Workers on an ethanol unit which used sulfuric acid in strong concentrations at a large refinery and chemical plant in Baton Rouge, Louisiana were reported, in 1979, at excess risk for upper respiratory cancer. The carcinogen implicated by indirect evidence was diethyl sulfate. However, with the continued use of sulfuric acid in the same plant, and with additional cases not attributable to the ethanol process, the hypothesis of an association between sulfuric acid exposure and upper respiratory cancer was tested. Each of 50 confirmed cases of upper respiratory cancer diagnosed between 1944 and 1980, was matched to at least three controls on sex, race, age, date of initial employment, and duration of employment. Thirty-four of the 50 cases were laryngeal cancers. Data were obtained from existing plant records. Retrospective estimates of exposure were made without regard to case or control status. Findings from conditional logistic regression techniques were supported by other statistical methods. Among workers classified as potentially highly exposed, four-fold relative risks for all upper respiratory cancer sites combined were exceeded by the relative risk for laryngeal cancer specifically. Exposure-response and consistency across various comparisons after controlling statistically for tobacco-use, alcoholism and other previously implicated risk factors, suggest increased cancer risk with higher exposure.

Along with our submission to the National Death Index (NDI) of a cohort of more than 23,000 petrochemical and refinery workers, we included 1,449 known U.S. deaths to determine the completeness of death ascertainment through the NDI. A number of factors that may affect follow-up were examined including sex, race, age and reporting area. Overall, NDI detected 97.1% of known deaths. Follow-up was slightly better for males (97.2%) than for females (92.0%) and was significantly better for whites (97.6%) than for nonwhites (92.0%). Analyses by reporting area showed very complete follow-up from all locations (93.4%-100%) except for the New York City area (71.4%). These findings indicate that NDI is an extremely useful source for vital status follow-up, though follow-up may be somewhat less complete for certain subgroups of an occupational study cohort.

No abstract

A cohort study covering the eight year period, 1970-1977, identified eight deaths attributed to brain cancer among employees at a refinery/chemical plant. Five of the cases were primary brain cancers and three were either metastatic from another site or the site was undetermined. A ninth case of brain tumor unspecified was identified. Five primary brain cancers would have been expected, based on national statistics. Some issues and dilemmas related to such a cohort study were discussed.

No abstract