Air pollution, aeroallergens and admissions to pediatric emergency room for respiratory reasons in Turin, northwestern Italy

Abstract

Background: Air pollution can cause respiratory symptoms or exacerbate pre-existing respiratory diseases, especially in children. This study looked at the short-term association of air pollution concentrations with Emergency Room (ER) admissions for respiratory reasons in pediatric age (0-18 years).

Methods: Daily number of ER admissions in a children's Hospital, concentrations of urban-background PM2.5, NO2, O3 and total aeroallergens (Corylaceae, Cupressaceae, Gramineae, Urticaceae, Ambrosia, Betula) were collected in Turin, northwestern Italy, for the period 1/08/2008 to 31/12/2010 (883 days). The associations between exposures and ER admissions were estimated, at time lags between 0 and 5 days, using generalized linear Poisson regression models, adjusted for non-meteorological potential confounders.

Results: In the study period, 21,793 ER admissions were observed, mainly (81 %) for upper respiratory tract infections. Median air pollution concentrations were 22.0, 42.5, 34.1 μg/m(3) for urban-background PM2.5, NO2, and O3, respectively, and 2.9 grains/m(3) for aeroallergens. We found that ER admissions increased by 1.3 % (95 % CI: 0.3-2.2 %) five days after a 10 μg/m(3) increase in NO2, and by 0.7 % (95 % CI: 0.1-1.2 %) one day after a 10 grains/m(3) increase in aeroallergens, while they were not associated with PM2.5 concentrations. ER admissions were negatively associated with O3 and aeroallergen concentrations at some time lags, but these association shifted to the null when meteorological confounders were adjusted for in the models.

Conclusions: Overall, these findings confirm adverse short-term health effects of air pollution on the risk of ER admission in children and encourage a careful management of the urban environment to health protection.

Keywords: Airborne pollutants; Pediatric emergency room; Pollens; Short-term respiratory effects; Time-series analysis.

Fig. 1

Monthly distribution of daily ER admissions for respiratory diseases during the study period

Fig. 2

Monthly distribution of daily PM_2.5, NO₂, O₃ and aeroallergen concentrations (panels a − d, respectively) during the study period

Fig. 3

Pairwise distributions and Pearson’s r correlation coefficients of air pollutant concentrations and meteorological variables. All the hypothesis tests of no correlation were statistically significant (all p <0.001)

Fig. 4

Estimates of association of daily PM_2,5, NO₂, O₃, and aeroallergen concentrations (panels a − d, respectively) with ER admissions for respiratory diseases at different time lags, adjusted for medium/long-term trend function, day of the week, influenza outbreaks, holidays and summer population decrease.* *Single-pollutant models, see “model A” in the Statistical analysis section. Relative risks (RR) with 95%CIs are given for a 10 μg/m³ increase in PM_2,5, NO₂, O₃ concentrations or a 10 grains/m³ increase in aeroallergen concentrations

Fig. 5

Estimates of association of daily PM_2,5, NO₂, O₃, and aeroallergen concentrations (panels a − d, respectively) with ER admissions for respiratory diseases at different time lags, adjusted for medium/long-term trend function, day of the week, influenza outbreaks, holidays, summer population decrease, and for the meteorological variables (daily temperature, daily relative humidity, cumulative daily precipitations).* * Single-pollutant models, see “model B” in the Statistical analysis section of the article. Relative risks (RR) with 95%CIs are given for a 10 μg/m³ increase in PM_2,5, NO₂, O₃ concentrations or a 10 grains/m³ increase in aeroallergen concentrations