Understanding the interplay between air pollution, physical activity and human mobility using large-scale behavioural data

Exposure to air pollution is the second leading cause of deaths from non-communicable diseases (NCDs). Based on small-scale experimental studies, it is suggested that exposure to air pollutants is deeply intertwined with individual lifestyle and behaviour. In the project I am going to study pollution exposure and how it affects human behaviour and its consequences on health benefits.

In my project, I will first estimate individuals levels of exposure to pollutants performing an “home-based” and a “mobility-based” estimation. In fact, most studies estimate exposure considering only the air quality in the area surrounding individuals’ home location. I propose to test that the home-location estimates are biased due to individuals spending a substantial amount of time outside their residential location. My study will build upon similar studies that focused on aggregated data for selected US cities, where it was shown that there are substantial differences in exposure according to racial and socioeconomic factors.

I will then concentrate on studying how changes in exposure to pollutants drive changes in physical activity behavior, while accounting for confounders such as weather conditions and media alerts.
In particular, it is suggested that air pollution may prevent people from being physically active during extreme air-pollution events or from engaging in physical activity overall in highly polluted environments.

Moreover I will study how increase in pollution exposure leads to long lasting loss in health benefits associated with physical activity. Here, I expect to find that individuals undergoing different levels of exposure carry out physical activity at different levels, showing that high levels of exposure and of physical inactivity account as significant risk factors for NCDs