Pulmonary function testing in epidemiological studies of particulate air pollution: Interpretative issues

Abstract

Epidemiological studies relying on pulmonary function tests (PET) to measure health effects from particulate matter (PM) air pollution have demonstrated varying results at ambient exposure levels, from no effect through mean PFT decrements of a few percent. The extent of PFT variability indicates the need for caution when interpreting the meaning of these small percentage changes. This review explores the sources and magnitude of PET variability, focusing on their impact on epidemiological studies of PM air pollution. Emphasis is placed on the potential for temperature, indoor air pollution exposures, aeroallergens, respiratory infections, and psychogenic factors to affect PFT results. Any differential distribution of sources of PFT variability over time or between exposed and control groups could bias or confound study results. These concerns mandate the need for rigorous epidemiological protocols when using PFTs in the study of PM air pollution, including the incorporation of independent control groups, an accounting for potential confounding factors, and an expanded role for experimental chamber exposures.