Lack of association between patients' measured burden of disease and risk for hospital readmission.

Identifying patients at increased risk for hospital readmission is important for clinicians, health policy-makers, hospital administrators, and researchers. We used a retrospective case-control design to compare the clinimetric properties of five validated indices that measure a patient's disease burden. The study was conducted on a random sample of patients discharged from the general medicine service at the Durham Department of Veterans Affairs Medical Center. Trained observers (two research assistants, one nurse, and two physicians) blinded to readmission status abstracted the required data elements from the medical record for three indices (Charlson, Kaplan-Feinstein, Index of Coexistent Disease). The hospital's computer provided data elements for two indices (Smith, adapted Charlson). Indices varied in the time required to complete, the ability to capture individual heterogeneity, and inter-observer variability. However, none of the indices discriminated among patients who did and those who did not have 6-month hospital readmissions. Factors other than summary scores derived from these indices should be used to identify patients at high risk for readmission.