Can population characteristics account for the variation in health impacts of air pollution? A meta-analysis of PM10-mortality studies

摘要

In this paper a regression analysis is undertaken using the largest sample of air pollution mortality studies to date, from both developing and developed countries, in anudattempt to further the understanding of the relationship between suspended particles and mortality. Applying Empirical Bayes meta-analysis, it is estimated that mortality rates on average increase by 6 per cent per 100-mg/m3 increase in Particulate Matter (PM10) concentrations, with greater effects in countries with high income inequality. We further find evidence that education and income have an influence on the effects of PM pollution.