Abstract Accurate estimation of the disease burden attributable to ambient particulate matter (PM2.5) requires reliable exposure data with an adequate spatial resolution. While coarse-resolution models are commonly used for national assessments, the impact of spatial resolution on environmental burden of disease (EBD) estimates remains understudied. We compared two modelling approaches for PM2.5 exposure in Germany: a chemical transport model (REM-CALGRID RCG, 2 × 2 km² resolution) and a high-resolution hybrid geostatistical land-use regression model (PMR, 100 × 100 m² resolution). EBD estimates were calculated for five health outcomes using established concentration-response functions. Across all outcomes, the PMR model consistently produced slightly higher EBD estimates—approximately 2%–3% higher disability-adjusted life years (DALYs) than the coarser RCG model. The difference is due to improved detection of local pollution hotspots and more accurate exposure assignments in the high-resolution model. A larger share of the population was classified into higher exposure categories under the PMR model. Although the added value of high-resolution modelling is limited at national level, it offers clear benefits for small-area analyses and applications in environmental justice and health equity. Our findings support the integration of fine-scale exposure data into public health surveillance and burden of disease (BoD) assessments where feasible.
Wallek et al. (Sat,) studied this question.