Use of GAMs to assess effects of air pollution on human health: Our ACAPS experience

The generalized additive model (GAM) is a useful analytical tool to assess the short-term health effects of air pollution using time series data. The modeling endeavor requires various decisions regarding the fitting of the GAM. We present a sensitivity analysis to assess the robustness of the risk estimate of the current day PM10 levels on daily cardio-pulmonary hospital admissions that was seen in the Alleghany County Air Pollution Study (ACAPS). The sensitivity analysis includes different lag models for the PM10, different composite measures that combine PM10 levels from multiple monitoring sites, different smoothing functions, and degrees of smoothing on weather and time covariates.
Results from our sensitivity analysis indicate that current day PM10 level is related to an increase in daily cardio-pulmonary hospital admissions in a robust way

Department of Biostatistics, Graduate School of Public Health, University of Pittsburgh, Pennsylvania 15261, USA.