Health and respirable particulate (PM10) air pollution: A causal or statistical association?

Abstract

Numerous studies have reported weak but statistically significant acute health effects of particulate air pollution. The associations are observed at levels below the current U.S. standard of 150 μg/m3 (24 hr). Health effects include acute increased mortality from cardiopulmonary conditions and acute morbidity such as hospital admissions for related diseases. We reviewed recent epidemiology studies to evaluate whether criteria for causality are met, and we conclude that they are not. The weak associations are as likely to be due to confounding by weather, copollutants, or exposure misclassification as by ambient particulate matter (PM). The results from the same metropolitan areas are inconsistent, and PM explains such a small amount of the variability in mortality/morbidity that the association has little practical significance. Finally, experimental chamber studies of susceptible individuals exposed to PM concentrations well above 150 μg/m3 provide no evidence to support the morbidity/mortality findings. None of the criteria for establishing causality of the PM/mortality hypothesis are clearly met at ambient concentrations common in many U.S. cities.