Abstract 6: Early Telephone Follow-up Fails to Reduce Readmission Rates for CHF & AMI.

Background: To reduce readmissions rates for patients admitted with acute myocardial infarction (AMI) or congestive heart failure (CHF), Duke University Hospital piloted a program in which patients were contacted by telephone within 72 hours of discharge and asked a set of structured questions believed to identify gaps in care to reduce the risk of readmission. Methods: A cohort of 838 patients admitted between December 2012 and May 2013 with a primary diagnosis of AMI (n=374) or CHF (n=464) were identified by DRG for prospective telephone follow-up and analyzed retrospectively. For primary outcome analysis, readmission rates were compared between two groups: one that received the telephone intervention and one that did not. If successfully contacted, patients were asked about filling and complying with their prescriptions, knowledge of their follow-up appointment, and understanding their discharge instructions. A secondary outcome analysis evaluated readmission rates among the telephone-eligible subgroup. Results: Patient characteristics among those contacted vs. not contacted did not differ substantially for either AMI or CHF discharges. Among AMI patients, there was no association between contact attempt and readmission within 30 days (p = 0.719) or 90 days (p=0.803). This finding persisted after adjusting for the effect of age at hospitalization, disease group, race, gender, and payor (OR = 1.28, 95% CI = 0.61 - 2.68, p = 0.512 at 30 days; OR = 1.24, 95% CI = 0.65 - 2.37, p = 0.510 at 90 days). Among CHF patients, there was no association between contact attempt and readmission within 30 days (p = 0.156) or 90 days (p=0.904). This finding persisted after adjusting for the effect of age at hospitalization, disease group, race, gender, and payor (OR = 0.67, 95% CI = 0.35 - 1.25, p = 0.207 at 30 days; OR = 0.94, 95% CI = 0.56 - 1.55, p = 0.797 at 90 days). There was no association between telephone-eligible patients and readmission rates at 30 or 90 days for either condition. Among patients successfully contacted, there was no association between answers pertaining to post-discharge behavior and readmission rates for CHF or AMI. Conclusions: Early telephone follow-up targeting post-discharge behaviors was not observed to impact 30- or 90-day readmission rates for patients admitted to an academic medical center with CHF or AMI. ![][1] [1]: /embed/graphic-1.gif