Distributed lag associations between respiratory illnesses and mortality with suspended particle concentration in Tula, a highly polluted industrial region in Central Mexico

Abstract

We aimed to evaluate the association between changes in airborne particulate matter concentration (PM) with changes in cases of mortality, acute respiratory infections (ARI) and asthma over 2004-2008 in an industrialized and polluted region in central Mexico. A generalized linear model with a Poisson distribution and a negative binomial analysis was used to evaluate the influence of PM and temperature on all-cause mortality (All-cause-M), cause-specific mortality (Cause-specific-M), ARI and asthma, using cubic spline functions and distributed lags of PM. Estimated changes in relative risk were calculated for an exposure corresponding to each increase of 10 μg/m(3) in PM level. Associations between PM and mortality and morbidity were statistically most consistent for total suspended particulate (TSP) than for particulate matter <10 μM aerodynamic diameter (PM10). The greatest effects in mortality were observed with a 3-week lag, and effects were greater for Cause-specific-M. We also found a displacement effect up to 4-week lag for Cause-specific-M and TSP. The greatest effects in morbidity were observed at 0-week lag, yet they were statistically marginal and were greater for asthma. We found a displacement effect at 4-5-6-week lag for asthma and TSP. All associations of mortality and morbidity, expressed as change in relative risk, were greater with PM10; however, all of them were statistically marginal. Increased respiratory morbidity and mortality is associated with weekly changes of PM air pollution in the region. A reduction in air pollutants from industrial sources would benefit life quality and health of the exposed population.