AI-powered EHR alerts make a world of difference for cardiologists, heart patients
Artificial intelligence (AI)-enabled clinician alerts can significantly improve care for heart patients, according to new data presented at ACC.26 and simultaneously published in JACC.[1] The study’s authors found that patients were more likely to undergo valve transcatheter aortic valve replacement (TAVR) and other critical valve interventions if their care team utilized automated AI alerts.
“The likelihood of receiving valve intervention varies substantially by treating physician, with marked heterogeneity even among cardiologists, and patients managed by clinicians with lower rates of valve intervention experience higher risk-adjusted mortality,” wrote lead author Wayne B. Batchelor, MD, a veteran interventional cardiologist and president of Inova’s medicine service line, and colleagues. “Access to valve intervention is also uneven, with women, racial and ethnic minority populations, older adults, patients with greater social deprivation and rural residents less likely to receive intervention even after adjustment for comorbidities and disease severity. These disparities reflect a complex interplay of patient-, clinician-, and system-level factors.”
With the ALERT trial, Batchelor et al. aimed to learn if AI could help patients receive more consistent care by alerting the care team when a valve intervention may be necessary. They focused on the use of the Tempus Next platform, an advanced AI software from Chicago-based Tempus that scans echocardiography results and flags any suspicious findings. This study was specifically focused on identifying signs of symptomatic severe aortic stenosis and mitral regurgitation, two life-threatening conditions that can be treated with minimally invasive procedures if detected in time.
More than 700 U.S. clinicians ordering more than 2,000 echocardiograms participated in the ALERT trial. Patients were randomized to either receive AI alerts after echocardiograms were ordered or carry on with usual care.
Overall, after 30 days, patients treated with AI alerts were 27% more likely to have their case evaluated by a multidisciplinary care team and they experienced a 40% increase in their odds of receiving TAVR or another valve intervention.
“These findings highlight the value of real-time clinical alerts to accelerate diagnosis and specialist referral, helping ensure that more patients—regardless of race, ethnicity, geography, hemodynamics, or other factors—have access to guideline-directed, life-saving care,” Batchelor said in a statement.
Medtronic and Tempus both funded this clinical trial.
“By helping clinicians identify patients sooner and connect them to care, technologies like AI‑enabled alerts have the potential to ensure more people can benefit from lifesaving therapies like TAVR, regardless of race, geography, or background,” Jorie Soskin, vice president of Medtronic’s structural heart business, said in the same statement. “This work underscores our commitment to advancing structural heart care and the future of TAVR.”
“The ALERT trial shows that data-driven clinical decision support positively impacts patients suffering from valvular heart disease,” Brandon Fornwalt, MD, PhD, senior vice president of cardiology at Tempus, added in a separate statement. “Being a part of this work further underscores our commitment at Tempus to building technology that helps every patient have the chance to receive the best possible care.”
Click here to read the full study in JACC, the American College of Cardiology’s flagship journal.
Michael has more than 19 years of experience as a professional writer and editor. He has written at length about cardiology, radiology, artificial intelligence and other key healthcare topics.