Comparison of Long-Term Air Pollution Exposure from Mobile and Routine Monitoring, Low-Cost Sensors, and Dispersion Models

This report, available for downloading below, presents a study led by Gerard Hoek at Utrecht University, the Netherlands. Hoek and colleagues compared the performance of a suite of long-term exposure assessment models in the Netherlands for four air pollutants: ultrafine particles, black carbon, fine particulate matter, and nitrogen dioxide. These pollutants can vary widely in space and time, posing challenges for predicting long-term air pollution exposure.

The investigators compared the predictions of exposure models at 20,000 random residential addresses in the Netherlands and tested them on existing and new validation data over a 20-year period. The study included a comparison of health effects estimates of the different models in three cohort studies.

The models generally resulted in moderately to highly correlated exposure predictions for all pollutants except fine particulate matter. Similar findings were reported on the presence of an association with various mortality and morbidity outcomes among the various exposure estimates, albeit with sometimes notable differences in the magnitude of the associations.