This is a feasibility study using a cross-over design to implement and compare a best practice alert (BPA) with an automated in-basket message to inform providers when a patient with heart failure and reduced ejection fraction (HFrEF) is not on appropriate medical therapy. The data from this pilot study will lead to a randomized controlled trial to compare the effectiveness of the BPA versus an automated in-basket message, versus usual care (no intervention).

An estimated 68,000 deaths per year nationwide can be attributed to gaps in care for patients with heart failure and reduced ejection fraction (HFrEF), with the majority being due to lack of mineralocorticoid receptor antagonists (MRA). Despite proven benefits in randomized trials, class I guideline recommendations, and published clinical performance measures, patients with HFrEF are often not on guideline-directed medical therapy (GDMT). While successful interventions for improvement in prescription of GDMT have often included multidisciplinary approaches with dedicated staff, the relatively high cost of hiring additional personnel has led to an interest in electronic health record (EHR)-based interventions. Prior studies on EHR-based interventions in this arena have mainly been conducted in the inpatient setting, which is limited to one encounter during acute hospitalization, a setting often complicated by renal dysfunction or hypotension that can limit prescription of MRA. The development and study of outpatient EHR-based alerts for HFrEF GDMT are needed. Two types of outpatient EHR-based interventions include best practice alerts (BPA) and automated in-basket messages. Both of these methods have limited data, with some studies showing benefit and others demonstrating provider fatigue and burnout. To our knowledge, there is no study that has directly compared these different types of EHR-based interventions.

This is a feasibility study using cross-over design at two outpatient clinics in a large health system to implement and compare a best practice alert (BPA) and an automated in-basket message to inform providers when a patient with heart failure and reduced ejection fraction (HFrEF) is not on appropriate medical therapy. The data from this study will lead to a randomized controlled trial to compare the effectiveness of the BPA versus an automated in-basket message, versus usual care (no intervention).

Inclusion Criteria:

• Cardiologist visit
• Transthoracic echocardiogram with the most recent EF >= 40%

Exclusion Criteria:

• Hypotension: SBP < 95
• Hyperkalemia: most recent K > 5.1, or any K >5.5
• Renal dysfunction: eGFR < 30
• Ventricular assist device
• Hospice care

Langone cardiology outpatients