Medication regimen complexity and risk of bleeding in people who initiate oral anticoagulants for atrial fibrillation: a population-based study

BACKGROUND: Oral anticoagulants (OACs) are high-risk medications often used in older people with complex medication regimens. This study was the first to assess the association between overall regimen complexity and bleeding in people with atrial fibrillation (AF) initiating OACs. METHODS: Patients diagnosed with AF who initiated an OAC (warfarin, dabigatran, rivaroxaban, apixaban) between 2010 and 2016 were identified from the Hong Kong Clinical Database and Reporting System. Each patient's Medication Regimen Complexity Index (MRCI) score was computed. Baseline characteristics were balanced using inverse probability of treatment weighting. People were followed until a first hospitalization for bleeding (intracranial hemorrhage, gastrointestinal bleeding, or other bleeding) and censored at discontinuation of the index OAC, death, or end of the follow-up period, whichever occurred first. Cox regression was used to estimate hazard ratios (HR) between MRCI quartiles and bleeding during initiation and all follow-up. RESULTS: There were 19 292 OAC initiators (n = 9 092 warfarin, n = 10 200 direct oral anticoagulants) with a mean (standard deviation) age at initiation of 73.9 (11.0) years. More complex medication regimens were associated with an increased risk of bleeding (MRCI > 14.0-22.00: aHR 1.17, 95% confidence interval [CI] 0.93-1.49; MRCI > 22.0-32.5: aHR 1.32, 95%CI 1.06-1.66; MRCI > 32.5: aHR 1.45, 95%CI 1.13-1.87, compared to MRCI ≤ 14). No significant association between MRCI and bleeding risk was observed during the initial 30, 60, or 90 days of treatment. CONCLUSION: In this cohort study of people with AF initiating an OAC, a more complex medication regimen was associated with higher bleeding risk over periods longer than 90 days. Further prospective studies are needed to assess whether MRCI should be considered in OAC prescribing.