The development of a measure of multiple physical environmental deprivation for epidemiological research. After United Kingdom and New Zealand, Portugal
Methods: PT-MEDIx was built at municipality-level (n=278, average population=35,000 inhabitants). The approach followed five stages: 1) identify health-relevant environmental factors, based on an extensive literature review; 2) acquire datasets about selected environmental factors and calculate municipality-level measures using Geographical Information Systems; 3) test associations between selected environmental factors and mortality using negative binomial models, adjusting for age, sex and socioeconomic deprivation; 4) construct a summary measure (PT-MEDIx) and 5) assess its association with mortality.
Results: We selected four dimensions of the physical environment: air pollution (particulate matter, carbon monoxide and nitrogen dioxide), climate (mean temperature), drinking water quality (trihalomethanes and nitrates), green space availability & proximity to polluting industry. PT-MEDIx score ranged from -1 (least environmental deprivation) to +4 (most). The index identified a clear spatial pattern within Portugal: least environmentally deprived municipalities in the inner north and center and most deprived in urban settings. Comparing with the intermediate group (PT-MEDIx=0) there was a significant positive association between environmental deprivation and all-cause (IRR=0.980 [0.961-1.001] in the least deprived, 1.024 [1.004-1.045] in the most deprived); cardiovascular (IRR=0.942 [0.912-0.974], 1.062 [1.028-1.097]) and cancer mortality (IRR=0.944 [0.909-0.973], 1.111 [1.084-1.128]). A dose-response relationship was particularly evident for cancer mortality.
Conclusions: Earlier UK/NZ indexes have good transferability to Southern Europe, meaning that the replication of MEDIx might contribute to untangle the complex pathways that link health, socioeconomic and physical environment.