Predictors of Hospital Length of Stay in University Renal Transplant Programs

摘要

Two hundred thousand people have chronic renal disease and 11,000 will receive a renal transplant this year. Renal transplants cost the national medical care system a billion dollars per year and the length of the hospital stay is the best predictor of the total cost of each transplant. The University HealthSystem Consortium (UHC) recognized this expense and initiated a benchmarking study in an effort to identify the best practices. This study used the UHC raw data in an attempt to identify the variables which contributed the most unique variation to length of stay (LOS). If one could identify the variables which contribute to the LOS, the transplant center could concentrate on those variables and bring down LOS while maintaining the quality of care. In the capitated environment of managed care, the length of hospital stay can mean the difference between making or losing money. A hierarchical multivariate regression analysis was performed and the variables prior transplant experience, total cold ischemic time, routine ICU admission and transplant center were significant at p < .05, F-29.07. The transplant center variable contained six discrete variables designating specific centers. The four variables accounted for nearly 25 percent of the unique variation in LOS and the largest portion resulted from the transplant center variable, implying the there may be regional variation. This sample reflects a very strong correlation between total cold ischemic time and LOS which supports the argument in favor of lowering the total cold ischemic time. Further research needs to be done in regional variation and the effects of clinical pathways.