The effect of nursing home quality on patient outcome.

This historical cohort study investigated the relationship between nursing home quality and the group outcomes of mortality, rehospitalization, and discharge. Multiple logistic regression was used to determine if nursing home quality indices increased the prediction of these outcomes when patient severity of illness and case-mix differences were in the model. Three hundred ninety veterans discharged to 11 nursing homes were followed for 6 months. Nursing home quality was assessed using indices from the Multiphasic Environmental Assessment Procedure and the Rush-Medicus Methodology for Monitoring Quality of Nursing Care. An increased likelihood of death was associated with the diagnoses of cancer and heart disease and with being rehospitalized. Four nursing home quality indices significantly improved the prediction of mortality (RN hours, nursing process, security, and mean quality). Rehospitalization was associated with the patient factors of heart disease, hypertension, race, and level of care and with size of the nursing home. Discharge from the facility was inversely associated with the diagnosis of cancer and directly related to the index of nursing process. The results support the notion that group outcomes may be related to nursing home quality and suggest the need for further studies to investigate the specific elements of nursing home quality which relate to improved outcome.