The impact of air pollution on hospital admissions: Evidence from Italy

In this paper we study the impact of air pollution on hospital admissions for chronic obstructive pulmonary disease for 103 Italian provinces, over the period from 2004 to 2009. We use information on annual mean concentrations of carbon monoxide, nitrogen dioxide, particulate matter, and ozone measured at monitoring station level to build province-level indicators of pollution. Hence, we estimate a regression model for hospital admissions, where we allow our aggregate measures of pollution to be subject to measurement error and correlated with the error term. We also adopt standard errors for estimates that are robust to serial and spatial correlation in the error term, to allow for temporal persistence and geographical concentration of unobservable risk factors. We find that higher levels of particulate matter are associated with higher levels of hospitalisation for children, while ozone plays an important role in explaining hospital admissions of the elderly. Other factors that appear to have an effect on hospital admissions for chronic obstructive pulmonary disease are precipitation and provincial unemployment rate.