ACC 2022. An EHR alert tailored to prompt clinicians to check guideline-directed medical therapy for individual HF patients was integrated carefully into clinical workflow.
Guideline-directed medical therapy (GDMT) for heart failure was prescribed significantly more often by physicians who received a customized digital alert via their electronic health record (EHR) system than by those who did not, according to authors of the PROMPT-HF trial.
The researchers presented their findings at the American College of Cardiology’s (ACC) 71st Annual Scientific Session, held April 2-4, in Washington, DC and virtually.
The study, in which an EHR alert occurred while a clinician was entering medical orders into the system shortly after a patient visit, met its primary endpoint of an increase in the number of classes of GDMT therapy prescribed 30 days after physician participant randomization. PROMPT-HF, according to an ACC statement, is now the first randomized trial to demonstrate significant improvements in use of heart failure GDMT using a clinician prompt that is “low-cost” and “scalable.”
“The results were quite remarkable,” said lead study author Tariq Ahmad, MD, chief of the Yale Heart Failure Program at Yale School of Medicine and Yale New Haven Health in the ACC statement. “We were able to dramatically increase the number of guideline-directed medical therapies used in a very rapid fashion.”
For patients with heart failure with reduced ejection fraction (HFrEF) there is compelling evidence for the use of the recommended combination of drugs from 4 different classes; health outcomes improve significantly. The authors say, however, that real-world data show persistent suboptimal use of the treatment, despite “resource-intensive interventions” to improve guideline adherence.
Ahamd and colleagues posited that an element missing from previous strategies to improve GDMT is that they have not been integrated into the existing clinical flow.
To assess whether that integration would make a difference, the researchers designed a pragmatic, multicenter EHR-based, randomized comparative effectiveness trial that enrolled 100 Yale New Haven Health clinicians caring for the largest number of patients with HFrEF.
The clinicians were randomized to receive a real-time tailored EHR-based alert while entering patient follow-up orders or to usual care. The customized alert, the authors explained, provided patient-specific clinical data commonly considered when adjustments are made to heart failure medications, which medications the patient was on, and indicators for which of the recommended GDMT classes the patient was not taking.
Ahmad et al designated the study’s primary endpoint as an increase in the number of GDMT classes prescribed at 30 days. The key secondary outcome of interest was any increase in dose of currently prescribed GDMT. Other secondary outcomes included filling of prescriptions, total health care costs, hospitalizations, emergency department visits, and death. Investigators also surveyed participating providers on user experience with the alert and knowledge of HFrEF guidelines.
PROMPT-HF clinician participants were enrolled between April and October 2021. Of the 100 clinicians, 69% were physicians and 31% advanced practice providers; together they cared for a population of 1310 patients. Baseline patient characteristics, the authors wrote, were well-balanced between the study groups. Medications prescribed at baseline were broken out as follows:
Beta blockers 84% ACE/ARB or ARNI 71% MRA 29% SGLT2 inhibitor 11%
During the study period, 75% of patients triggered 1 alert, 17% triggered 2 alerts, and 8% triggered ≥3 alerts.
The primary outcome, an increase in the number of GDMT classes prescribed at 30 days occurred, in 25.7% of the alert arm vs in 18.7% in the usual care arm (adjusted RR, 1.41 [95% CI, 1.03-1.93]; P=.03). The investigators also report numerical increases for each GDMT class in both study arms, with greater increases seen among those in the alert group. Ahmad et al calculated the number of alerts needed for one medication class addition as 14.
When data were analyzed for the study’s secondary outcomes, Ahmad and colleagues found an increase in dose or addition of a GDMT class in 36.2% of the alert arm and 26.2% of the usual care arm (aRR, 1.39 [95% CI, 1.08-1.79]; P=.01). The number needed to alert for this outcome was 10. They report no significant differences between the study arms for rate of emergency department visits or hospitalizations at 30 days. Similarly, no differences were seen for safety outcomes between the alert and usual care groups.
When they reviewed participant survey responses, the team found that 79% of clinicians in the alert group agreed or strongly agreed the prompts were effective at improving prescription of GDMT. Among these clinicians, 25% accepted the recommendations made by the prompt, 48% said they will address the suggested interventions in the future, 14% indicated that the patients do not meet the criteria, and the rest did not acknowledge the alerts. When asked a reason for not acting on a recommendation the primary responses were hypotension, renal failure/decompensation, unable to tolerate the medication, and patient refusal.
The authors note several study limitations including use of a single academic health system, deployment of the alert in a single (Epic) HER, and a focus on an increase in medication initiation vs an increase in dosing.
Reference: Ghazi L, Yamamoto y, Riello RJ, et al. Pragmatic trial aimed at improving use of guideline directed medical therapy in outpatients with HF. Presented at and simultaneously published in the Journal of the American College of Cardiology.