Abstract P1: Limitations of Self-Reported Readmission Data After Acute Myocardial Infarction: Insights from the TRIUMPH Registry
Background: Self-reported readmission rates are frequently reported in the medical literature, yet the validity of these data is controversial. Few studies describe the accuracy of self-report of readmission following an AMI in comparison with physician-adjudicated data. Methods: We studied 4,340 AMI patients enrolled in the 24-US center TRIUMPH registry. Patients were interviewed at 1, 6 and 12 months after their AMI, and were asked to report all hospitalizations since their last contact, including the hospital name, date and reason. After obtaining consent from the patient and each hospital, all hospitalization records within the first year after the patient's index MI were requested and adjudicated by a physician panel. Accuracy of patients’ report of hospitalization and reason for admission (sensitivity, specificity) were assessed. Results: Of 4,340 patients, 3,633 (84%) completed follow-up interviews, reporting a total of 2,016 readmissions. Of these, hospital records were successfully obtained on 1,373. Record review revealed that 501 (36%) were not actual rehospitalizations (e.g., emergency department only, outpatient visits, admissions prior to study enrollment). Interestingly, when obtaining hospital records, we identified another 394 readmissions that were not reported by patients. Sensitivity of self-reported reason for admission was modest to poor for cardiac-cause rehospitalization, AMI and percutaneous coronary intervention (Table). Conclusions: We identified several limitations of self-reported readmission rates in this multi-center AMI cohort. Emergency department and outpatient visits were frequently reported as hospital admissions, nearly 400 hospitalizations were not reported to study personnel at the time of the follow-up interview, and accuracy of patient-reported reason for admission was modest at best. These data underscore the importance of verifying self-reported follow-up outcomes data. Accuracy of Patient Self-Reported Reason for Admission Adjudicated Reason for Admission Sensitivity Specificity Any cardiac 37% 88% AMI 43% 94% PCI 66% 93%