A Prediction Model to Identify Acute Myocardial Infarction (AMI) Patients at Risk for 30-Day Readmission

摘要

Reductions in hospital readmissions have been identified by Congress and President Obama as a source for reducing Medicare spending. We aimed to build a prediction model for identifying acute myocardial infarction (AMI) patients who are at risk for unplanned readmission. This is a retrospective study in patients who suffered with AMI during the period from 2010 to 2012 at OSF HealthCare, a multi-site healthcare service. Among 3,058 AMI admissions, the average 30-day readmission rate was 8.9%, and it was more likely to occur among those who were Black American (18.2%), elderly (14.5% for age≥85 years), having a longer length of stay (LOS) [20.2% for LOS>7 days], who were without cardiology services (18.4%), who lived in a metropolitan area (10.6%), and those with comorbidities (10%-23%) except for obesity. For the prediction model, the area under the receiver operating characteristic curve (AUROC) was 0.739 as well as having 70.2% sensitivity and 67.7% specificity given a cut-off point of 0.08. Other three cut-off points (0.06, 0.12 and 0.20) were also selected for classifying patients into four risk levels: low, medium, high and higher. It is feasible to use routine electronic medical record (EMR) data to identify AMI patients at risk of 30-day readmission. Multi-level interventions could be developed and tailored according to individual risk of readmission.