In the November issue of the journal Circulation, I wrote a letter to the editor in response to a study published in January, entitled, “Association Between Hospital Volume, Processes of Care, and Outcomes in Patients Admitted With Heart Failure: Insights From Get With The Guidelines-Heart Failure.”

My letter commented on the study’s findings, in which researchers concluded, “In evaluating the quality of inpatient heart failure care, patients and policy makers should consider how well a hospital meets clinical care guidelines.”

In my letter to the Circulation editors, I reflected on the fact that, according to the New England Journal of Medicine, nearly half of chronically ill participants often do not receive guideline-directed medical therapy (GDMT). This is in line with what Circulation editors found in their January study, as well as with other literature throughout the cardiology field. For example, another study, from the Journal of the American College of Cardiology, looked at the use of GDMT for heart failure patients with reduced ejection fraction. It found that the recommended therapy—a prescription for both a renin-angiotensin inhibitor and a heart failure-approved beta-blocker within 90 days before the placement of a preventive implantable cardioverter-defibrillator—was implemented only 61 percent of the time.

In my experience, the reason GDMT is not always followed is threefold: First, physicians have too many patients, too much administrative burden, and not enough time to continually research the latest guidelines in order to implement them at the point of care. Second, medical advancements and discoveries are occurring faster than can be digested and applied in real-time. And finally, while the transition from fee-for-service to value-based care has helped providers look at big-picture strategies for managing patients’ health, perhaps it’s made individual treatment guidelines and recommendations easier to overlook.

Fortunately, industry has reached the point where AI and machine learning can help with many of these challenges. By creating a “digital twin” from patient data including electronic health records, insurance companies, and information from remote monitoring technology (devices that are wearable or implantable, or can be used in the home), HealthReveal can compare personalized patient information against the latest GDMT in real-time, and send “Reveals” directly to physicians and others on the care team, allowing them to intervene before adverse health events occur. We are excited to use AI technology to democratize care and make the best treatments available to everyone—the future of healthcare is now.

Click here to read my letter to the editor in Circulation.