Reanalysis of non-occupational exposure to asbestos and the risk of pleural mesothelioma

Since these data are available online to readers, I present an analysis that replaces the Camus et al’s4 study of 1998 with the more recent data. Results The results of the analyses are shown in Table 1.Table 1 Meta-regression of the relative risk of pleural mesothelioma in relation to explanatory variables Risk factor Base model (Use all the data in table 1 of Marsh, but replace the Camus study with that of the Quebec Institute of Public Health.) Sensitivity analysis 1 (Drop the McDonald and McDonald’s study5 of 1980 from the analysis because it has been replaced by a later update.) Sensitivity analysis 2 (McDonald and McDonald5 describe cases for whom domestic exposure was to clothing of an employee engaged in ‘insulation or factory work.’ Reclassify exposure to mixed fibre type.) Relative risk (95% CI)* Relative risk (95% CI) Relative risk (95% CI) Relative risk (95% CI) Baseline (any exposure vs no exposure to asbestos) 6.09 (2.14 to 17.3) 6.09 (2.35 to 15.7) 6.15 (2.41 to 15.7) 5.71 (2.32 to 14.03) Neighbourhood exposure versus household 1.39 (0.38 to 5.09) 1.35 (0.41 to 4.42) 1.22 (0.32 to 4.65) 1.20 (0.42 to 3.41) Chrysotile versus mixed fibres 0.61 (0.10 to 3.85) 1.06 (0.18 to 6.18) 1.38 (0.16 to 12.2) 3.15 (0.40 to 24.8) Amphibole versus mixed fibres 2.85 (0.32 to 25.5) 2.89 (0.38 to 22.1) 3.07 (0.40 to 23.5) 3.33 (0.49 to 22.9) Test of the hypothesis that not all of the explanatory variables are 1.0 P=0.60 (Rejected) P=0.63 (Rejected) P=0.61 (Rejected) P=0.37 (Rejected) (Continued below) *These are multiplicative models, so the risk attributable to any set of factors is the baseline risk multiplied by the factor-specific risk.