Tailored EHR Alerts Fail to Improve Acute HF Medication Prescription at Discharge

Clayton D – stock.adobe.com

Despite the implementation of a real-time, targeted, tailored electronic health record (EHR) alert system for patients with acute heart failure (AHF), the PROMPT-AHF trial did not find a significant increase in overall guideline-directed medical therapy (GDMT) prescriptions at hospital discharge.

Findings were published in European Heart Journal.

Patients hospitalized for AHF are at particularly high risk for adverse clinical events, and they rarely receive an adequate number of GDMT prescriptions at hospital discharge despite evidence of their benefits and safety at this time. Instead, this typically occurs during post-discharge care, which can delay or prevent patients from receiving prescriptions for life-saving therapies.

To determine whether an EHR-based alert system would increase the rates of GDMT prescription at discharge, PROMPT-AHF—a pragmatic, multicenter, randomized clinical trial—automatically enrolled patients 48 hours after hospital admission if they met specific criteria for AHF hospitalization.

These criteria included:

• Being aged 18 years or older
• Having N-terminal pro-B-type natriuretic peptide (NT-proBNP) levels > 500 pg/mL
• Having left ventricular ejection fraction ≤ 40%
• Having received intravenous (IV) diuretics within 24 hours of hospital admission
• Not being on any of the 4 classes of GDMT, including beta blockers, angiotensin-converting enzyme inhibitors (ACE-I)/angiotensin receptor blockers (ARB)/angiotensin receptor-neprilysin inhibitors (ARNI), mineralocorticoid receptor antagonists (MRA), and sodium-glucose cotransporter 2 inhibitors (SGLT2i)

Patients were excluded if they opted out of EHR-based research, were receiving hospice care, were unable to tolerate oral medication administration, were currently hospitalized in an intensive care unit, or were receiving IV inotropic agents.

Between May 2021 and November 2022, a total of 1012 patients were enrolled in the study, with 502 in the intervention arm and 510 in the control arm. The median age of the participants was 74 years, with 26% being female and 24% being Black. At the time of the alert, 85% of patients were already on beta blockers, 55% were on ACE-I/ARB/ARNI, 20% were on MRA, and 17% were on SGLT2is.

In the intervention arm of the study, providers received an alert during order entry that included relevant patient characteristics and individualized GDMT recommendations with links to an order set.

The study's primary outcome was the increase in the number of GDMT prescriptions at discharge. Surprisingly, the results showed that the tailored EHR alert system did not significantly impact prescriptions, with both the alert and no alert groups exhibiting a similar rate of GDMT prescriptions at discharge (adjusted risk ratio [RR], 0.95; 0.81-1.12, P = .99)

However, patients randomized to the alert arm were more likely to have an increase in MRA prescriptions at discharge (adjusted RR, 1.54; 1.10-2.16; P = .01).

At the time of discharge, only 9 (11.2%) patients in the study received all 4 classes of GDMT, indicating that comprehensive evidence-based regimens for HF were not being consistently prescribed.

The underlying reasons for these findings are not entirely clear, but one possible explanation the authors gave could be that these patients receive care from multiple health care providers whose primary focus may not be enhancing adherence to GDMT during an AHF hospitalization. This can be attributed to various factors, such as the need to shorten hospital stays, addressing various concurrent medical conditions, exposure to numerous EHR alerts, a diffusion of responsibility among care providers, and the expectation that optimizing GDMT for a chronic disease primarily falls under outpatient care. According to the authors, these findings align with established psychological concepts like the bystander effect and the prominence of certain issues over others.

The authors also said the phenomenon of “alert fatigue” could be a key cause.

“This is likely due to the largely unregulated increase in alerts used as clinical decision-support tools to implement best practices,” they explained. “Our study highlights the importance of such alerts being studied in a randomized fashion before widespread implementation, and with input from frontline clinicians, in order to achieve their stated purpose and to reduce burnout.”

Overall, PROMPT-AHF showed that GDMT prescription remains suboptimal at hospital discharge.

“Further refinement and improvement of such alerts and changes to clinician incentives are needed to overcome barriers to the implementation of GDMT during hospitalizations for AHF,” the study authors concluded.

Reference

Ghazi L, Yamamoto Y, Fuery M, et al. Electronic health record alerts for management of heart failure with reduced ejection fraction in hospitalized patients: the PROMPT-AHF trial. Eur Heart J. Published online August 31, 2023. doi:10.1093/eurheartj/ehad512