In-Hospital Management of Atrial Fibrillation: The CHADS2 Score Predicts Increased Cost

Abstract Background Hospitalizations for atrial fibrillation (AF) impose a substantial burden on our health care system, and AF management strategies are increasingly focused on hospitalization reduction. The objectives of this study were to determine the cost of hospitalization for AF and to identify the main determinants of this cost in a Canadian setting. Methods Our study population consisted of patients hospitalized for AF and/or atrial flutter at a tertiary care hospital in Canada between April 1, 2001, and March 31, 2007. Patient-level demographics and data on clinical resource use and cost of treatment were collected from a computerized resource use and cost accounting system. The main determinants of in-hospital costs were identified through Bayesian model averaging. Results Data were collected on 325 consecutive hospitalizations for AF. The median length of stay was 5 days (interquartile range [IQR], 3-9). The mean cost of an AF admission was CAD$4740 (SD = CAD$4457), and the median was CAD$3532 (IQR, CAD$2013-CAD$5944). Multivariate analysis identified 2 independent predictors of increased cost: CHADS2 score (relative increase in cost: 1.24; 95% CI, 1.16-1.33) and warfarin use (relative increase in cost: 1.41; 95% CI, 1.20-1.67). These 2 variables were also independent predictors of increased length of stay. Conclusions The main clinical determinants of increased cost and increased length of stay were CHADS2 score and warfarin use. Strategies for reducing AF-related costs should focus on preventing hospitalization or decreasing its length in patients with high CHADS2 scores and on finding alternatives to the use of warfarin or using outpatient bridging anticoagulation to facilitate earlier hospital discharge.