A Computable Algorithm for Medication Optimization in Heart Failure With Reduced Ejection Fraction

Area of Science:

• Cardiology
• Medical Informatics
• Clinical Trials

Background:

• Optimizing guideline-directed medical therapy (GDMT) is crucial for improving outcomes in patients with heart failure with reduced ejection fraction (HFrEF).
• Suboptimal GDMT is a common challenge in clinical practice, even within structured clinical trials.

Purpose of the Study:

• To evaluate a novel computable algorithm designed to recommend GDMT for HFrEF.
• To assess if the algorithm's recommendations align with actual medication adjustments and if its output correlates with patient outcomes.

Main Methods:

• Utilized clinical trial data from GUIDE-IT and HF-ACTION studies.
• Applied a computable medication optimization algorithm to generate GDMT recommendations and a medication optimization score (MOS).
• Compared algorithm-based recommendations to actual medication changes and analyzed the association between MOS and clinical endpoints using Cox proportional-hazards models.

Main Results:

• The algorithm identified significant opportunities for initiating and titrating GDMT (ACEI/ARB, beta-blockers, MRAs) that were underutilized in the trials.
• A higher baseline MOS was significantly associated with a reduced risk of cardiovascular death or heart failure hospitalization in GUIDE-IT (HR: 0.41) and all-cause death/hospitalization in HF-ACTION (HR: 0.61).
• The algorithm demonstrated accuracy in identifying patients who could benefit from GDMT optimization.

Conclusions:

• The computable algorithm effectively identifies patients requiring GDMT optimization in HFrEF.
• The algorithm-generated MOS is a predictor of improved clinical outcomes, highlighting its potential clinical utility.
• Implementation of this algorithm in clinical settings could address suboptimal GDMT and improve care for HFrEF patients.