A spatial-temporal regression model to predict daily outdoor residential PAH concentrations in an epidemiologic study in Fresno, CA

Background clic aromatic hydrocarbons (PAHs) are generated as a byproduct of combustion, and are associated with respiratory symptoms and increased risk of asthma attacks. ives ign daily, outdoor exposures to participants in the Fresno Asthmatic Children’s Environment Study (FACES) using land use regression models for the sum of 4-, 5- and 6-ring PAHs (PAH456). s ta were collected daily at the EPA Supersite in Fresno, CA from 10/2000 through 2/2007. From 2/2002 to 2/2003, intensive air pollution sampling was conducted at 83 homes of participants in the FACES study. These measurement data were combined with meteorological data, source data, and other spatial variables to form a land use regression model to assign daily exposure at all FACES homes for all years of the study (2001–2008). s del for daily, outdoor residential PAH456 concentrations accounted for 80% of the between-home variability and 18% of the within-home variability. Both temporal and spatial variables were significant in the model. Traffic characteristics and home heating fuel were the main spatial explanatory variables. sions e spatial and temporal distributions of PAHs vary on an intra-urban scale, the location of the child’s home within the urban setting plays an important role in the level of exposure that each child has to PAHs.