Space-time Statistics for Environmental Epidemiology
Abstracts: ISEE 20th Annual Conference, Pasadena, California, October 12–16, 2008: Symposium Abstracts
Abstracts published in Epidemiology have been reviewed by the organizations of Epidemiology. Affliate Societies at whose meetings the abstracts have been accepted for presentation. These abstracts have not undergone review by the Editorial Board of Epidemiology.
A recurrent and challenging task in environmental epidemiology is the analysis and synthesis of data collected at different spatial scales and times. The integration of (geo)statistics into Geographical Information Systems (GIS) has enabled the incorporation of multiple layers of information and space-time correlation in the mapping of contaminant concentrations. However, there remain substantial hurdles when assessing the lifetime exposure to contaminants for individuals that move frequently and for pollution levels that change in time (e.g. air pollution). To protect patient privacy publicly available data are often aggregated to a sufficient extent to prevent the disclosure or reconstruction of patient identity. The information available for human health studies is thus restricted to mortality rates within areas, such as ZIP codes or counties. Associations can then be investigated between these areal data and environmental, socio-economic or demographic covariates. Traditional analysis of health data fails to account for the fact that mortality rates tend to be spatially correlated and are less reliable when computed for sparsely populated areas. Another weakness is that the association between rate data and covariates is assumed stationary or spatially invariant. This assumption is likely unrealistic for large regions that display substantial geographic variations in demographic, social, economic, and environmental conditions. The goal of this symposium is to bring together international experts in the conceptual development and application of spatial and space/time analysis methods and to encourage their application for the joint analysis of exposure and health data. We have built on the expertise at Universities, Research and Development Companies and Agencies to put together a list with international expertise in spatial and space/time analysis. This symposium will offer a platform to foster interactions between modelers, software developers and health scientists that are critically needed to advance the field of space/time statistics in environmental epidemiology. The proposed format of this symposium is to have one session of two hours with four presenters, which should leave ample time for discussion.© 2008 Lippincott Williams & Wilkins, Inc.