Cancer Mortality and Industrial Pollution in Spain

Abstract

P-169 Introduction: The European Pollutant Emission Register (EPER) is a public register of industrial emissions into air and water (direct or via a sewerage system) of 50 pollutants created by the Decision 2000/479/CE (http://www.eper.cec.eu.int/). The aim of this study was to evaluate the statistical association between air-pollutant emissions from industrial facilities included on EPER and cancer mortality in populations living in Spanish towns around these facilities. Methods: This ecological study analysed 22 different cancer types linked to environmental exposures. Observed cancer deaths were provided by the Spanish National Statistics Institute from 1989 to 1998. Expected deaths were calculated for each of the 8077 Spanish towns with the overall mortality for Spain by age group and sex taken as reference for each cancer location. Data exposure was obtained from EPER database for 2001 year. Relative risks (RR) of cancer were estimated using Poisson regression models, with observed cancer cases as dependent variable, expected cases as offset, presence of industrial facility as explicative variable, and socio-demographic variables (population size, income level and percentages of older-than-65, farming, unemployed, illiterate population) provided by 1996 municipal roll as covariates. For those cancers showing a statistically significant mortality risk excess, we applied the spatial Poisson model of Besag, York and Mollié (BYM), fitted by Markov chain Monte Carlo simulation methods. Results: Presence of EPER industrial facilities in Spanish municipalities was associated with an excess mortality risk, independently of socio-demographic variables, for the following cancers (RR, IC95%): oesophagus (1.049, 1.014–1.086), colorectal (1.018, 1.003–1.033), lung (1.037, 1.025–1.049), connective tissue (1.161, 1.078–1.250), bladder (1.035, 1.009–1.061), kidney (1.100, 1.058–1.144), Non-Hodgkin lymphoma (1.053, 1.018–1.089), leukaemia (1.033, 1.003–1.063). Non-statistically significant mortality risk excess was observed among other analysed cancers. Discussion and Conclusions: We used the presence of industrial facilities, instead of a distance measure, due to the fact that the geographical coordinates provided by EPER databases need to be validated. The presence of industrial facilities registered in EPER with air-pollutant emissions in Spanish towns seems to be associated with the increase of mortality risk due to some cancers. Both Poisson regression and BYM models suggested similar findings. The risk differences by sex allow to differentiate indirectly the risk derived from occupational exposure. EPER information can be useful to determine the relationship between industrial pollution and health outcomes in Europe. This project was supported by: FIS 040041and RCESP-FIS C03/09.