Is particulate air pollution at the front door a good proxy of residential exposure?

The most advanced epidemiological studies on health effects of air pollution assign exposure to individuals based on residential outdoor concentrations of air pollutants measured or estimated at the front-door. In order to assess to what extent this approach could cause misclassification, indoor measurements were carried out in unoccupied rooms at the front and back of a building which fronted onto a major urban road. Simultaneous measurements were also carried out at adjacent outdoor locations to the front and rear of the building. Two 15-day monitoring campaigns were conducted in the period June-December 2013 in a building located in the urban area of Bologna, Italy. Particulate matter metrics including PM2.5 mass and chemical composition, particle number concentration and size distribution were measured. Both outdoor and indoor concentrations at the front of the building substantially exceeded those at the rear. The highest front/back ratio was found for ultrafine particles with outdoor concentration at the front door 3.4 times higher than at the rear. A weak influence on front/back ratios was found for wind direction. Particle size distribution showed a substantial loss of particles within the sub-50 nm size range between the front and rear of the building and a further loss of this size range in the indoor data. The chemical speciation data showed relevant reductions for most constituents between the front and the rear, especially for traffic related elements such as Elemental Carbon, Iron, Manganese and Tin. The main conclusion of the study is that gradients in concentrations between the front and rear, both outside and inside the building, are relevant and comparable to those measured between buildings located in high and low traffic areas. These findings show high potential for misclassification in the epidemiological studies that assign exposure based on particle concentrations estimated or measured at subjects' home addresses.