© 2006 American Thoracic Society
Residential Proximity to Naturally Occurring Asbestos and Mesothelioma Risks: Further Consideration of Exposure Misclassification and Occupational ConfoundingTo the Editor:We read with interest the recent article by Pan and coworkers (1) and the subsequent letter to the editor and response (2, 3). The letter points out important study design limitations of Pan and coworkers' study (e.g., long latency of mesothelioma, exposure misclassification, and bioavailability of asbestos fibers). We believe that study design limitations and effects of exposure conditions should be further addressed. The fact that many of the regions classified as "exposed" have experienced a substantial influx of population migration (4) strongly suggests exposure misclassification when relying on current residence information. In addition, important problems with exposure misclassification are likely because of the following: (1) the asbestos content in ultramafic rock varies considerably; (2) there are other nonultramafic sources of environmental asbestos exposure; (3) the assumption that meteorologic conditions would distribute asbestos fibers with radial symmetry around a source is contradicted by actual conditions, which heavily favor a downwind direction; (4) the distances assigned as "exposed," which ranged up to 100 km, are not realistic given that the effect being tracked is extremely small relative to the much larger effects of contributions from other factors (e.g., occupational exposures); and (5) lack of consideration of the potential difference in fiber type (amphibole vs. chrysotile), which has important implications for potential public health risk (5). In their response, Schenker and coworkers note that confounding by prior occupations is an unlikely explanation of their findings, based on a survey of union workers' migration patterns and analytical control for previous occupation (3). This survey reflects residential migration of a small number of workers from one union in one region of California and is of limited value in generalizing about migration of formerly exposed workers (6). In addition, the occupational information reflects only at most one job title that was available in cancer registry records. This incomplete work history information may miss relevant amphibole asbestos exposure among mesothelioma cases. Furthermore, occupational information was missing on approximately 25% of the cases and 37% of the controls. Given these limitations, occupational confounding may not have been adequately addressed and can potentially swamp the effect studied, which is acknowledged to be much smaller. The large number of mesothelioma cases in the California cancer registry provides the opportunity for further study; however, before appropriate risk communication and public health policy can be developed, substantial improvement in assessment of exposure and confounders is needed.
Exponent, Inc., Menlo Park, California and UCLA School of Public Health, Los Angeles, California
Aeolus, Inc., Albany, California
The Graduate School and University Center of the City University of New York, New York, New York FOOTNOTES Conflict of Interest Statement: The employer of M.A.K, Exponent, has received funding from the National Stone, Sand & Gravel Association (NSSGA), for research work on naturally occurring asbestos. D.W.B. received grants from the NSSGA to conduct several independent research projects on naturally occurring asbestos (NOA). He also consulted for NSSGA over the same time period. In 2006, he also consulted for the El Dorado County, Department of Education, on NOA-related issues. For many years he has conducted research for the U.S. EPA related to development of analytical methods and risk assessment procedures for assessing the hazards of asbestos in general and has consulted for the agency to assist with site-specific assessment of asbestos exposure and risk. At the end of 2004, the Agency for Toxic Substances and Disease Registry retained him to conduct a course in asbestos exposure and risk and provide software for conducting life table analyses for assessing asbestos risk. In 2002 and 2003, he conducted a retrospective risk assessment on NOA and developed a monitoring plan to protect neighbors during construction in the City of San Francisco. A.M.L. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. This letter was independently prepared by the signed authors and was not subject to review and approval by NSSGA or its members. REFERENCES
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
