The relationship between Elder Risk Assessment Index score and 30-day readmission from the nursing home.

Objective: Preventable early readmission to the hospital is expensive, and identification of patients at risk is an important task for health care providers. The objective of this study was to determine the relationship between a high score on the Elder Risk Assessment (ERA) Index and 30-day readmission to the hospital in older patients discharged to a nursing home.

Patients: Patients aged > 60 years residing in the community on January 1, 2005 and subsequently admitted to a local nursing home following hospitalization were included. The cohort was selected from all patients in a primary care internal medicine practice in Rochester, MN.

Methods: This was a retrospective cohort study that used an electronically archived administrative risk index, the ERA Index, which was derived from demographic and clinical factors. The primary outcome was hospital readmission within 30 days following initial admission to a nursing home. The predictor variable was the ERA Index score. Univariate association between the total ERA Index score and individual components of the ERA Index and 30-day rehospitalization were determined. The ERA Index score cutoff with optimal sensitivity and specificity for hospital readmission was also identified.

Results: Of 12 650 patients in the population, 800 were admitted to a facility between 2005 and 2007. Thirty-day readmission was not higher in the group with the highest ERA Index score (top quartile), with a relative risk of 1.72 (95% confidence interval [CI], 0.93-3.56) compared with the lowest-scoring group. The second- and third-highest quartiles were significantly associated with higher 30-day readmission. The individual component of the ERA Index that had the strongest association with early readmission was dementia, with an odds ratio of 2.69 (95% CI, 1.71-4.23). A cutoff score of 5 on the ERA Index resulted in a sensitivity of 0.81 and a specificity of 0.34 with an area under the curve of 0.55.

Discussion: Those with the highest ERA Index score, the top quartile, were not at risk for early hospital readmission. The ERA Index does not predict readmissions from the nursing home to the hospital. There is a need to develop a unique index to predict rehospitalizations in nursing home residents.