Exposure-specific lung cancer risks in Chinese chrysotile textile workers and mining workers

Abstract

ObjectiveWhether there is a difference in the exposure–response slope for lung cancer between mining workers and textile workers exposed to chrysotile has not been well documented. This study was carried out to evaluate exposure-specific lung cancer risks in Chinese chrysotile textile workers and mining workers. Subjects and methodsA chrysotile mining worker cohort and a chrysotile textile worker cohort were observed concurrently for 26 years. Information on workers’ vital status, occupational history and smoking habits were collected, and causes and dates of deaths were verified from death registries. Individual cumulative fiber exposures were estimated based on periodic dust/fiber measurements from different workshops, job title and duration, and categorized into four levels (Q1–Q4). Standardized mortality ratios (SMRs) for lung cancer were calculated and stratified by industry and job title with reference of the national rates. Cox proportional hazard models were fit to estimate the exposure-specific lung cancer risks upon adjustment for age and smoking, in which an external control cohort consisting of industrial workers without asbestos exposure was used as reference group for both textile and mining workers. ResultsSMRs were almost consistent with exposure levels in terms of job titles and workshops. A clear exposure–response relationship between lung cancer mortality and exposure levels was observed in both cohorts. At low exposure levels (Q1 and Q2), textile workers displayed higher death risks of lung cancer than mining workers. However, similarly considerably high risks were observed at higher exposure levels, with hazard ratios of over 8 and 11 at Q3 and Q4, respectively, for both textile and mining workers, after both age and smoking were adjusted. ConclusionThe chrysotile textile workers appeared to have a higher risk of lung cancer than the mining workers at a relatively low exposure level, but no difference was observed at a high exposure level, where both cohorts displayed a considerably high risk.