![A majority of medical devices involved in Class I recalls were never required by the U.S. Food and Drug Administration (FDA) to undergo premarket or postmarket clinical testing, according to new research published in Annals of Internal Medicine.[1]](/sites/default/files/styles/top_stories/public/2024-09/istock-1209664264.jpg.webp?itok=gQInU1vO)
Screening electrocardiogram (ECG) results with artificial intelligence (AI) is a cost-effective strategy for identifying patients who may face a heightened risk of heart failure, according to new research published in Mayo Clinic Proceedings: Digital Health.[1] The process was especially affordable when implemented in outpatient settings.
The study’s authors previously confirmed AI-powered ECG screening was a safe, effective and convenient technique for identifying patients with a low left ventricular ejection fraction who may not present with other heart failure symptoms.[2] For this updated analysis, they explored those same patients to examine whether or not the screening was also cost-effective.
The group reviewed data from more than 22,000 patients with a median age of 63 years old. All patients underwent an ECG and presented with no prior symptoms associated with heart failure, and advanced AI algorithms were used to identify any patients who may require follow-up care.
"We categorized patients as either AI-ECG positive, meaning we would recommend further testing for low ejection fraction, or AI-ECG negative with no further tests needed,” senior author Xiaoxi Yao, PhD, MPH, a professor of health services research at Mayo Clinic, explained in a prepared statement. “Then we followed the normal path of care and looked at what that would cost. Did they have an echocardiogram? Did they stay healthy or develop heart failure later and need hospitalization? We considered different scenarios, costs and patient outcomes.”
Overall, this practice was linked to an additional $386 in healthcare costs per patient and a cost-effectiveness ratio of $27,858 per quality-adjusted life year. When only looking at outpatient settings, meanwhile, that ratio improved to just $1,651 per quality-adjusted life year.
According to Yao, the next step of their research is to identify ways to streamline the entire process. Also, the team did note that additional costs may be involved with implementing a process for the use of AI-powered ECG to screen for heart failure.
“Dedicated resources will be needed for monitoring the algorithm’s performance, identifying reasons for performance shifts if there are any, and potentially fine-tuning the algorithm,” the authors wrote. “Different kinds of algorithms, depending on how they were trained and what data the algorithms are run on, have different propensities for performance shifts. However, regardless of the propensity, dedicated resources to support such monitoring and retraining, potentially as part of the licensing cost, will need to be considered during the initial implementation phase and re-evaluated regularly during long-term implementation.”
Click here to read the group’s full analysis.
Michael has more than 18 years of experience as a professional writer and editor. He has written at length about cardiology, radiology, artificial intelligence and other key healthcare topics.