Time to death and health expenditure: an improved model for the impact of demographic change on health care costs

Background: Obtaining well-founded estimates of the effect of demographic change on future health expenditures is a pressing issue in all developed countries. Thus far, expenditure projections have examined the effect of age on health care costs, but fail to account for the influence of remaining life expectancy on costs. Objective: This paper seeks to create a more accurate projection model that considers the concentration of costs towards the end of life, and to compare this model with the more traditional approach that holds age- and sex-specific per capita expenditures constant. Methods: We used a longitudinal hospital dataset which followed 90 929 patients aged 65 and older from 1970 to death, to create an economic model of hospital costs based on patient age and time remaining to death. We then applied the model to England population projections to predict the effect of demographic changes on hospital expenditures from 2002 to 2026. Results: The decline in age-specific mortality rates over time postpones death to later ages, pushing back death-related costs. Accounting for this in expenditure projections gave a predicted annual growth rate of 0.40%-half of the rate predicted with a traditional method. Conclusions: Using richer data and more refined methods than have hitherto been employed, this study strongly confirms that the pressure of population increases and ageing demographic structure on hospital expenditures will be partially countered by the postponement of death-related hospital costs to later in life-a finding consistent with emerging epidemiological evidence, and heartening for policy makers and physicians alike.