• Carl van Walraven and Alan J Forster.
• Faculty of Medicine, University of Ottawa, Ottawa Hospital Research Institute, ASB1-003 1053 Carling Avenue, Ottawa, Ontario K1Y 4E9, Canada. carlv@ohri.ca
• J Clin Epidemiol. 2013 Jun 1; 66 (6): 688-90.

ObjectivesHospitals have strong incentives to decrease readmission rates. Not all hospital readmissions are potentially avoidable. Therefore, only a component of all hospital readmissions can be influenced by interventions designed to decrease them. In this study, we determined how effective interventions must be to attain specific reductions in hospital readmission rates.Study Design And SettingA conceptual model of all readmissions and potentially avoidable readmissions was used to derive a mathematical relationship between the relative reduction in the total number of readmissions, the relative reduction in potentially avoidable readmissions, and the proportion of readmissions that are potentially avoidable.ResultsWhen 22% of readmissions were potentially avoidable, achieving a 20% reduction in the total number of readmissions required a 91% reduction in potentially avoidable readmissions; decreasing potentially avoidable readmissions by 20% reduced total readmissions by 4.4%.ConclusionThese results highlight that relative reductions in the total number of readmissions are notably lower than that for potentially avoidable readmissions. This separation in relative reduction of all and potentially avoidable readmissions increases as the proportion of readmissions deemed potentially avoidable decreases. These results have important implications for health care planners and researchers.Copyright © 2013 Elsevier Inc. All rights reserved.

10 keywords...

hide…