Arrhythmias & electrophysiology
Heterogeneity of Treatment Effect of ABC Pathway–Based Care in Older Atrial Fibrillation: Insights from the mAFA-II and MIRACLE-AF Trials Using a Causal Machine Learning Approach
Authors' conclusion
ABC pathway-based integrated care demonstrates clinically meaningful treatment heterogeneity among older patients with AF. These findings support more personalised implementation of integrated care and highlight the potential of causal ML to guide precision management in AF.
Study context
- Population
- Patients aged ≥65 years with atrial fibrillation (AF)patients aged ≥65 years from the mAFA-II and MIRACLE-AF trials
- Participants
- 3,209 patients
- Comparison
- Heterogeneity of treatment effect (HTE) of ABC pathway-based care, across patients predicted to benefit vs predicted to derive harm
- Primary endpoint
- Composite of:
- All-cause mortality
- Ischaemic stroke
- Intracranial haemorrhage
- Heart failure
- Acute coronary syndrome
- Major bleeding
Methodological detailsmodels developed in mAFA-II and externally validated in MIRACLE-AF
- Models developed in mAFA-II and externally validated in MIRACLE-AF
Primary result
As reported in the abstract
- mAFA-II testing cohort: ABC pathway-based care associated with a lower risk of the primary outcome in patients predicted to benefit; no benefit and possible harm in those predicted to derive harm
- P-interaction=0.002
Further findings
Model performance
- The uplift random forest demonstrated the best performance and identified meaningful HTE
Characteristics of patients predicted to derive limited benefit
- Greater clinical complexity, including higher comorbidity burden and CHA2DS2-VASc scores
Predictors of treatment effect
- Sex, heart failure, and hypertension were the strongest predictors of treatment effect
Original publication
Heterogeneity of Treatment Effect of ABC Pathway–Based Care in Older Atrial Fibrillation: Insights from the mAFA-II and MIRACLE-AF Trials Using a Causal Machine Learning Approach. Heart Rhythm. Published 30 October 2026. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →