Estimating historical exposure to silica among mine and pottery workers in the people's Republic of China

Publisher Website
Google Scholar
Abstract
A quantitative retrospective exposure assessment method was developed for use in a nested case‐control study of lung cancer among mine and pottery workers exposed to silica dust in the People's Republic of China. Exposure assessment was carried out in 20 mines (10 tungsten, 6 iron/copper, and 4 tin) and nine pottery factories. A job title dictionary was developed and used in both the collection of historical exposure information and work histories of 1,668 (316 cases and 1,352 controls) study subjects. Several data abstraction forms were developed to collect historical and current exposure information and employees' work histories, starting in 1950. A retrospective exposure matrix was developed on the basis of facility/job title/calendar year combinations using available historical exposure information and current exposure profiles. Information on the amount of respirable, thoracic, and free silica content in total dust was used in estimating exposure to silica. Starting in 1950, 6,805 historical estimates had been carried out for 14 calendar‐year periods. We estimated the average total dust concentration to be 9 mg/M3, with a range from 28 mg/M3 in earlier years to 3 mg/M3 in recent years. Several exposure indices [such as cumulative dust, average dust, cumulative respirable (<5 μ in particle size) and thoracic (<10 μ in particle size) silica dust, average respirable and thoracic silica dust, exposure‐weighted duration, and the highest/longest exposure] were calculated for individuals by merging work history and historical exposure matrix for each study subject. We developed these various measures of exposure to allow investigators to compare and contrast different indices of historical exposure to silica. © 1993 Wiley‐Liss, Inc. 1 This article is a US Government work and, as such, is in the public domain in the United States of America.

This publication has 18 references indexed in Scilit: