| Province-wide use of the Minimum Data Set 2.0 allows for the development of placement algorithms based on standardized continuous assessment to balance need for continuing care with the least-restrictive environment. Further knowledge of the predictors of care serves to improve these placement algorithms. Due to increasing regionalization of health care services in Ontario and diversity of ways services interact, analytic techniques are required that embrace the contextual nature of health services. The purpose of the current evaluation was two-fold: (1) to explore the degree to which ratings of a patients' health are dependent on the context in which continuing care takes place (i.e. variation in scores attributable to hospital or region); and (2) understand the influence of both symptoms at admission and throughout hospitalization on subsequent need for continuing care. This evaluation examined 24 231 standardized quarterly health symptoms nested within 15 904 patients over age 50 that were admitted and discharged from Ontario's 124 Complex Continuing Care facilities between April of 2007 and March of2009. Symptoms associated with the need for continuing care (activities of daily living, cognitive impairment, frailty, and aggression) and resource utilization were employed as dependent variables in multilevel modeling analyses. With respect to the first research question, as much as one-third of the variance in patient symptoms and resource utilization were associated with the hospital attended (i.e. occurred between-hospitals rather than solely within). This variation was only minimally accounted for by differences in the average age, sex, and length of stays between hospitals. With respect to the second research question, individuals referred from acute care and private homes were similar in symptom profiles but dissimilar in resource-intensity, while individuals from long-term care homes (LTCH) displayed higher levels of cognitive impairment and aggression. Over time, patients from acute care showed greater declines in symptoms, frailty, and resource utilization than other groups. Individuals discharged to LTCH were rated highest in queried symptoms (activities of daily living, cognitive impairment, aggression) and frailty, while those discharged to private homes were higher in resource intensity. Individuals discharged to acute care also experienced steeper declines in resource utilization on average than those discharged to acute care facilities. Future focused research into predictors of between hospital variability in outcomes and the surprising resource-intensity findings for those discharged to health care are suggested. |
