Population Models of Factors Affecting Health Trends

Duke University, Durham NC

Investigators

Linked publications & trials

Abstract

DESCRIPTION (provided by applicant): Declines in chronic disability from1982 to 1999 among the elderly raise issues of fundamental scientific and policy importance. Prof. Manton s program project will enhance our understanding of the underlying biomedical and physiological reasons for these declines. In this supplemental application, we propose rigorous investigation of the implications of these declines for Medicare, projecting the number of disabled and nondisabled elderly persons in the U.S., 2000 to 2080, by age, sex, race, education, income/assets, and marital status with the use of acute Medicare and long-term care services. These estimates and projections will facilitate a comprehensive assessment of the impact of the declines in chronic disability and their implications for the use of acute and long-term care services. We will examine three types of forecasting models: (1) static component methods, (2) Markov chain methods, and (3) grade of membership methods. Consideration will be given to uncertainty/confidence intervals for the forecasts, and changes in transition rates or similar parameters over time in a manner consistent with extrapolations of the ongoing declines in chronic disability. An issue of considerable importance that may not be evident in the 1982 to 1999 data, but which may be identified in the projection, is that small policy changes can have large, nonlinear effects on Medicare financing. This has been driven home by the 1997 BBA which extended Medicare insolvency from 2001 to 2029 in just four years. The primary data source will be the 1982-1999 National Long-Term Care Survey (linked to Medicare). The second major data source used will be the 1992-1998 Health and Retirement Study & Asset and Health Dynamics Among the Oldest-Old Study in order to examine disability changes for ages 51 to 64. Other data sets will be used as needed for specific estimates, such as changes in the length of nursing home stays (e.g., the 1985 and 1995-1997 National Nursing Home Study, the 1968-2000 national multiple cause of death microdata mortality files).