Abstracts: ISEE 20th Annual Conference, Pasadena, California, October 12-16, 2008: Contributed Abstracts
ChemRisk, Houston, TX, USA
Abstracts published in Epidemiology have been reviewed by the organizations of Epidemiology. Affliate Societies at whose meetings the abstracts have been accepted for presentation. These abstracts have not undergone review by the Editorial Board of Epidemiology.
Human exposure to cobalt (Co), thallium (Tl), tungsten (Tu), and uranium (Ur) is primarily occupational. Although a biological exposure index has been established for Co, current reference values for Tl, Tu, and Ur have not been previously characterized in detail. Using the 2001–2002 and 2003–2004 NHANES data, we calculated reference statistics for each of the four metals and evaluated the demographic characteristics that had the greatest impact on levels of these metals in urine.
Demographic information and urinary concentration data for creatinine, cobalt, thallium, tungsten and uranium from the 2001–2002 and 2003–2004 NHANES surveys were combined into a single dataset and utilized in these analyses. All concentrations of metals were adjusted for urinary creatinine. Reference concentrations, including the geometric mean and various percentiles, were characterized for each metal by region and age, gender, race/ethnicity, or tobacco smoke exposure. Adjusted associations between the various factors and levels of metals were evaluated for each metal independently using linear regression. The natural log of cobalt, thallium, tungsten or uranium concentrations were used as the dependent variable in all regression models. All analyses were completed using SAS and SUDAAN software.
The overall geometric mean (GM) levels of creatinine-adjusted urinary cobalt, thallium, tungsten, and uranium for the general, U.S. population were 0.35 μg/g (95% CI: 0.33-0.36), 0.16 μg/g (95% CI: 0.15–0.17), 0.076 μg/g (95% CI: 0.071–0.081), and 0.0081 μg/g (95% CI: 0.0074–0.0088), respectively. Generally, concentrations of cobalt and thallium varied little across geographic regions. For tungsten, some variation was observed among the different regions, but only concentrations of uranium varied significantly by region. Interestingly, adjusted GM levels of all four metals were considerably higher in young children (one to five years of age) and slightly higher in females and individuals living in the West region. For cobalt, tungsten and uranium, adjusted geometric mean concentrations were higher for smokers. While non-Hispanic blacks had lower levels of cobalt, thallium and uranium than non-Hispanic whites, Mexican Americans had higher levels of all four metals than non-Hispanic whites.
The National Health and Nutrition Examination Surveys provide a rich source of data with which to determine reference values for various compounds that may be encountered through occupational exposure. Additionally, the data is useful in evaluating which populations are at greatest risk of having elevated urinary concentrations of these compounds. In utilizing the NHANES data to determine referent values of these metals, we have provided needed information with which to compare results from biological monitoring studies of cobalt, thallium, tungsten, and uranium in potentially exposed individuals and populations.