Reducing potentially avoidable hospitalizations of nursing home residents: results of a pilot quality improvement project.

OBJECTIVES Hospitalizations expose nursing home (NH) residents to disruptions in care, iatrogenic events and related morbidity, and result in excess health care costs. Research has shown that a substantial proportion of these hospitalizations may be avoidable and that reducing such hospitalizations could save Medicare dollars that could be re-invested in improving the quality of care in US NHs. The objective of this project was to pilot test tools and strategies designed to assist NH professionals in reducing potentially avoidable hospitalizations. DESIGN Six-month prospective quality improvement initiative conducted by the Georgia Medical Care Foundation, the Medicare Quality Improvement Organization (QIO) for Georgia. Participating NHs were provided with communication and clinical practice tools and strategies designed to assist in reducing potentially avoidable hospitalizations, and on-site and telephonic support by an advance practice nurse. A retrospective review of acute care transfers was conducted by facility staff. Outcome data were compared to measures collected retrospectively over a 15-month baseline period. SETTING Three NHs in Georgia selected based on high rates of hospitalization that volunteered to participate. MEASUREMENTS Use of the tools and strategies were monitored every 2 weeks during the intervention with on-site visits by the advance practice nurse. Baseline data on hospitalization rates were determined using the Minimum Data Set (MDS), and hospitalizations were rated by an expert panel as potentially avoidable using a structured implicit record review process similar to that used in a previous study of the appropriateness of hospitalizations of NH residents. All hospitalizations during the 6-month intervention were ascertained, and all hospitalizations of residents whose hospital stay was reimbursed by Medicare were reviewed by the expert panel to determine the proportion that was potentially avoidable. RESULTS Although NH staff viewed the tools favorably, their use of them in the 3 facilities varied and none of the facilities fully implemented all of the tools. Despite only partial implementation, the quality improvement initiative was associated with a 50% reduction in the overall rate of hospitalizations during the 6-month intervention period compared to baseline. The proportion of hospitalizations rated as potentially avoidable was also reduced by 36%-from 77% at baseline to 49% during the intervention. CONCLUSION The quality improvement strategies and tools tested in this pilot project show promise for assisting NHs in reducing potentially avoidable hospitalizations. The results must be interpreted cautiously because this was not a controlled study, and was conducted in only 3 highly selected NHs. Refinement of the tools and implementation strategies and testing in a larger and more diverse sample of NHs is under way.