Background: Readmission rates for patients discharged with heart failure approach 50% within 6 months. Identifying factors to predict risk of readmission in these patients could help clinicians focus resource-intensive disease management efforts on the high-risk patients.

Methods: The study sample included patients 65 years of age or older with a principal discharge diagnosis of heart failure who were admitted to 18 Connecticut hospitals in 1994 and 1995. We obtained patient and clinical data from medical record review. We determined outcomes within 6 months after discharge, including all-cause readmission, heart failure-related readmission, and death, from the Medicare administrative database. We evaluated 2176 patients, including 1129 in the derivation cohort and 1047 in the validation cohort.

Results: Of 32 patient and clinical factors examined, 4 were found to be significantly associated with readmission in a multivariate model. They were prior admission within 1 year, prior heart failure, diabetes, and creatinine level >2.5 mg/dL at discharge. The event rates according to number of risk predictors were similar in the derivation and the validation sets for all outcomes. In the validation cohort, rates for all-cause readmission and combined readmission or death were 26% and 31% in patients with no risk predictors, 48% and 54% in patients with 1 or 2 risk predictors, and 59% and 65% in patients with 3 or all risk predictors.

Conclusions: Few patient and clinical factors predict readmission within 6 months after discharge in elderly patients with heart failure. Although we were unable to identify a group of patients at very low risk, a group of high-risk patients were identified for whom resource-intensive interventions designed to improve outcomes may be justified.