We read with interest the study by Brunetti et al. on oral anticoagulant (OAC) nonprescription in older atrial fibrillation patients with reduced expected survival [1]. The authors should be commended for addressing a clinically challenging question. However, we identify two methodological errors, neither acknowledged in the limitations, that fundamentally invalidate their comparison of thromboembolic and bleeding events between OAC-prescribed and nonprescribed groups.
