Forecasting health status changes in an aging U.S. population: Assessment of the current status and some proposals

Much interest has been recently focused on forecasting population health changes such as population aging and life expectancy increases at advanced ages, qualitative changes in health and survival, changes in federal reimbursement systems, and the need to forecast for long-range capital investment in specialized health facilities. Recent efforts at forecasting health and morbidity changes in the population are reviewed in this paper and the recent lack of methodological innovation in health forecasting strategies is examined. Two forecasting models-stochastic compartmental systems and mixed multivariate continuous state-discrete state processes-are described. We show how these models can be applied to health forecasting using existing data, how they can be used in a complementary fashion to improve health projections made from any one type of data, and how the mixed multivariate stochastic process model can be used to evaluate assumptions made in the compartment modeling approach. © 1987 D. Reidel Publishing Company.

Duke Scholars

Author Kenneth G. Manton Trinity College of Arts & Sciences