Established Populations for Epidemiologic Studies of the Elderly

In recognition of the rapidly growing size of the elderly population, the National Institute on Aging initiated a project in 1980 to investigate health risks that affect adults as they age. This initiative led to the Established Populations for Epidemiologic Studies of the Elderly (EPESE) project. The specific goals of the EPESE project were to provide estimates of the magnitude of health problems of community-dwelling older adults, to examine risk factors for these health problems for the purpose of informing appropriate interventions to prevent disability and postpone mortality, and to provide basic information on the social, mental, behavioral, and economic conditions of older adults and the relation of these conditions to health and health changes over time. Initially, three studies were funded: the East Boston Senior Health Project, the Iowa 65+ Rural Health Study, and the Yale Health and Aging Project. In 1985, the Piedmont Health Survey of the Elderly was added as a fourth study.

The EPESE studies were planned to have a number of important design features in common to facilitate cross-study comparisons and analyses. These design features served two overriding goals: (1) to obtain reliable prevalence estimates of important health problems and (2) to provide unbiased estimates of the associations between risk factors and major health outcomes. To achieve these goals, each of the four studies was designed as a population-based longitudinal study with standardized assessment of major risk factors and outcomes, relying on well-established and previously validated measures wherever possible. The cohorts in each study were selected to be representative of the target population, defined on the basis of being age 65 years and older, with sufficient heterogeneity in target populations across the projects to achieve a reasonably adequate representation of the entire elderly population in the United States during the early 1980s. It should be recognized that certain subpopulations that have grown substantially in size and proportion over the past 20 years, such as older Latinos and Asian Americans, were not represented in the EPESE studies.

The East Boston Senior Health Project focused on a working-class population of predominantly Italian American background who lived in the geographically defined community of East Boston, an island located in Boston Harbor. All residents age 65 years and older were eligible for the study. The Iowa 65+ Rural Health Study targeted all noninstitutionalized older adults living in two rural counties in east-central Iowa (Iowa and Washington counties). The Yale Health and Aging Project conducted a random sample of all older adults living in the city of New Haven, Connecticut. The random sample was stratified by sex and housing type (public housing, age-restricted housing, or community). The Piedmont Health Survey of the Elderly conducted a stratified random sample of the older population of five counties in the north-central Piedmont area of North Carolina, employing a sampling design to obtain a cohort that would contain approximately 55% African Americans.

The longitudinal design of each of the projects was the second common design feature. This feature served to provide optimal information on risk factors that are prospectively linked to age-related changes in health. Each study conducted a total of seven interviews at yearly intervals. Face-to-face interviews were conducted at baseline and repeated at the third and sixth anniversaries of the baseline interviews. [Page 199]Telephone interviews were conducted during intervening years, corresponding with the first, second, fourth and fifth anniversaries of baseline. Uniformly high participation rates were achieved at all sites, with baseline participation rates ranging from 80% to 85% and follow-up participation rates generally exceeding 95% of survivors.

...