Medea

Acción Estratégica Medea

Methodology

This is a cross-sectional ecological and time-series study of mortality in the selected cities, which uses census tracts as the unit of analysis. For each area and period the smoothed Standardised Mortality Ratio will be calculated using full Bayesian hierarchical models, specifically the model proposed by Besag, York and Mollie, and assuming a Poisson distribution for the observed deaths. Indicators of exposure, both socio-economic and environmental, will be obtained. Gender-sensitive socio-economic deprivation indices will be constructed through principal components analysis. The effects of socio-economic level and of environmental pollution on mortality risk will be assessed. Auto-regressive spatio-temporal modelling will also be carried out with the yearly data. All analyses will be carried for men and women separately.

The project involves 14 different research groups in different parts of Spain, coordinated by the Agència de Salut Pública de Barcelona (Barcelona Public Health Agency). Seven working groups have been set up to coordinate different sections of the project.

The analysed cities and areas are the following: A Coruña, Alicante, Avilés, Barcelona, Bilbao, Capitales de provincia de Andalucía, Castellón, Comunidad de Madrid, El Camp de Tarragona, Ferrol, Gijón, La región metropolitana de Barcelona, La región sanitaria de Girona, Las Palmas de Gran Canaria, Logroño, Lugo, Municipio de Cartagena y la Unión, Pamplona, Pontevedra, San Sebastián, Santa Cruz de Tenerife, Santiago de Compostela, Oviedo, Ourense, Valencia, Vigo, Vitoria, Zaragoza.