was from the inhalation of uranium compounds, rather than from external gamma radiation. The process used to enrich the 235U content of uranium involved a mass spectrographic unit, with two enrichment stages that exposed workers to uranium dust. The first step enriched the 235U content to about 15 percent. Records indicate that many workers had high levels of exposure to uranium dust even by standards at that time (150 μg/m3 of ambient air). Air samples from the area involving the first enrichment stage included many readings higher than the standard; in the chemistry process area of the first stage, average levels were 250–500 μg/m3, whereas levels averaged 25–50 μg/m3 in the first-stage mass spectrographic area and in the second-stage area. After 1945, the high levels in the first-stage processing were reduced when UF6 began to be used as the starting material. Although UF6 is a gas and more soluble, it was immediately converted to less soluble oxides (UO4 and UO3) or to UF4. Although data on particle size are limited, the air contained small particles (<1 μm in diameter) that can be inhaled deeply into the alveoli of the lung and transported to other organs. According to the authors, probably only a few workers used respirators. Data on urinary uranium levels were too few to provide individual exposure estimates. However, about one-third of the workers had urinary uranium levels greater than 0.05 μg/ml, which corresponds approximately to the level expected at the maximum permissible concentration of uranium dust in the air. Thus, many employees, especially those involved in the early steps of uranium processing, were heavily exposed to uranium dust.
This study addressed the hypothesis that working up to several years in areas with high average levels of uranium dust was associated with increased mortality over a period 25–30 years after employment. The authors defined subgroups of workers by the department in which an individual worked and the average levels of uranium dust in these departments. The authors used an internal comparison group of employees with minimal exposure to uranium dust (e.g., workers in buildings where uranium was not processed) and an external comparison group of all U.S. white males. As noted above, deaths (reported by 1974) were ascertained through the SSA by record linkage; the authors estimated that they identified 94 percent of the deaths by this method. The authors used total and cause-specific death rates for U.S. white males, specific for age and calendar year, to compute expected numbers of deaths for the calculation of SMRs for subgroups of employees.
Oak Ridge nuclear materials fabrication workers (1948–1974) (Checkoway et al., 1988). This analysis by Checkoway and colleagues (1988) was based on a retrospective follow-up of 6,781 white men employed in the Oak Ridge nuclear materials fabrication plant (the Y-12 plant) between 1947 and 1979. The individuals included in this study cohort were hired after 1947, and therefore the cohort does not overlap with the workers studied by Polednak and Frome (1981).
Mortality follow-up was conducted primarily using SSA records, supplemented with other sources; vital status was determined for 96 percent of workers. In this facility, exposure to uranium was primarily as airborne dust resulting