Frailty, Cognitive Impairment, and Anticoagulation Among Older Adults withNonvalvularAtrial Fibrillation

BACKGROUND/OBJECTIVES Oral anticoagulation (OAC) is challenging in older patients with nonvalvular atrial fibrillation (NVAF) who are often frail and have cognitive impairment. We examined the characteristics of older NVAF patients associated with higher odds of physical and cognitive impairments. We also examined if these high-risk patients have different OAC prescribing patterns and their satisfaction with treatment because it may impact optimal management of their NVAF. METHODS The patients in the Systematic Assessment of Geriatric Elements in Atrial Fibrillation (SAGE-AF study cohort 2016-2018) had NVAF, were aged 65 and older, and eligible for the receipt of OAC. Measures included frailty (Fried Frailty scale), cognitive impairment (Montreal Cognitive Assessment Battery), OAC prescribing and type (direct oral anticoagulant [DOAC] or vitamin K antagonist [VKA]), depressive symptoms (Patient Health Questionnaire-9), bleeding, stroke risk, and treatment benefit (Anti-Clot Treatment Scale). RESULTS Patients (n = 1,244) were 49% female, aged 76 (standard deviation = 7) years. A total of 14% were frail, and 42% had cognitive impairment. Frailty and cognitive impairment co-occurred in 9%. Odds of having both impairments versus none were higher with depression (odds ratio [OR] = 4.62; 95% confidence interval [CI] = 2.59-8.26), older age (OR = 1.56; 95% CI = 1.29-1.88), lower education (OR = 3.81; 95%CI = 2.13-6.81), race/ethnicity other than non-Hispanic White (OR = 7.94; 95% CI = 4.34-14.55), bleeding risk (OR = 1.43; 95% CI = 1.12-1.81), and stroke risk (OR = 1.35; 95% CI = 1.13-1.62). OAC prescribing was not associated with CI and frailty status. Among patients taking OACs (85%), those with both impairments were more likely to take DOAC than VKA (OR = 1.69; 95% CI = 1.01-2.80). Having both impairments (OR = 1.87; 95% CI = 1.08-3.27) or cognitive impairment (OR = 1.56; 95% CI = 1.09-2.24) was associated with higher odds of reporting lower treatment benefit. CONCLUSION In a large cohort of older NVAF patients, half were frail or cognitively impaired, and 9% had both impairments. We highlight the characteristics of patients who may benefit from cognitive and physical function screenings to maximize treatment and enhance prognosis. Finally, the co-occurrence of impairment was associated with low perceived benefit of treatment that may impede optimal management.