A study of the epidemiological coincidences of selected causes of death and toxic chemical releases in Texas counties

Abstract

Environmental justice proponents have argued that demographic factors unevenly affect the location of manufacturing and wastes facilities, and thus, differences in exposure risks and outcomes. In this research, statistical relationships among demographic factors, toxic chemical wastes released by Texas manufacturing facilities who participated in the Toxic Release Inventory (TRI) in 1990, and cancer-related death rates were examined at the county level of analysis. Chemical-waste data were obtained from the Environmental Protection Agency, mortality data from the National Center for Health Statistics, and 1990 demographic data from the US Bureau of the Census. Age, gender, and ethnicity specific death rates of Texas counties were calculated and standardized to the state population for each of three causes of death: all cancer deaths, lung-related cancer deaths, and all cancer deaths less lung-related cancer deaths. In addition to the volume of toxic chemical wastes released and the number of manufacturing facilities in a county, cancer-related death rates were regressed against the following county-level variables: percent Black population, percent Hispanic population, county population density, percent of the dependent population (population under the age of 14 years and 65 years of age and older), percent female population in the child bearing ages of 15 to 44, per capita income, percent employed in manufacturing industries, and volume of TRI releases. In the cancer-related regression models, population density and proportion of dependent population were significant. Although some support was found for racially based correlations with the location of TRI bivariate correlation analysis, very little evidence was found among demographic and industrial factors that explained differences in cancer mortality rates in the multivariate analyses. Implications of the findings for future research are discussed.