Epidemiological Findings of Cohorts
Chrysotile exposures without identified amphibole fibers do not appear to increase the risk of being diagnosed with mesothelioma based on the results of epidemiological cohort studies of over 220,000 individuals. The 7 cases of mesothelioma reported in the 71 cohort studies (i.e., from baseline, well-defined populations) where no amphibole exposures were identified are summarized in Table 4. However, careful review of these few cases illustrate that their exposures were likely mixed, the diagnosis questionable, and/or the latency period inadequate or unstated. This analysis has not identified a case of mesothelioma from the cohort studies that is definitely documented as solely exposed to raw chrysotile fibers uncontaminated by amphiboles. Morphological and recent toxicological evidence implicates balangeroite as an amphibole-acting contaminant; therefore the five cases reported from the Balangero asbestos mine are in Table 2.
Simply for illustrative purposes, among approximately 32,853 subjects exposed to amphiboles, 404 cases of mesothelioma ( 1.23%) were reported (Table 1 ), whereas only 7 cases (at most) were observed for 32,039 subjects exposed to chrysotile, or 0.04% (Table 3). Mixed fiber exposures resulted in an intermediate percentage of 0.67% for cases (994/147,384) (see Table 2). Clearly the trend is greatly slanted towards amphiboles, as found by other reviewers (Hodgson and Darnton, 2000; Berman and Crump, 2003). These latter two sets of cohort studies do not differ greatly when compared to those summarized in Table 1 in terms of time of follow-up, exposures levels, time frames of exposures, and diagnostic methods, so these factors cannot account for the paucity of mesothelioma cases in asbestos cohort studies where amphibole exposures were not identified.
As described earlier, a confounder is a factor or exposure associated with the disease and the exposure of interest. In order for confounding to substantially affect estimates of risk, the association of the potential confounder with disease must be stronger than the observed association between the exposure of interest and the disease. The relative potency of amphiboles in causing mesothelioma is very great relative to chrysotile, assuming chrysotile has any mesotheliogenic potency (ATSDR, 2003, pp. 94-95). From the data in Tables 1 and 2 herein, one can arrive at the same conclusion: The risk of mesothelioma is primarily if not solely from exposure to amphiboles.
Replication by well-designed, relevant studies for confirmation of hypotheses is absolutely necessary to establish an association. Consistency of results in different studies testing the same hypothesis was a guiding principle of the U.S. Surgeon General's report on smoking and health (Bayne-Jones, 1964). Replication is one of the bulwarks of the scientific method that helps distinguish true from false claims. Under its "criteria for causality" the World Health Organization (WHO) states that associations that are replicated are more likely to imply causality. To quote a criterion used by the WHO and International Agency for Cancer Research (IARC) for basing their opinions, "When several epidemiological studies show little or no indication between an exposure and cancer, the judgment may be made that, in the aggregate, they show evidence of lack of carcinogenicity" (IARC, 2000).