Factors affecting the mesothelioma detection rate within national and international epidemiological studies: insights from Scottish linked cancer registry-mortality data.

ICD-9 code 163 (malignant neoplasm of pleura) listed as underlying cause of death detected only 40% of Scottish mesothelioma cases (all body sites) from the cancer registry in 1981-1999. This is lower than both the previously published 55% figure, derived from UK mesothelioma register data 1986-1991, which is based on any mention of mesothelioma on death certificates, cross-referenced to cancer registry data, and the 44% figure derived from Scottish mortality data 1981-1999, which captured any mention of mesothelioma on the death certificate. Detection from cancer registry data increased to 75% under ICD-10 in Scotland, confirming earlier predictions of the benefit of ICD-10's more specific mesothelioma codes. Including the accidental poisoning codes E866.4 (ICD-9) and X49 (ICD-10), covering poisoning by 'unspecified' and 'other' causes, which appear to have been used as coding surrogates for mesothelioma when asbestos exposure was explicitly mentioned in deaths suggestive of a mesothelioma, and which are recorded as the underlying cause of death in 4-7% of mesotheliomas, may improve the mesothelioma detection rate in future epidemiological studies.