Background and objective: The impact of air pollution on morbidity and mortality is well established, especially for respiratory diseases. There are only few studies, however, that have evaluated the form of the exposure-response relationship and the role of individual effect modifiers. We studied the exposure-response relationship between daily PM10 and respiratory mortality in Italy, and considered the potential effect modifiers. Methods: A case-crossover analysis was carried out in 35+ year old residents in 10 Italian cities who died from respiratory causes (ICD-9 codes 460-519) in 2001-2005. Daily data of PM10, NO2 and O3 were obtained for the same period. Individual information on sociodemographic characteristics and chronic conditions from previous hospital admissions were collected. A conditional logistic regression analysis was carried out in each city and the pooled association was estimated in a random-effect meta-analysis. The exposure-response relationship was explored using a meta-smoothing approach. Results: There were 19,629 respiratory deaths (7.1% of natural causes). The risk of dying for respiratory causes increased by 2.29% (95% Confidence Intervals, CI, 1.03-3.58) for 10mg/m3 increase of PM10 (lag 0-3). The strongest associations were observed for chronic obstructive pulmonary disease (COPD) (2.79%) and pneumonia (2.30%) both at lag 0-5. There was no evidence of a threshold in the exposure-response curve and the relationship appeared to be linear. The PM10 effect was not confounded by co-pollutants (NO2 or O3). No statistically significant effect modifiers were detected although very elderly (85+ years), females and subjects with a chronic disease had the strongest effect. Conclusions: We found a no-threshold linear effect of PM10 on respiratory mortality in Italian cities. COPD and pneumonia showed the strongest association. Neither confounding from NO2 was indicated nor a clear effect modification was found.
Assessment of Short-Term Effects of Ambient Particulate Matter on Respiratory Mortality in Italian Cities.
VIGOTTI, MARIA ANGELA;
2009-01-01
Abstract
Background and objective: The impact of air pollution on morbidity and mortality is well established, especially for respiratory diseases. There are only few studies, however, that have evaluated the form of the exposure-response relationship and the role of individual effect modifiers. We studied the exposure-response relationship between daily PM10 and respiratory mortality in Italy, and considered the potential effect modifiers. Methods: A case-crossover analysis was carried out in 35+ year old residents in 10 Italian cities who died from respiratory causes (ICD-9 codes 460-519) in 2001-2005. Daily data of PM10, NO2 and O3 were obtained for the same period. Individual information on sociodemographic characteristics and chronic conditions from previous hospital admissions were collected. A conditional logistic regression analysis was carried out in each city and the pooled association was estimated in a random-effect meta-analysis. The exposure-response relationship was explored using a meta-smoothing approach. Results: There were 19,629 respiratory deaths (7.1% of natural causes). The risk of dying for respiratory causes increased by 2.29% (95% Confidence Intervals, CI, 1.03-3.58) for 10mg/m3 increase of PM10 (lag 0-3). The strongest associations were observed for chronic obstructive pulmonary disease (COPD) (2.79%) and pneumonia (2.30%) both at lag 0-5. There was no evidence of a threshold in the exposure-response curve and the relationship appeared to be linear. The PM10 effect was not confounded by co-pollutants (NO2 or O3). No statistically significant effect modifiers were detected although very elderly (85+ years), females and subjects with a chronic disease had the strongest effect. Conclusions: We found a no-threshold linear effect of PM10 on respiratory mortality in Italian cities. COPD and pneumonia showed the strongest association. Neither confounding from NO2 was indicated nor a clear effect modification was found.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
