Abstract 19943: International Variation in 30-day Unscheduled Cardiovascular Rehospitalization Rates Following Acute Myocardial Infarction in the Global Registry of Acute Coronary Events (GRACE)
Background: Readmission after AMI is common and costly. There is paucity of data to define international rates of 30-day readmission. We performed an analysis of the GRACE registry to characterize predictors of and international variation in readmission rates after AMI.
Methods: Data were evaluated from 26,268 patients with AMI enrolled in GRACE from 14 countries between 1999 and 2007. Patients who had an unscheduled hospitalization for cardiovascular disease within 1 month of discharge were identified. Stepwise logistic regression analysis was performed to evaluate for independent predictors of 30-day readmission. In analyses adjusted for individual predictors of GRACE score and patient-level length of stay (LOS), odds ratios for readmission for each country vs. the United States were calculated.
Results: 1856 patients (7.1%) were readmitted for a cardiovascular cause within 30 days. There was significant variation in unadjusted readmission rates across countries (P<0.001). There was a trend toward higher 6-month readmission rates in those countries with lower 30-day readmission rates (R=0.55, p=0.05). US enrollment was an independent predictor of readmission (OR, 1.50; 95% CI, 1.34-1.69) compared to rest of countries. After adjustment for country-level median LOS, US location was no longer an independent predictor of readmission. In analyses adjusted for individual predictors of GRACE score and patient-level LOS, the odds of readmission vs. the United States remained significantly lower in the following clusters: Argentina, Belgium, Canada, Germany, Poland, and Spain (Table).
Conclusions: There is significant international variation in readmission rates following AMI. The impact of US location as a predictor of readmission is significantly attenuated after adjustment for country-level LOS. Select countries have significantly lower risk-adjusted odds of readmission. This may reflect variations in in-hospital and/or post-discharge care.
Author Disclosures: P. Vaishnava: None. W. Huang: None. F. Anderson: None. D. Brieger: None. P. Steg: None. J. Gore: None. K. Fox: None. K. Eagle: None.
- © 2014 by American Heart Association, Inc.