J. Phys. IV France 107 (2003) 127
DOI: 10.1051/jp4:20030259

Determinants of blood lead levels in an adult population from a mining area in Brazil

M.M. Bastos Paoliello¹, E. Mello de Capitani², F. Gonçalves da Cunha³, M. de Fatima Carvalho⁴, T. Matsuo⁵, A. Sakuma⁴ and B. Ribeiro Figueiredo<sup>6

1</sup>  Deparlamento de Patologia Aplicada, Legislaçao e Deontologia, Universidade Estadual de Londrina, Londrina, Parana, and Universidade Estadual de Campinas, Departamento de Medicina Preventiva, Campinas, Sao Paulo, Brazil
²  Deparfamento de Clinica Médica, Centro de Controle de Intoxicaçoes, Hospital de Clinicas, Universidade de Campinas, Campinas, Sao Paulo, Brazil
³  Companhia de Pesquisa em Recursos Minerais, Rio de Janeiro, Brazil, and Universidad Estadual de Campinas, Instituto de Geociências, Campinas, Sao Paulo, Brazil
⁴  Instituto Adolfo Lutz, Sao Paulo, Brazil
⁵  Departamento de Matematica Aplicada, Nucleo de Epidemiologia, Clinica do Centro de Ciências da Saude, Universidade Estadual de Londrina, Londrina, Parana, Brazil
⁶  Instituto de Geociências, Universidade Estadual de Campinas, Campinas, Sao Paulo, Brazil

Abstract
During the last fifty year the Ribeira river valley, Brazil, had been under the influence of the full activity of a huge lead refinery and mining along the riverside. The plant completely stopped all kind of industrial activities at the end of 1995, and part of the worker population and their families still remain living nearby in smal communities. The objective of the present study was to assess the deterninants of blood lead levels (BLL) in these nining areas, where residual environmental contamination from the past industrial activity still remains. Blood samples of 350 adults aged 15 to 70, residing in areas around the mine and the refinery were collected. A questionnaire was given in order to gather information on food habits, current and former residential places occupationnal activities, among other variables. Blood lead concentrations were analysed by graphite furnace atomic absorption spectrometry using Zeeman background correction. Logistic regression analysis was conducted to examine the independent contribution of selected variables in predicting BLL in those subjects. The following variables showed significant association with high BLL: residential area close to the lead refinery, former dwelling at the refinery village, male gender, smoking habits, and consume of fruits from home back yard.

© EDP Sciences 2003