Nursing home case-mix classification and residents suffering from cognitive impairment: RUG-II and cognition in the Texas case-mix data base

Abstract

Case-mix reimbursement systems for nursing home care are now in use or under consideration in over 20 states. Though many in the long-term care field still debate the relative merits of the various systems, those systems based on the Resource Utilization Groups-II (RUG-II) system used in New York seem to be increasingly popular. However, one of the more serious questions raised concerning RUG-like systems is whether they fully acknowledge the role played by cognitive deficits in determining nursing home residents' care needs. The results of this research indicate: 1) that resource provision and cognitive status are correlated, 2) that residents with higher levels of cognitive impairment appear most frequently in RUG-II categories that have higher case-mix weights, 3) that simply adding some measure of cognitive function does not significantly increase the power of these models, and 4) that the RUG-II model predicts the level of care resources provided to the cognitively intact much better than it predicts those provided to residents suffering from relatively severe disorientation. These results imply that such classification systems may create disincentives for the admission of the most cognitively impaired residents and that the care of these individuals, more than others, may be dependent on facility-to-facility variation