Sound empirical and quantitative analysis on the relationship between different patterns of urban expansion and environmental or social costs of mobility are still very rare in Europe and the few studies available provide only a qualitative discussion on this. Recently, Camagni et al. (2002) have performed an empirical analysis on the metropolitan area of Milan, aimed at establishing whether different patterns of urban expansion generate different levels of land consumption and heterogeneous impacts of urban mobility. Results confirm the expectation that higher environmental impact of mobility is associated with more extensive and sprawling urban development, more recent urbanisation processes and residential specialisation. The present paper enlarges further the empirical analysis to seven Italian metropolitan areas (namely, Bari, Florence, Naples, Padua, Perugia, Potenza and Turin) to corroborate previous results for the Italian context. The novelty of the present paper is threefold. Firstly, we are interested in exploring the changes occurred to the intensity of the mobility impact across a ten-year period, from 1981 to 1991, corresponding to the Italian economic boom years. Secondly, using an econometric analysis in cross-section, we consider several metropolitan areas at once, being therefore able to explore whether there are significant differences in the way the model explains variations in the mobility impact across various Italian urban areas. Finally, we propose a conceptual interpretation of the causal chain in the explanation of the mobility impact intensity and we test it using Causal Path Analysis.