Cluster-Randomized Trial Comparing Ambulatory Decision Support Tools to Improve Heart Failure Care
Authors' conclusion
An automated, patient-specific, electronic health record-embedded alert increased MRA prescribing compared to both a message and usual care.
Study context
- Population
- Adults with heart failure with reduced ejection fraction (HFrEF), no active MRA prescription, no contraindication to mineralocorticoid receptor antagonists (MRAs), and an outpatient cardiologist in a large health system
- Participants
- 2,211 adults
- Comparison
- EHR-embedded alert vs message vs usual care on MRA prescribing
- Alertn = 755
- Messagen = 812
- Usual caren = 644reference
- Primary endpoint
- New MRA prescribing
Methodological details3-arm, pragmatic, cluster-randomized by cardiologist (60 per arm)
- 3-arm
- Pragmatic
- Cluster-randomized by cardiologist (60 per arm)
Primary result
As reported in the abstract
Relative risk (95% CI)P value
Alert vs Usual care
2.53(1.77–3.62)
P < 0.0001Alert vs Message
1.67(1.21–2.29)
P = 0.002- NNT with alert for an additional MRA prescription
- 5.6
Original publication
Cluster-Randomized Trial Comparing Ambulatory Decision Support Tools to Improve Heart Failure Care. J Am Coll Cardiol. Published 5 March 2023. doi:10.
Summarised from the published abstract; every number is checked against it