Kindig DA, Wang H, Remington PL; Academy for Health Services Research and Health Policy. Meeting.

Abstr Acad Health Serv Res Health Policy Meet. 2000; 17: UNKNOWN.

Presented by: David A. Kindig, M.D., Ph.D., Department of Preventive Medicine, University of Wisconsin/Madison School of Medicine, WARF Building, 610 Walnut Street, Madison, WI 53705-2397. Tel: 608-263-6294. Fax: 608-262-6404. E-mail: dakindig@facstaff.wisc.edu

Research Objective: Previous research has identified substantial variation in age-adjusted mortality rates and the factors associated with these differences. As a part of an effort to develop life-stage specific measures of health adjusted life expectancy, we determined the variation in age-specific mortality rates across the nation as well as the most important independent variables associated with each life-stage outcome.Study Design: Multiple-group comparison of 801 Health Service Areas (HSAs) was included in this study. The dependent variables were age-adjusted and age-specific mortality rates; with life-stage age groups of <1, 1-14, 15-24, 25-64, 65-74, and 75+. The independent variables used included 8 socio-economic variables, 4 medical resource variables, and 3 risk behavior variables from the Area Resource File and the 1992-1994 Behavioral Risk Factor Surveillance System. Weighted least square regression with the cluster-robust variance estimation was used for the data analysis in order to take into account for the unequal variance of the dependent variables and correlated residuals within the clusters (defined as states).Population Studied: The population of the United States in 1989-1991.Principal Findings: Geographic variation in mortality rate was the highest in the 1-14 age group (CV 0.473) compared to all age mortality (CV 0.107), with other high variation in age 15-24 (CV 0.413) and infant mortality (CV 0.278). Mapping the rates showed higher mortality in the south and lower Midwest, particularly for infant, 25-64, and 65+s'. Regression analysis explained from 31% of the variation in age 1-14 to 82% for age 25-64. Several variables were consistently associated with higher mortality in each life stage (such as race, education and divorce), but additional associations were seen at later life stages (such as health resources, and risk behavioral indicators).Conclusions: Geographic variations in age-adjusted mortality rates differ by life stage. Several independent variables are associated with the mortality at every life stage, with additional factors becoming significant at later life stages. Further comparative analysis with individual data should be carried out in order to approach causal understanding.Implications for Policy, Delivery or Practice: Understanding the reasons for the variation in health outcomes is necessary for overall improvement as well as reduction in inequality across individuals and populations. Life stage analysis of mortality (and of HRQoL as well) will be useful in guiding public and private policy decisions with these goals.Primary Funding Source: Milbank Memorial Fund

Publication Types:

• Meeting Abstracts

Keywords:

• Aged
• Catchment Area (Health)
• Continental Population Groups
• Humans
• Infant
• Infant Mortality
• Life Expectancy
• Neoplasm Staging
• Population
• Population Dynamics
• Socioeconomic Factors
• United States
• classification
• mortality
• statistics & numerical data
• hsrmtgs

Other ID:

• GWHSR0000668

UI: 102272342

From Meeting Abstracts