Comparative analysis of European cities in terms of air quality and health outcomes using Data Envelopment Analysis

Abstract

Air quality in contemporary cities is not only an environmental issue but also an important determinant of public health. This paper examines the relative ability of European cities to achieve favourable health outcomes under comparable levels of air pollution and urbanisation pressure. The analysis draws on a balanced panel of 20 European cities for 2015–2024 (200 city–year observations). An output-oriented BCC Data Envelopment Analysis (DEA) model was applied, with inputs comprising PM2.5 concentration, NO 2 concentration, and population density, and outputs comprising life expectancy and a reciprocally transformed rate of respiratory hospitalisations. Most cities lie close to the efficiency frontier (mean = 0.977; SD = 0.018; range = 0.933–1.000), implying an average gap of approximately 2.4% in desirable health outcomes. Bratislava (0.997) and Budapest (0.996) recorded the most favourable long-term profiles, whereas Prague (0.946) and Helsinki (0.944) recorded relatively weaker performance. Descriptive trend analysis suggests statistically significant changes in selected cities including Krakow (β = 0.0006; p = 0.037), Brussels (β = − 0.0044; p = 0.018), and Prague (β = −0.0041; p = 0.049). PCA results indicate that the selected pressure indicators capture largely distinct dimensions, although the overall sampling adequacy remains limited. The study links public health, air quality, and city benchmarking to provide a multidimensional perspective relevant for urban policy discussions. The findings represent relative efficiency estimates rather than causal effects or comprehensive measures of urban quality of life and should therefore be interpreted in light of the model specification, limited variable set, and data constraints.