Predictors of Home Care Costs among Persons with Dementia, ALS and MS in Ontario
MetadataShow full item record
Purpose: The purpose of this project was to look at the costs of individuals with Alzheimer’s disease and related dementias (ADRD), ALS, and MS in long stay home care in Ontario, Canada. The specific goals were to produce estimates of costs for these individuals, as well as identify clinical and personal characteristics associated with these costs. This project also tested the effectiveness of the Resource Utilization Group for home care case-mix system for use in these special populations. Methods: This project was conducted using a secondary analysis of assessment data from the Canadian Staff Time Resource Intensity Verification Project, a 13-week study of home care costs (N=435 141). The project was guided by the Andersen and Newman (1973) framework for healthcare resource utilization. Descriptive characteristics and mean costs were produced using bivariate frequency and means procedures for each of three conditions. Predictors of costs were identified for each of the three neurological conditions through multivariate regression analysis conducted separately for each condition. In total 41 independent variables were included into the bivariate and multivariate analyses. The dependent variable was the total weekly formal and informal home care costs across all multivariate analyses. Results: In total, ADRD, ALS, or MS diagnoses were present in 16% of the assessments. The mean costs for the three conditions combined were $594.81. The mean costs for ADRD, ALS, and MS were $593.32, $898.41, and $574.92, respectively. Characteristics that were predictive of cost across all conditions included the Resource Utilization Group for home care case-mix system, ADL functionality, IADL functionality, cognitive performance, unsteady gait, stair use, difficulty swallowing, respiratory challenges, and bowel incontinence. The Resource Utilization Group for home care case-mix system had the highest level of explained variance of any single item tested in this project across all conditions. However, other clinical characteristics also contributed substantial levels of explained variance to the models for each of the three conditions. Conclusions: The findings from this project suggest that although diagnosis of ADRD, ALS, and/or MS can describe cost, clinical characteristics are the most important predictors of costs for individuals with these conditions. In addition, the Resource Utilization Group for home care case-mix system can adequately predict costs of individuals with these conditions. The addition of some clinical characteristics would likely improve the predictive abilities of the Resource Utilization Group for home care case-mix system.