Introduction: Atrial fibrillation (AF) is the most common cardiac arrhythmia and can lead to significant comorbidities and mortality. Persistence with oral anticoagulation (OAC) is crucial to prevent stroke but rates of discontinuation are high. This systematic review explored underlying reasons for OAC discontinuation.
Methods: A systematic review was undertaken to identify studies that reported factors influencing discontinuation of OAC in AF, in 11 databases, grey literature and backwards citations from eligible studies published between 2000 and 2019. Two reviewers independently screened titles, abstracts and papers against inclusion criteria and extracted data. Study quality was appraised using Gough's weight of evidence framework. Data were synthesised narratively.
Results: Of 6,619 sources identified, 10 full studies and 2 abstracts met the inclusion criteria. Overall, these provided moderate appropriateness to answer the review question. Four reported clinical registry data, six were retrospective reviews of patients' medical records and two studies reported interviews and surveys. Nine studies evaluated outcomes relating to dabigatran and/or warfarin and three included rivaroxaban (n = 3), apixaban (n = 3) and edoxaban (n = 1). Bleeding complications and gastrointestinal events were the most common factors associated with discontinuation, followed by frailty and risk of falling. Patients' perspectives were seldom specifically assessed. Influence of family carers in decisions regarding OAC discontinuation was not examined.
Conclusion: The available evidence is derived from heterogeneous studies with few relevant data for the newer direct oral anticoagulants. Reasons underpinning decision-making to discontinue OAC from the perspective of patients, family carers and clinicians is poorly understood.
Keywords: adherence; atrial fibrillation; discontinuation; older people; oral anticoagulation.
© The Author(s) 2021. Published by Oxford University Press on behalf of the British Geriatrics Society. All rights reserved. For permissions, please email: email@example.com.