Novel Intervention to Increase GDMT in HF Exploits Noncardiac Admissions

For patients with heart failure with reduced ejection fraction (HFrEF) who are admitted to the hospital for noncardiac reasons, a system whereby their treating clinicians can connect directly with heart failure specialists not only increases the likelihood of guideline-directed medical therapy (GDMT) initiation but also improves dose optimization at discharge.

Communication between heart failure specialists and doctors in other, noncardiology disciplines is relatively rare. “Usual care usually consists of a primary medical team reaching out to general cardiology consult through a formal consultation if there is a clinical question that requires an answer or assistance,” lead author Vishal N. Rao, MD, MPH (Duke University School of Medicine and Duke Clinical Research Institute, Durham, NC), told TCTMD. But if a cardiology consult is not deemed necessary for a heart failure patient hospitalized for causes like pneumonia, complications from a fall, COVID-19, or kidney injury, he continued, it’s unlikely that the medical team would ever speak with a cardiologist about the patient’s care plan.

Study after study has shown that initiation and uptitration of lifesaving heart failure medication is abysmal, so out-of-the box strategies for improving GDMT are urgently needed.

According to Rao, his multidisciplinary team of cardiologists, internists, pharmacists, and nurses was motivated to design a peer-to-peer communication system within their hospital’s electronic health record (EHR) in order to “best identify patients with heart failure and move the needle with regard to their clinical prognosis at every single opportunity possible with regards to medical care and frequent touch points. . . . Seeing that there's rising rates of patients with heart failure and the high rates of hospitalizations overall, we aimed to develop a study to improve the quality of care across our services beyond just cardiology alone.”

The clinician-level “peer-to-peer” system developed for and used in this study was automatically embedded in their institution’s EHR. It consisted of a detailed note written by someone on the multidisciplinary heart failure team which provided GDMT recommendations and information on medication costs. The rounding physician also had the option to call and speak with a heart failure physician, pharmacist, or nurse in real time for further consultation.

Anand Shah, MD, MBA (Duke University School of Medicine), is scheduled to present the study at a poster session during the American Heart Association 2022 Scientific Sessions later this week; the paper was published this week in Circulation: Heart Failure.

Sean Pinney, MD (University of Chicago, IL), who reviewed the study for TCTMD, said: “My immediate reaction was: Why didn't I think of this?

“It's a great implementation-science piece that is critically important for us to try to come up with new ways to increase the initiation and escalation of guideline-directed medical therapy,” he continued. “We have these therapies that have been shown to be life-extending, life-improving, and the real challenge to us now is coming up with ways to get these medicines into patients. The fact that they targeted admissions on a medical floor and provided peer-to-peer guidance was really fantastic.”

Increase in GDMT

For the study, Rao and colleagues identified patients already diagnosed with HFrEF who had been admitted to their hospital for noncardiac reasons between May and September 2021, although those with end-stage renal disease, hemodynamic instability, concurrent COVID-19 infection, and current enrollment in hospice care were excluded.

Rao described the communication system as “essentially a virtual consultation note and communication between various providers across specialties” that provided information about the patient, including imaging, laboratory findings, and past medical history, to help clinicians gauge potential eligibility for GDMT. In addition, there would be an option to engage in a telephone consultation in order to discuss patient suitability for heart failure medications, estimated prescription drug costs, completion of prior authorizations or prescription coverage options, and postdischarge care transition and follow-up.

We have these therapies that have been shown to be life-extending, life-improving, and the real challenge to us now is coming up with ways to get these medicines into patients. Sean Pinney

Of the 91 patients included, 52 were randomized to medical teams using the intervention (mean age 63; 23% female) and 39 to medical teams providing usual care (mean age 66; 26% female). Mean ejection fraction was similar in the two groups (33.1% vs 31.2%; P = 0.28)

GDMT use on admission was similar for both groups, specifically for beta-blockers (60% vs 67%; P = 0.49), ACE inhibitors/ARBs/ARNIs (62% vs 53%; P = 0.40), mineralocorticoid antagonists (MRAs; 33% vs 26%; P = 0.51), and sodium-glucose cotransporter 2 (SGLT2) inhibitors (20% vs 16%; P = 0.64).

Following randomization, the intervention cohort reported more GDMT initiation or continuation during the study than those who received usual care, most notably for ACE inhibitors/ARBs/ARNIs (71% vs 49%; P = 0.04), MRAs (40% vs 21%; P = 0.05), and SGLT2 inhibitors (26% vs 14%; P = 0.19).

Expressed as an optimal medical therapy (OMT) score at discharge, scores increased for patients randomized to the automatic peer-to-peer consults and decreased for controls, even after adjusting for OMT score on admission (+0.44 vs -0.31; P = 0.041).

At 30 days, however, there was no difference between the intervention and usual-care groups in the rate of follow-up with a primary care physician (64% vs 63%; P = 0.95) or heart failure specialist/general cardiologist (31% vs 33%; P = 0.85). No patients from the intervention cohort died, whereas the mortality rate in the usual-care group was 7.7%.

Rao specified that this was a feasibility study to explore the possibilities for implementing this kind of virtual-peer-to-peer consultation during noncardiac patient visits, something they’d initially anticipated would be challenging. “We were fairly surprised by the overwhelming positive reception,” he said. “We opened it up to the overall department of medicine, including hospitalists, providers, and medical residents and trainees, and said, ‘If you don't want to participate in this, you don't have to.’” In the end, “nobody declined participation, and in fact, many people were very eager to learn about it.”

As for the results, Rao called them “favorable” and “surprising.” The intervention was well received and “definitely opened doors to our future endeavors, as we are hopeful that this may be a type of mechanism that [we] can add on to other quality improvement-based strategies to engage providers and patients in routine heart failure care.”

Rao said his team is planning future research to assess how individual providers feel about this type of intervention and whether it could be replicated and possibly scaled up at other institutions.

Direct Decision Support

Pinney said that in one of his previous roles, his team tried to design an automatic intervention that provided guidance through the EHR to providers, “but that was not terribly effective, in part because it's really hard to write that type of an algorithm.” Where this study succeeds, he continued, “was coming up with [multidisciplinary] decision support directly to providers, which was immediate and actionable.”

While he would have expected the intervention to lead to a greater uptake of SGLT2 inhibitors, “I think that speaks to the challenge that we have right now across the country of trying to get these medicines into people,” Pinney said. “Some of this is overcoming knowledge gaps and some of it is overcoming inertia, but I think that unaffordable drug costs and the need for preapprovals are also pretty significant barriers to be overcome. And I bet you that also hindered the uptake of the SGLT2 inhibitors.”

In thinking about how to scale an intervention like this, Pinney said he would like to see how it performs in other academic medical centers as well as community hospitals and rural settings. He said he is already considering how to enact something similar at his institution.

Some of the challenges he expects might get in the way include the complexity of the patient population, provider inertia, type of EHR and what it allows, and appropriate staffing. “But those are solvable,” Pinney said.

Lastly, he said the study highlighted the additional hurdle of appropriate transitional care. “There were such a low rates of primary care and cardiology follow-up 30 days, and it's not from a lack of trying,” Pinney said. “You kind of fumble the ball at the one-yard line. You've taken all of this time, you've invested in the resources—both the EMR and the human resources, the medical expertise—in designing the intervention, you get it tweaked while they're in the hospital, and then it just all falls apart because further dose titration doesn't take place because they don't keep that follow-up appointment. I think that's where this paper also succeeds, and it points us again in the right direction. . . . We have to come up with another type of intervention to make sure that our transitional care plans are successful and get patients into the doctors’ offices after they leave the hospital.”