While there is great concern about the potential impact of aging populations on health care systems in the developed world, evidence from recent decades has shown at best a weak relationship between population aging and health expenditures at the aggregate level. This thesis explores the literature that frames the relationship between age and health care utilization in the context of reduced mortality and shorter periods of morbidity at the end of life. We add to this literature with an empirical study of individual health expenditures of the British Columbia senior population in the years 1991-2001 in the categories of hospital services, continuing care,
doctor billings, and pharmaceutical prescriptions. Expenditures for decedent and survivors of the same age are compared and are fitted to a model using age and time-to-death as explanatory factors. The partial derivative of the model with respect to age is analyzed for empirical
estimates of the effect of age after controlling for time-to-death. Results show that decedent costs rose over the study period while costs for survivors fell, particularly in continuing care, so
that the relative cost of dying increased. The effect of age, after controlling for time to death, was muted or negative for hospitals, doctors, and drugs, but strongly positive for continuing care and, as a result, for all services combined. Overall, these results suggest that age is not a ‘red
herring’, as some researchers have suggested, with respect to forecasting future demands on health systems. While future reductions in mortality and morbidity could mitigate pressures on hospitals, aging populations will put increased pressure on long-term residential care and other
forms of social care.
Identifer | oai:union.ndltd.org:TORONTO/oai:tspace.library.utoronto.ca:1807/26379 |
Date | 23 February 2011 |
Creators | Payne, Greg Jason |
Contributors | Coyte, Peter C. |
Source Sets | University of Toronto |
Language | en_ca |
Detected Language | English |
Type | Thesis |
Page generated in 0.002 seconds