BMJ Open. 2026 Jun 22;16(6):e112736. doi: 10.1136/bmjopen-2025-112736.
ABSTRACT
OBJECTIVES: This study aimed to investigate the association and potential causal relationship between ischaemic stroke/transient ischaemic attack (TIA) and oral anticoagulant (OAC) adherence using Bradford Hill's criteria and to determine the factors influencing OAC adherence following ischaemic stroke in atrial fibrillation (AF) patients.
DESIGN: We conducted a scoping review.
DATA SOURCES: MEDLINE, Embase and Web of Science from inception up to March 2026.
RESULTS: The analysis included 43 studies on OAC adherence across different phases: initiation (n=1), implementation (n=15), implementation and discontinuation (n=7) and persistence (n=20). Of these, 26 studies showed a positive association between ischaemic stroke history and OAC adherence while one showed a negative association. The evaluation of the body of evidence suggests a mixed association between ischaemic stroke and subsequent OAC adherence. While causality is not definitively established due to the observational nature of the included studies, they fulfil several of the classic Bradford Hill criteria, including temporality, consistency, biological gradient, coherence and plausibility. Factors like age, chronic obstructive pulmonary disease, cancer, heart failure, hypertension, myocardial infarction, anaemia, dementia, smoking, prior warfarin use, education level, stroke severity and recurrence were associated with decreased OAC adherence, whereas experiencing a first stroke was associated with increased adherence.
CONCLUSIONS: Existing evidence shows mixed associations between ischaemic stroke and OAC adherence and causality also remains inconclusive. There are multiple factors that modify this relationship. Further observational studies using causality methods are needed to definitively confirm this relationship.
PMID:42331577 | DOI:10.1136/bmjopen-2025-112736