Abstract Background and Significance
1 Introduction

Mortality rates among the oldest old population have dropped significantly in developed countries during the last half of this century [27]. Our understanding of declining mortality rates and variations in patterns of decline are based on, for the most part, death registration and census data. However, there are questions about the quality and consistency of these data, especially the quality of age-specific data [25]. Data collected for the administration of social support and health care programs for the elderly offer a new and potentially significant source of information on mortality at older ages. In this paper we evaluate the consistency of age-specific mortality patterns for older Americans using the Health Care Financing Administration’s (HCFA’s) Medicare Enrollment Data Base (EDB), a data set that includes over 30 million records. We evaluate the consistency of cohort and period age-specific mortality patterns across regions of the United States by sex and race.

In the following section we review the declines in mortality at older ages throughout the developed world with special attention to patterns in the United States, thus setting the context for the importance of this data set. We discuss the black mortality crossover and the various interpretations and implications of this pattern. We then describe the Medicare EDB and discuss previous research using it as well as evaluations of it for further research. The methods for estimating mortality rates and the strategy for evaluating consistency are then presented. Lastly, we present and evaluate the cohort mortality patterns, followed by the period mortality patterns.

Abstract Background and Significance

logo70.gif (2450 bytes)

Evaluation of U.S. Mortality Patterns at Old Ages
Using the Medicare Enrollment Data Base.

Allan M. Parnell and Cynthia R. Owens
1999 - 2000 Max-Planck-Gesellschaft ISSN 1435-9871