Method Fifty workers from two yacht factories participated, including 23 FRP laminators and 27 non-FRP workers. Pre-shift colour confusion index (CCI) was measured on the first workday of a week by using Lanthony Desaturated Panel D-15d. Concentrations of volatile organic compounds related to FRP laminating were collected by a stainless canister, and then analysed using a GC/MS. Cumulative working hours for FRP laminating was obtained from daily administrative records of the companies. Logistic regression was used to assess correlation between cumulative working hours (past 0.5, 2, and 8 years) of FRP laminating and pre-shift CCI, where work time and CCI were stratified by median and the model was adjusted for age and regular alcohol consumption. Results The mean styrene exposure during FRP laminating was 4.2 ppm. The median of cumulative working hours for FRP laminating in the past 0.5, 2 and 8 years was 2, 49, 248 h respectively. Higher cumulative working hours for FRP laminating was associated with poor CCI (past 0.5 yrs, odd ratio (OR)=3.1, p = 0.1; past 2 yrs, OR=4.8, p = 0.03; past 8 yrs, OR=6.5, p = 0.01). The effect of long-term exposure to styrene appeared to be stronger than short-term exposure. Conclusions Long-term exposure to styrene from FRP laminating was associated with colour vision impairment. Objectives To study the associations between head and neck cancer risk and occupational exposure to asbestos, mineral wools and silica. Method ICARE is a population based case-control study conducted in France. Analyses were restricted to men and included 1833 cases of head and neck squamous cell carcinomas (HNSCC) and 2747 controls. Complete occupational history was collected. Occupational exposures were assessed through job-exposure matrices. Logistic models were used to estimate adjusted odds ratios (ORs) and 95% confidence intervals (CI). Results Exposure to asbestos was associated with an elevated risk of HNSCC (OR=1.9, CI 1.6-2.3), and the risk increased significantly with the probability, duration and cumulative level of exposure. Significantly increased risks were found for all cancer sites: larynx (OR=2.1, CI 1.6-2.8), hypopharynx (OR=2.0 CI 1.5-2.8), oropharynx (OR=1.6 CI=1.3-2.1) and oral cavity (OR=1.9 CI 1.4-2.6). Conversely, after adjustment for asbestos exposure, exposure to mineral wools was not associated with HNSCC risk (OR=0.8 CI 0.6-1.0), for any of the cancer sites. Exposure to silica was not associated with HNSCC risk overall (OR=0.9 CI 0.7-1.2), but non-significantly elevated ORs were observed for the highest level of cumulative exposure for oropharyngeal (OR=1.6 CI 0.8-3.2) and hypopharyngeal cancer (OR=1.9 CI 0.9-4.1). Conclusions Our findings confirm the role of asbestos exposure in laryngeal cancer, and suggest that asbestos exposure increases also the risk of oral and pharyngeal cancer. There is some evidence of an association between silica exposure and pharyngeal cancer. Exposure to mineral wools was not associated with HNSCC risk in our study. Results Compensation statistics...