Mesothelioma risk among those exposed to chrysotile asbestos only and mixtures that include amphibole: a case–control study in the USA, 1975–1980

OBJECTIVES: Occupational asbestos exposure is causally linked to mesothelioma. However, whether exposure to only chrysotile asbestos is associated with mesothelioma risk, and the heterogeneity in risk by different fibre types/lengths remains unclear. We investigated whether mesothelioma risk differs among workers exposed to only chrysotile asbestos compared with chrysotile and ≥1 amphibole (ie, amosite, tremolite, anthophyllite and crocidolite) over the working lifetime. METHODS: We analysed next-of-kin interview data including occupational histories for 580 white men (176 cases and 404 controls) from a case-control study of mesothelioma conducted in the USA in 1975-1980. Asbestos exposure was determined by an occupational hygienist using a job-exposure matrix and exposure categories included chrysotile only and nine chrysotile-amphibole mixtures. Logistic regression models were used to estimate the ORs and 95% CIs of mesothelioma, comparing each asbestos category to the unexposed group, adjusted for age at death and data source. Analysis of contrasts was used to assess overall heterogeneity and pair-wise differences in risk. RESULTS: Exposure to long and short chrysotile only was associated with increased mesothelioma risk compared with the unexposed (OR=3.8 (95% CI 1.3 to 11.2)). The complex mixture of extra-long amosite, short and long chrysotile, tremolite and anthophyllite was associated with the highest risk (OR=12.8 (95% CI 4.1 to 40.2)). There was evidence for overall heterogeneity among the asbestos exposure categories (p heterogeneity=0.02). However, the lower risk observed for exposure to chrysotile only compared with the complex mixture was not significant (p difference=0.10). CONCLUSIONS: Our findings suggest that policies aimed at regulating asbestos should target both pure chrysotile and mixtures that include amphibole.