Career Development Award: Center for Appalachian Research in Environmental Sciences: Using Commercially Available Residential Histories for Exposure Assessment in Environmental Epidemiologic Analysis of Cancer Risk

While there are known health risks associated with an individual’s occupational exposure to agents generated by coal mining activities (e.g., pneumoconiosis), researchers over the past several years have also explored potential environmental exposures relevant to population health. Multiple ecological and cross-sectional studies have demonstrated, for example, that the risk of lung cancer is elevated among populations in coal mining regions, even after accounting for rates of cigarette smoking, the foremost cause of this disease (1-3). Rigorous individual-level studies of lung cancer risk and environmental exposures derived from coal mining activities have not been reported, however, despite the population-based evidence. Case-control study designs in particular could provide stronger evidence of a relationship, if one truly exists. Latency is a major challenge for case-control studies of environmental exposures and cancer. Due to the decades-long latency of lung cancer, estimating exposures occurring over decades is sometimes necessary to accurately calculate risk. Residential histories with sufficient spatial and temporal detail can provide important data for long-term exposure assessment, but these are often hindered by participants’ difficulty with recall of particular details, especially full street addresses including house or building number. Without such precise locational data, it is much more difficult to accurately assess potential exposures among research participants, since most geocoding algorithms will assign individuals to locations, such as ZIP code or county centroids, that might be miles from their actual residence, especially in rural regions. The use of address data from LexisNexis, Inc., a commercial provider of data services, has previously demonstrated promise for robust analysis of cancer risk in relation to residential history. Jacquez and colleagues conducted an analysis of bladder cancer in coordination with the Michigan Cancer Registry, and Hurley and colleagues recently completed a similar analysis of breast cancer using data from the California Teachers Study (CTS) (4, 5). Both studies demonstrated that address data from LexisNexis provided information that further enhanced the residential history survey data already collected, but also presented some notable challenges, including duplicate records and invalid dates of residence. Automated sources such as these have inherent flaws, like surveys, but appear to contribute key information for exposure assessment in environmental epidemiologic studies of cancer. Still, these innovative techniques have not yet been employed in Central Appalachia, a rural region with some of the highest rates of cancer in the U.S. The ultimate goal of our research is to understand the effect of environmental exposures on risk for cancer in Kentucky and Central Appalachia. “Mountaintop removal” mining has been practiced in the region since the 1990s, and is concerning in terms of potential environmental exposures because it requires displacement of large quantities of rock and soil “overburden” to reach thin seams of coal. We will examine the risk of lung cancer in relation to duration of residence in proximity to such mining sites by leveraging survey data from a previous case-control study of lung cancer in the Central Appalachian region and augmenting it with LexisNexis data. Additionally, we will conduct a case-control study of cancers potentially related to Superfund sites using Kentucky Cancer Registry (KCR) data linked to LexisNexis data. We hypothesize that personyears of residential proximity to both sites is positively associated with cancer risk. Our Specific Aims are to: 1. Assess the utility of LexisNexis address data in augmenting residential history data from an existing casecontrol study of lung cancer in Central Appalachia. This aim is comprised of two sub-aims: one addressing potential biases when using LexisNexis data, and another relating to the effect on analytical findings. A. Examine characteristics of cancer cases and controls potentially associated with robust address data available from LexisNexis. It is likely that individuals from some demographic and socioeconomic groups have more of their residential history information captured by LexisNexis’s system. This subaim will examine the availability of previous addresses to determine if such biases exist, and subsequently inform the following sub-aim and Aim 2. B. Conduct parallel analyses of lung cancer risk in relation to lifetime residential proximity to mountaintop removal mining using (i) only the existing survey data, (ii) only LexisNexis data, and (iii) a combined data set. Comparison of these results will highlight the effect on measures of association when using the LexisNexis data to supplement survey data on residential history. 2. Conduct a case-control study of selected Superfund sites and the risk of other cancers, using only population-based cancer registry data and LexisNexis data. This analysis will examine use of LexisNexis data as the sole source of residential history information for cases of selected cancers and controls made up of other cancer cases in the cancer registry. Less non-residential information will be available in this analysis, but since it is registry-based there will be many more cases and controls than in Aim 1. Accomplishing these Specific Aims will inform design and implementation of future environmental epidemiologic studies, identify geographic areas of concern in Kentucky that may be targeted for biomarkers research and public health intervention, and support the career development of the Principal Investigator.