American College of Cardiology studies will explore whether AI can improve clinicians' guideline adherence

Editor's Note: This story has been changed to clarify that Practice Fusion is a brand within the Veradigm business.

The American College of Cardiology (ACC) is planning a trio of studies that will measure whether personalized clinical guideline support delivered by an AI tool at the point of care can improve heart patients’ outcomes.

The TRANSFORM studies will each use HealthReveal’s software to generate and deliver care recommendations based on current best practices. These insights are surfaced to clinicians through the electronic health record (EHR) system and include medical literature informing the software’s recommendations.

The studies are part of ACC's TRANSFORM initiative, a new kind of quality research program launched in 2019. These programs leverage EHR data, office-based interventions and partnerships to include the pharmaceutical and medical device industry, health plans, employers, clinicians and patients.

As part of the TRANSFORM program, ACC partnered with health information technology company Veradigm, owned by Allscripts, to develop clinical quality improvement and research networks to power the research. Veradigm owns and operates the Pinnacle Registry, which focuses on coronary artery disease, hypertension, heart failure, atrial fibrillation and diabetes in the outpatient setting, and the Diabetes Collaborative Registry.

All of theTRANSFORM programs use these outpatient registries to help identify providers and patients who are eligible for participation in the program.

RELATED: Early AI adopters beginning to see success predicting readmissions, ED visits

As part of its work with HealthReveal, the ACC's first study will be the TRANSFORM HFrEF study. It will be led by Massachusetts General Hospital cardiology researcher Jim Januzzi, M.D., and focus on the treatment of patients with heart failure with reduced ejection fraction (HFrEF).

Effective strategies have already been established for treating the condition, said Januzzi, who is also a professor of medicine at Harvard Medical School and a trustee of the ACC.

However, a 2018 study reported that less than a quarter of HFrEF patients received the drug treatments as recommended by current clinical guidelines and that less than 1% received those prescriptions at the appropriate dosage. This nonadherence can result in worsened patient outcomes and higher costs for additional care.

“There’s a lot that clinicians have to consider during the standard 20-minute visit with their patients,” Januzzi told Fierce Healthcare. “Many other medical considerations need to be reviewed, and then, of course, there’s the patient in the equation. Changes in therapies may sometimes be met with some hesitation, so understanding how we can best educate our clinicians and patients to really focus on getting them to the optimal medical care … is really the goal of the TRANSFORM HFrEF study.”

Januzzi and Lonny Reisman, M.D., founder and CEO of HealthReveal, said the strengths of the clinical AI software are its scalability and its ability to update recommendations over the course of care. The tool is also clear about how it is reaching its conclusions, Reisman said, which is important in gaining clinicians’ trust and ensuring that care remains adherent to the guideline recommendations.

RELATED: Industry Voices—Next-generation artificial intelligence needs transparency of process to build trust and acceptance

“I want to emphasize that this is not a ‘black box’ sort of an AI thing, where HealthReveal conjured up a way to treat heart failure patients,” Reisman told Fierce Healthcare. “We have created a bridge between the medical literature and the clinical trials.

“Our ability is basically to say: Here’s your patient’s data, this is why they’re eligible for this intervention, here’s the literature that you don’t have to read in its entirety … and the very concrete, actionable step we’re asking you to take,” he said.

The six health system study is set to recruit the first of its target 3,072 participants “within the next couple of weeks,” Januzzi said.

The patients will be randomized to receive either the treatment as usual or the intervention. The latter will consist of clinician support via HealthReveal’s software and patient-facing educational materials designed to help participants “understand what information clinicians are interested in hearing from them,” Januzzi said.

The primary endpoint will be achievement of superior heart failure care over a six-month period, Januzzi said, with other secondary outcomes consisting of patient-reported health status, hospitalizations and death.

TRANSFORM HFrEF is being funded by the ACC and a grant from Novartis Pharmaceuticals. It is set to conclude partway through 2022 and will begin reporting results within a few months of the final patient visit, Januzzi said.

RELATED: Healthcare is ramping up AI investments during COVID. But the industry is still on the fence about Google, Amazon. Here's why

The ACC will be following this with two smaller trials that will similarly employ HealthReveal’s clinical AI.

TRANSFORM CVRiD will be looking at cardiovascular outcomes among patients with Type 2 diabetes. The ACC said that it will begin enrolling patients into the 12-month study period this fall.

The final study will be focused on patients with atrial fibrillation, which Januzzi noted is insufficiently or incorrectly treated with blood thinners. The ACC is planning to formally introduce the investigation next spring.

All three studies will make a point to recruit participants that reflect the U.S. patient demographic, Reisman and Januzzi said.

“We are particularly interested in finding means by which to deliver high-quality care to populations of patients who might not have, to date, received the highest quality of care, including underserved populations,” Januzzi added. “That’s going to be a major focus in all the studies, [and] the atrial fibrillation study in particular.”

Some providers have tapped data-driven clinical decision support as an area of opportunity for AI.

RELATED: Beth Israel Deaconess Medical Center joins Ascension to pilot Google's EHR search tool

Mayo Clinic, for instance, launched a pair of AI companies earlier this month that will tap into providers’ wide stream of patient data to enable early detection and accelerated interventions.

On the other hand, Practice Fusion—purchased by Allscripts in 2018 and now a brand within the Veradigm business unit—was dinged $145 million by regulators to resolve criminal and civil charges it engaged in a kickback scheme that used its EHR software to influence physician prescribing of opioid pain medications.

As AI is increasingly used in healthcare, stakeholders are now laying out road maps for how the industry will support or regulate new healthcare AI products.

Over the last several months, the Consumer Technology Association brought together a host of health and technology companies to publish industry standards around terminology and trust for healthcare AI. The American Medical Association adopted its official policy on healthcare AI as far back as 2018, while the U.S. Food and Drug Administration more recently laid out an AI and machine learning action plan outlining its approach to regulating software and encouraging rigor among healthcare AI developers.