Introduction: Rehospitalization after discharge for a wide variety of medical conditions is common and is associated with increased morbidity and cost. We investigated the rates of rehospitalization for stroke in a population of patients hospitalized with atrial fibrillation (AF) who were at risk of stroke.
Methods: Using the Premier Healthcare Database (representing 1 in 5 US hospital discharges), we identified 1,147,866 patients hospitalized from January 1, 2011 through June 30, 2015 with AF and a guideline indication for an oral anticoagulant based on a CHA2DS2-VaSC score ≥ 2. The primary outcome was readmission within 90 days due to acute ischemic stroke (AIS) or intracranial hemorrhage (ICH). Multivariable logistic regression was used to compare rehospitalization rates for patients receiving various oral antithrombotic therapies, while controlling for numerous patient risk factors, reasons for current hospitalization, and hospital characteristics. Confidence intervals and p-values were adjusted for multiple comparisons using Holm-Bonferroni.
Results: Patients were 53% female with a mean age of 77 (±11) years. The distribution of patients on various antithrombotic agents at discharge was 23% with no therapy, 20% ASA only, 1% dual-antiplatelet therapy (DAPT), 17% non-vitamin K oral anticoagulants (NOAC), and 39% warfarin. Table 1 displays rehospitalization rates, unadjusted (therapy only), and adjusted for all covariates using logistic regression analysis.
Conclusions: AF patients discharged on ASA alone or no antithrombotic therapy, compared to warfarin, were more likely to have early rehospitalization for acute ischemic stroke. Patients treated with NOACs had early ischemic stroke rehospitalization rates similar to warfarin, but were less likely to be have early rehospitalization for ICH when compared to warfarin.
Table 1: 90-Day Rehospitalization Rates by Antithrombotic Agents