Modeling multiple environmental exposures and childhood leukemia risk

点击翻译按钮获取中文摘要

Incidence for many cancers is suspected to be associated with chemicals through environmental routes of exposure. Individuals are exposed to a large number of diverse environmental chemicals simultaneously and evaluation of multiple chemical exposures is important for identifying cancer risk factors. Increasingly, exposures are being measured for a large number of chemicals in epidemiologic studies to allow for a more comprehensive assessment of cancer risk factors in the exposome. However, most existing studies of environmental chemical exposures and cancer use a single-chemical approach that evaluates chemicals independently as risk factors. Traditional statistical methods used in existing studies are significantly challenged by the typically strong correlation observed among many environmental chemical exposures, as well as other environmental and socioeconomic variables. There is a need for development and assessment of statistical methods to model environmental cancer risk that consider a large number of diverse chemicals with different effects for different chemical classes. The specific aims of this research are 1) to develop more comprehensive exposure risk models and apply them to a case-control study of childhood leukemia in California that contains concentrations measured for a large number of diverse chemicals, 2) to account for uncertainty associated with imputing chemical concentrations below the limit of detection when estimating chemical mixture effect, and 3) to account for neighborhood socioeconomic status and residual confounding when estimating chemical exposure effects. The study of childhood leukemia will benefit from environmental chemical risk analysis because it is a cancer with an unclear etiology and established risk factors that account for only a small proportion of the total annual cases in the United States. The expected outcomes of this research will be 1) new statistical approaches to model environmental cancer risk that consider environmental exposures more comprehensively while also accounting for uncertainty related to missing data, and 2) an evaluation of the effects of exposure to chemicals from many chemical classes and an identification of the important chemicals for childhood leukemia. The significance of this research is two-fold. First, the development and evaluation of new statistical approaches to risk analysis that consider multiple diverse environmental exposures while accounting for uncertainty will advance the field of environmental epidemiology research. Second, this will be the first environmental risk analysis of a case-control study of childhood leukemia that estimates effects for different chemical classes using a large number of correlated chemical exposures while also adjusting for known demographic risk factors at the household and neighborhood level and accounting for uncertainty related to missing data. The methodological approaches developed in this work will be applicable to many other cancers associated with multiple environmental exposures of differing types.