Bleeding risk and healthcare resource utilisation in elderly patients treated with edoxaban or vitamin K antagonists for atrial fibrillation in Italy

Abstract

Abstract Funding Acknowledgements Type of funding sources: Other. Main funding source(s): Daiichi Sankyo Europe GmbH Background Direct oral anticoagulants (DOACs) have been shown to be non-inferior to vitamin K antagonists (VKA) regarding both efficacy and safety outcomes in patients with atrial fibrillation (AF). However, elderly are underrepresented in the underlying clinical trials. Purpose The aim of this study was to compare risks of major bleeding events and compare healthcare resource utilization (HRU) between AF patients treated with edoxaban or VKA in elderly. Methods A cohort study was conducted using claims databases of Caserta and Palermo Local Heath Units in Italy. AF patients starting use of edoxaban or VKA between August 1st, 2016 and December 31st, 2018 were included. Date of the first dispensing was defined as the index date. The study population was matched based on a propensity score based on factors associated with the outcome. We restricted to patients aged ≥65, ≥1 year database history, and no use of the index drug in the year before index date. Incidence rates of bleeding outcomes and rates of HRU were assessed per 1,000 and 100 person-years follow-up (PY), respectively. Cox regression analyses to adjust for baseline covariates were used for comparisons of incidence rates of bleeding outcomes among all edoxaban and VKA users. Poisson regression analyses were used for comparisons of rates of HRU among all edoxaban and VKA users. Both analyses were adjusted for age. Sex, region and year of index date were considered for the adjusted models as well, using a backward stepwise approach to select eligible variables. Results 1,317 edoxaban users and 2,924 VKA users were included in the matched population. Mean age was 79 in both treatments groups, and 43% of the edoxaban users and 45% of the VKA users was male. Bleeding risks were significantly lower among edoxaban users compared to VKA users aged ≥65 (adjusted HR 0.39 (95% CI 0.19-0.83)) and among patients aged ≥75 (adjusted HR 0.37 (95% CI 0.16-0.86)). Among patients aged ≥65, edoxaban users were significantly less often hospitalised (RR 0.56 (95% CI 0.46-0.68)) and the total number of hospitalised days were also significant lower (RR 0.58 (95% CI 0.42-0.80)) compared to VKA users. Among patients aged ≥75, similar results were observed for the number of hospitalisations. Edoxaban users had significant less out-patient visits compared to VKA users (among patients aged ≥65 the RR was 0.44 (95% CI 0.39-0.50) and among patients ≥75 this was 0.40 (95% CI 0.35-0.47). Use of out-patient medication use was significantly lower among edoxaban users compared to VKA users among patients aged ≥65 (adjusted RR 0.91 (95% CI 0.88-0.95)) as well as among patients aged ≥75 (adjusted RR 0.91 (95% CI 0.87-0.95)). Conclusion Study results show a decreased bleeding risk of edoxaban compared to VKA in both age groups of patients with AF. Hospital based HRU has shown to be lower among edoxaban users compared to VKA users in both age groups. Out-patient HRU was also lower among edoxaban users.