Can a Predictive Model Identify AMI Patients at Risk of 90-Day Readmissions?

A clinical model that uses variables known at discharge may be effective in predicting 90-day readmission risk for patients with acute myocardial infarction (AMI), according to a study published Oct. 12 in Circulation: Cardiovascular Quality and Outcomes.

Using data from ACC's Chest Pain – MI Registry, Vinay Kini, MD, MSHP, et al., identified 86,849 Medicare beneficiaries who were discharged with a primary diagnosis of AMI and developed a model to assess a patient’s risk of readmission within 90 days.

The researchers randomly assigned 70 percent of the patients to the predictive model. According to the results, 23,912 patients (27.5 percent) were readmitted within 90 days. More than half of all readmissions (55 percent) occurred within 30 days, and 81 percent occurred within 60 days. Predictors of readmission included older age and a history of diabetes or heart failure.

The researchers then validated the model in the remaining 30 percent of patients. Results showed that there were no significant differences between the derivation and validation groups. The entire cohort was stratified into deciles of predicted readmission risk, ranging from a 13.1 percent risk for those in the lowest decile vs. 42.9 percent for those in the highest decile.  

The authors conclude that use of the model could identify patients at the highest risk of readmission within 90 days and “improve value by enabling” clinicians to target interventions toward these patients.

In an accompanying editorial, Cian P. McCarthy, MBBCh, and Ambarish Pandey, MD, write that predictive models “are a welcome first step in response” to value-based payment models and that future research is necessary to validate the results.

Keywords: Patient Readmission, Medicare, Heart Failure, Patient Discharge, Registries, Myocardial Infarction, Diabetes Mellitus, Chest Pain, Cohort Studies, National Cardiovascular Data Registries, Chest Pain MI Registry

< Back to Listings