Abstracts: ISEE 22nd Annual Conference, Seoul, Korea, 28 August-1 September 2010: Air Pollution - Exposure Characterization and Health Effects
There has been much interest in the association between ambient air pollution and birth outcomes, and different methods have been applied. Proximity to traffic is a proxy for traffic-related exposures and can be estimated using geographic information systems. The aim of this study was to assess birth outcomes in relation to the distance of the mother's home to the nearest road and the number of roads surrounding it.
We investigated the association between traffic exposure and gestational age, birth weight, and birth length of 970 mothers in Logan, Queensland. We used multiple regression models and controlled for mother's age, smoking, parity, and education. We used a spline to control for trends and seasonal patterns. We modeled the effect of distance to road using a nonlinear spine with 3 degrees of freedom to capture the expected exponential-like decay in risk. For the number of roads surrounding the mother's home (main roads, freeways, and highways), we used a parametric approach by fitting either a quadratic or linear model.
There was some evidence of shorter gestations for mothers living closer to a major road and a marginal increase in birth weight and birth length with increasing proximity to a highway. We also found evidence of shorter gestations with greater density of main roads and freeways. Only the association between a shorter gestation and greater road density was statistically significant. Living 2 km from a main road increased gestation by an average of 0.4 weeks compared with being adjacent to a main road. Modeling the effects of wind direction had little effect on the results.
Exposure to traffic had only a marginal negative influence on gestation length in this study. Future studies should include both temporal and spatial variation in exposure to air pollution when assessing its effects on birth outcomes.