Arrhythmias & electrophysiology
Identifying treatment heterogeneity in atrial fibrillation using a novel causal machine learning method
Authors' conclusion
Among patients with AF treated with a NOAC or warfarin, a causal ML method identified patient subgroups with differences in outcomes associated with OAC use.
Study context
- Population
- Patients initiating a NOAC or warfarin for nonvalvular atrial fibrillation (AF), OptumLabs Data Warehouse, 08/01/2010–11/29/2017
- Participants
- 34,569 patients
- Exposure
- Apixaban vs dabigatran vs rivaroxaban vs warfarin
- Outcome
- Composite of:
- Ischemic stroke
- Intracranial hemorrhage
- All-cause mortality
- Follow-up
- Mean 8.3 (SD, 9.0) months
Methodological detailscausal machine learning, machine learning (ML)-matched
- Causal machine learning
- ML-matched
Main finding
As reported in the abstract
Subgroups with heterogeneous head-to-head OAC effects on composite of:
- Ischemic stroke
- Intracranial hemorrhage
- All-cause mortality
Subgroups favoring apixaban over dabigatran
5
Subgroups favoring apixaban over rivaroxaban
2
Subgroup favoring dabigatran over rivaroxaban
1
Subgroup favoring rivaroxaban over dabigatran
1
Observational data: associations, not effects.
Further findings
Composite outcome and death, whole cohort · mean follow-up 8.3 (SD, 9.0) months
- Composite outcome
- 2,110 (6.1%)
- Died
- 1,675 (4.8%)
Original publication
Identifying treatment heterogeneity in atrial fibrillation using a novel causal machine learning method. Am Heart J. Published 7 March 2023. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →