TARGET-EU: Rivaroxaban and risk of major gastrointestinal bleeding in elderly patients with non-valvular atrial fibrillation

This entry contains the code list used in case study 4 that is part of the broader TARGET-EU project (EUPAS1000000539), which aims to advance the regulatory use of real-world data through the application of target trial emulation and estimand methodologies. Background: Non-valvular atrial fibrillation (NVAF) is a common arrhythmia associated with significant morbidity and mortality. Direct oral anticoagulants (DOACs) are recommended over vitamin K antagonists in NVAF patients. Evidence suggests clinically important differences in bleeding risk between DOACs, particularly in high-risk populations such as the elderly, who are often underrepresented in randomised controlled trials. Objectives: The primary objective is to estimate the effect of initiating rivaroxaban versus apixaban on time to a first major gastrointestinal (GI) bleeding event among adults aged 75 years or older with non-valvular atrial fibrillation. Methods: We will conduct an active comparator new-user cohort study using linked electronic health records from Denmark (Danish national registers) and Spain (SIDIAP). Eligible individuals are adults aged 75 years or older who initiated rivaroxaban or apixaban between 2013 and 2023. In the primary analysis, a while-on-treatment strategy is used for treatment-related intercurrent events (discontinuation and switching), and a while-alive strategy is applied to non-GI bleeding death. Inverse probability of treatment weighting (IPTW) is used to estimate treatment effects. The primary analysis uses a Cox proportional hazards model, with supplemental analyses using an accelerated failure time model to estimate restricted mean survival time at 1 and 2 years. Sensitivity analyses will assess the impact of departures from key assumptions, including censoring-at-random (via inverse probability of censoring weighting) and outcome misclassification (via probabilistic bias analysis).