Valvular & structural heart disease
Automated Alerts to Improve Timely Evaluation and Treatment of Valvular Heart Disease: The ALERT Trial
Primary resultResult — go to the resultWin ratio 1.27 95% CI 1.05–1.54Authors' conclusion
In this multisystem cluster randomized trial, automated ECN alerts improved timely guideline-directed evaluation and valve intervention for clinically significant AS and MR.
Study context
- Population
- Clinicians ordering echocardiograms with significant aortic stenosis (AS) or mitral regurgitation (MR), 5 US health systems, August 2024–September 2025
- Participants
- 765 clinicians · 2,016 echocardiograms
- Comparison
- Automated electronic clinician notification (ECN) alert vs no alert
- Primary endpoint
- Hierarchical composite:
- Time to surgical or transcatheter valve intervention
- Then time to multidisciplinary heart team clinic evaluation (stratified win ratio)
- Follow-up
- 90 days
Methodological detailsmultisystem, cluster-randomized, 5 US health systems, 35 hospitals
- Multisystem
- Cluster-randomized
- 5 US health systems, 35 hospitals
Primary result
As reported in the abstract
Win ratio
1.27
95% CI 1.05–1.54
P = 0.007
ECN alert vs Usual care
win ratio 1.27 (95% CI 1.05–1.54); P = 0.007
Secondary outcomes
Valve intervention (surgical or transcatheter) · 90 days
P = 0.005
Multidisciplinary heart team evaluation · 90 days
P = 0.005
Primary endpoint by valve pathology (subgroup)
- Win ratio in AS
- 1.29
- Win ratio in MR
- 1.23
- No heterogeneity by valve pathology
- Pint = 0.821
- Interaction P = 0.821
Primary endpoint across prespecified subgroups (age, sex, race, social deprivation index, inpatient vs outpatient, provider specialty, rurality)
- No evidence of heterogeneity
- Pint > 0.100 for all
Sensitivity analyses (modified ITT, ITT, per-protocol)
- Consistent results across populations
Original publication
Automated Alerts to Improve Timely Evaluation and Treatment of Valvular Heart Disease: The ALERT Trial. J Am Coll Cardiol. Published 29 March 2026. doi:10.
Summarised from the published abstract; every number is checked against it