The 131st Annual Meeting (November 15-19, 2003) of APHA

Abstract #57741

Aimée R. Dechter, PhD, Hyunjoon Park, MA, and Huey-Chi Chang, MA. Department of Sociology/Center for Demography and Ecology, University of Wisconsin-Madison, 4412 Social Science Building, 1180 Observatory Drive, Madison, WI 53706

Causes of racial and socioeconomic differentials in health may be due to disparities in the availability and quality of community resources and services, as well as biological, behavioral, environmental, and social factors. We examine the role of selected resources, services, and mean population behavioral characteristics in these health differentials. We estimate a multi-level model of age-standardized sex-specific county-level mortality rates. Explanatory variables include resources (e.g., the number of primary care physicians and hospitals per capita) and behaviors (e.g., prevalence of smoking and obesity) that may be disproportionately distributed across counties on the basis of socioeconomic status and racial composition. We also include a set of services, such as primary care programs (e.g., community health centers and federally qualified health center look-alikes), that are directed at these populations and may mediate their health disadvantage. We compile data for all counties in the U.S. Few nationally representative studies have included specific factors that account for socioeconomic and racial differentials. Most focus solely on population characteristics (e.g., average income in county, proportion unemployed, out-of-wedlock childbearing rate, etc.) and do not identify mechanisms through which these population characteristics operate to affect health and mortality. Moreover, these studies have not explored mediating factors, whereas our study focuses on specific services and means-tested state support which have implications for policy and intervention.

Learning Objectives:

• Participants will learn a statistical methodology for studying the impact of factors at multiple levels of observation (e.g., individual, community, county, state, etc.) on outcomes at the lowest level of observation (e.g., the individual). In this study, we estimate the impact of state and county level health resources and other state and county characteristics on county level mortality rates. These multi-level models can be used in public health research to evaluate the impact of specific programs, population characteristics and behaviors, resources and services available at various levels of organization and geography on one or more levels of observation. This statistical technique will allow participants to study one or more determinants of health/mortality while taking into account a broader range of factors and contexts at the individual, program and community level.
• Participants will be able to describe sources of variation in county level mortality and in particular, variation in socioeconomic and racial differentials in county-level mortality. Participants will be able to assess the importance of mean population behavior and characteristics, and county and state level health related resources for socioeconomic and racial differentials in mortality.
• Participants will be able to identify those services and resources that appear to mediate the mortality disadvantage of low income populations and African Americans.

Presenting author's disclosure statement:
I do not have any significant financial interest/arrangement or affiliation with any organization/institution whose products or services are being discussed in this session.