Mesothelioma in occupational cohort studies: methodological considerations.

OBJECTIVE: This article describes effective strategies for the identification and valid assessment of mortality due to mesothelioma. METHODS: We manually reviewed all death certificates for mention of mesothelioma for all International Classification of Diseases (ICD) revisions. We tested the accuracy of our ascertainment method by comparing New Jersey death certificate data from our health status registry with histologically confirmed cases from the New Jersey State Cancer Registry. RESULTS: We found reasonably good agreement between death certificate diagnoses and histologically confirmed cases, κ coefficient 0.86 (95% confidence interval, 0.76 to 0.95). Most mesothelioma deaths in our test and North American cohorts were coded to unspecified anatomical sites. CONCLUSIONS: Limiting ascertainment to pleura and peritoneum ICD codes underestimates mesothelioma deaths. Reviewing all ICD codes that could contain mesothelioma is the only effective method for complete capture of mesothelioma diagnoses.