Authors: Jay Christian*, University of Kentucky, Courtney Walker, University of Kentucky, Bin Huang, University of Kentucky
Topics: Medical and Health Geography, Spatial Analysis & Modeling, Geographic Information Science and Systems
Keywords: cancer, environmental epidemiology, coal, Appalachia, residential history
Session Type: Paper
Start / End Time: 8:00 AM / 9:40 AM
Room: Madison A, Marriott, Mezzanine Level
Presentation File: No File Uploaded
Many published epidemiologic studies have demonstrated associations between coal mining activity and chronic disease risk among residents in the Central Appalachian region, but the great majority have used ecological or cross-sectional study designs, which provide relatively weak evidence for causation. Case-control studies can provide stronger evidence, but should ideally include residential histories from all participants to enable long-term assessment of environmental exposures over time, rather than relying solely on residential location at diagnosis. Obtaining accurate address history information through surveys, however, requires substantial time and resources for even a few hundred participants, especially in distressed rural communities like many in Appalachia. We are currently investigating the utility of using address data procured from LexisNexis, a commercial credit reporting company, to enhance the analysis of data gathered through an existing case-control study of lung cancer and exposure to trace elements associated with coal deposits in Appalachian Kentucky. In this presentation, we discuss challenges of working with LexisNexis address data and the potential for bias in regard to precision and accuracy. Furthermore, we examine concordance with previous addresses reported by participants (n=520) of the existing case-control study in an in-person survey. Our results demonstrate that LexisNexis data can provide valuable information on the residential histories of many research participants at very low cost, and with comparable quality compared to in-person surveys, though biases related to socioeconomic status could complicate some epidemiologic analyses.