Our results confirm previous findings of an elevated metarelative risk for multiple myeloma among firefighters. In addition, a probable association with non-Hodgkin lymphoma, prostate, and testicular cancer was demonstrated.

BackgroundWe previously reported an association between infant wheezing and residence < 100 m from stop-and-go bus and truck traffic. The use of a proximity model, however, may lead to exposure misclassification.ObjectiveResults obtained from a land use regression (LUR) model of exposure to truck and bus traffic are compared with those obtained with a proximity model. The estimates derived from the LUR model were then related to infant wheezing.MethodsWe derived a marker of diesel combustion—elemental carbon attributable to traffic sources (ECAT)—from ambient monitoring results of particulate matter with aerodynamic diameter < 2.5 μm. We developed a multiple regression model with ECAT as the outcome variable. Variables included in the model were locations of major roads, bus routes, truck traffic count, and elevation. Model parameter estimates were applied to estimate individual ECAT levels at infants’ homes.ResultsThe levels of estimated ECAT at the monitoring stations ranged from 0.20 to 1.02 μg/m3. A LUR model of exposure with a coefficient of determination (R2) of 0.75 was applied to infants’ homes. The mean (± SD) ambient exposure of ECAT for infants previously categorized as unexposed, exposed to stop-and-go traffic, or exposed to moving traffic was 0.32 ± 0.06, 0.42 ± 0.14, and 0.49 ± 0.14 μg/m3, respectively. Levels of ECAT from 0.30 to 0.90 μg/m3 were significantly associated with infant wheezing.ConclusionsThe LUR model resulted in a range of ECAT individually derived for all infants’ homes that may reduce the exposure misclassification that can arise from a proximity model.

Epidemiologic studies of air pollution require accurate exposure assessments at unmonitored locations in order to minimize exposure misclassification. One approach gaining considerable interest is the land-use regression (LUR) model. Generally, the LUR model has been utilized to characterize air pollution exposure and health effects for individuals residing within urban areas. The objective of this article is to briefly summarize the history and application of LUR models to date outlining similarities and differences of the variables included in the model, model development, and model validation. There were 6 studies available for a total of 12 LUR models. Our findings indicated that among these studies, the four primary classes of variables used were road type, traffic count, elevation, and land cover. Of these four, traffic count was generally the most important. The model R 2 explaining the variability in the exposure estimates for these LUR models ranged from .54 to .81. The number of air sampling sites generating the exposure estimates, however, was not correlated with the model R 2 suggesting that the locations of the sampling sites may be of greater importance than the total number of sites. The primary conclusion of this study is that LUR models are an important tool for integrating traffic and geographic information to characterize variability in exposures.Epidemiologic studies of the health effects of air pollution require accurate exposure assessments at unmonitored locations (e.g., subjects' place of residence) in order to minimize exposure misclassification. Methodologies employed to accomplish this task include, but are not limited to, spatial interpolation (e.g., kriging), proximity models, and dispersion modeling (e.g., California Line Source Dispersion model, CALINE) and these and others have been reviewed . Intraurban air pollution is characterized, however, by high spatial variability of pollutants with rapid decay from the source (Briggs et al., 1997(Briggs et al., , 2000. For example, nitrogen dioxide (NO 2 ) has been shown to have two-to threefold differences within 50 m or less (Hewitt, 1991), sulfur concentrations have been demonstrated to decrease by onehalf between 50 and 150 m from a highway (Reponen et al., 2003), and ultrafine particles have been shown to be elevated above background concentrations to approximately 300 m from highways (Zhu et al., 2002). These small-scale variations in pollutant concentrations are not identifiable using most interpolation techniques based upon monitoring density and spatial distribution of traffic sources (Brauer et al., 2003). Furthermore, proximity models have a likelihood of exposure misclassification due to the assumption of isotropic dispersion and the use of a categorical exposure designation (e.g., residence <100 m = exposed, residence >100 m = unexposed) . In order to address these limitations, land-use regression (LUR) models have been developed and utilized to model traffic pollutants including NO 2 and PM 2.5 (Briggs et al., 1997(B...