Harvey Checkoway

Cancer Mortality and Solvent Exposures in the Rubber Industry

Some evidence suggests that solvent exposures to rubber industry workers may be associated with excess cancer mortality, but most studies of rubber workers lack information about specific chemical exposures. In one large rubber and tire-manufacturing plant, however, historical documents allowed a classification of jobs based on potential exposures to all solvents that were authorized for use in the plant. A case-control analysis of a 6678 member cohort compared the solvent exposure histories of a 20% age-stratified random sample of the cohort with those of cohort members who died during 1964-1973 from stomach cancer, respiratory system cancer, prostate cancer, lymphosarcoma, or lymphatic leukemia. Of these cancers, only lymphosarcoma and lymphatic leukemia showed significant positive associations with any of the potential solvent exposures. Lymphatic leukemia was especially strongly related to carbon tetrachloride (OR = 15.3, p less than .0001) and carbon disulfide (OR = 8.9, p = .0003). Lymphosarcoma showed similar, but weaker, associations with these two solvents. Benzene, a suspected carcinogen, was not significantly associated with any of the cancers.

A hematology survey of workers at a styrene-butadiene synthetic rubber manufacturing plant

A previous epidemiologic study of the U.S. rubber industry indicated that there has been an excess of leukemia and lymphoma mortality among hourly workers at one styrene-butadiene rubber manufacturing plant. This investigation was a combined industrial hygiene and hematology cross-sectional survey at the same plant. The objectives of the survey were to quantify exposure levels for styrene, butadiene, benzene and toluene, and to relate these levels to hematologic variation. Personal air samples and blood specimens were obtained from 157 production workers. All exposure levels for the four chemicals assayed were well below recommended standards. The higher mean styrene (13.67 ppm) and butadiene (20.03 ppm) concentrations were found in the Tank Farm area; in all other departments the mean levels for the four chemicals were less than 2 ppm. The Tank Farm workers had slightly lower levels of circulating erythrocytes, hemoglobin, platelets and neutrophils, and slightly higher mean corpuscular red cell volumes and neutrophil band counts than the other workers. Overall, in this population there was no pronounced evidence of hematologic abnormality, as determined from examination of peripheral blood.

Exponential models for analyses of time-related factors, illustrated with asbestos textile worker mortality data

In any study based on an occupational cohort, it is important to consider the variation in risk factors over time. Cumulative exposure is the most important time-related factor for exposure- response analyses, whereas other time-related factors such as age at risk, year at risk, and length of follow-up may be confounders and effect modifiers. This paper examines the family of exponential models which can be used for timerelated analyses of studies based on an occupational cohort. Analyses using Poisaon regression, the proportional hazards model, and the logistic model are presented, and their interrelationships explored. These models are illustrated with data from a cohort study of lung cancer mortality among asbestos textile plant workers. All three approaches yielded similar effect estimates. In particular, Poisson regression and the proportional hazards model yielded very similar findings, but Poisson regression has some conceptual and computational advantages.

Metabolic modeling of organ-specific doses to carcinogens as illustrated with alpha-radiation emitting radionuclides

Quantitative estimation of doses of carcinogens delivered to physiologic targets facilities specification of organ-specific dose-response functions. Typically, exposure measurements, such as air or water concentrations of carcinogens, are used as dose surrogates in epidemiologic studies. An illustrative exception to this usual situation is the case of airborne alpha radiation-emitting radionuclides, for which organ-specific doses can be derived. A metabolic modeling approach for estimating doses delivered to the lung, gastrointestinal tract and bone is described for three classes of radionuclides: soluble uranium, insoluble uranium and plutonium. The dose models are defined in terms of biological retention patterns and organ-specific depositions affinities. Application of the metabolic modeling approach is illustrated with a hypothetical example of excess lung cancer risk projection in a cohort of persons exposed to plutonium. Also, a simple example is presented to demonstrate how adherence to the metabolic model structure can avoid gross overestimation of doses in the case of multiple in vivo lung counting measurements taken in close temporal proximity following a large exposure intake.