Introduction: Medicare’s Hospital Readmissions Reduction Program assesses financial penalties for hospitals based on risk-standardized readmission rates after specific episodes of care, including acute myocardial infarction (AMI). Whether the algorithm accurately identifies patients with AMI who have preventable readmission is unknown.
Methods: Using administrative data from Medicare, we conducted physician-adjudicated chart reviews of all patients considered 30 day readmissions after AMI attributed to one hospital from July 2012-June 2015. We extracted information about revascularization during index hospitalization. For patients readmitted to the index hospital or an affiliate, we also extracted reason for readmission.
Results: Of 199 admissions, 66 (33.2%) received PCI and 19 (9.6%) underwent CABG on index hospitalization. The remainder of patients did not receive any intervention, i.e. 39 patients (19.6%) were declined due to procedural risk, 15 (7.5%) because of goals of care and 14 (7.0%) refused revascularization. Forty-six patients (23.1%) had troponin elevation in the absence of an MI and did not have an indication for revascularization. The most common diagnoses of the 161 (80.9%) patients readmitted to the index hospital or an affiliate were infections and cardiac and non-cardiac chest discomfort (Table 1).
Conclusions: Our results demonstrate that many AMI patients who count towards the Medicare penalty do not receive revascularization during the index hospitalization because of high procedural risk or patient preference. Focusing on these patients may improve readmission metric performance. Furthermore, adding administrative codes for prohibitive procedural risk may improve accuracy of the metric as a measure of quality.