Modelli biostatistici per lo studio degli effetti dell’inquinamento atmosferico su popolazioni fragili

The impact of air pollution on human health has become a hot topic in the last decades. Air pollution can mainly damage the respiratory, cardiovascular and neurological systems with short- and long-term effects. The harmful effects of exposure are observed in the general population as well as in vulnerable populations such as children and subjects with underlying pathologies. Numerous study designs and methods have been developed to assess the effect of exposure to air pollution on several outcomes; the choice of the best one mainly depends on the research question at hand and on the available data. The focus of this dissertation is the study of the effect of air pollution on vulnerable populations living in the Po valley, Italy, the most polluted area of Europe. In the first contribution, a cross-sectional study is conducted to assess the acute effect of Nitrogen Dioxide (NO2) and Particulate Matter of less than 10 µg/m3 (PM10) on the innate immune response of bronchial epithelial cells of children. Bronchial epithelial cells have been collected through a bronchos-copy and infected with rhinovirus, the expression of immune cells in response to the infection has been regressed against the short-term exposure to PM and NO2. A compromised ability of cells to face a viral infection was found in association with both the considered pollutants. A further contribution focuses on the Pediatric Emergency Department (PED) presentations for bronchiolitis in infants. A time-stratified case-crossover analysis is applied and combined with distributed lag non-linear models to evaluate if the exposure to air pollution could increase the risk of PED presentation and its time trend. PM has been found to impact at shorter lags, whereas NO2 increases the risk of PED presentations at longer lags. The overall cumulative effect of exposure has also been evaluated, showing a positive result for NO2. The last part of the thesis deals with the study of a cohort of children and adolescents discharged from the Pediatric Intensive Care Unit (PICU). Subjects are followed for one year and the first readmission for cardiovascular, respiratory or neurological diseases is regressed against the exposure to NO2, PM10 and Particulate Matter of less than 2.5 µg/m3 (PM2.5) subjects were monthly exposed to. This particular high-risk population shows an increased risk of readmission in association with all the considered pollutants, highlighting a risk that is not often considered in literature. Indeed, it is rare to find results of a cohort of high-risk patients followed over time to assess the risk of readmission associated with air pollution. By showing different results of the air pollution effects on vulnerable populations, this dissertation aims at evaluating the impact of NO2 and PM on different health outcomes in children living in the Po valley through different study designs and biostatistical methods.