Epidemiologic studies of the respiratory effects of uranium have involved miners and workers in uranium-processing plants (see Table 4-1). Their results are difficult to interpret because of workers’ coexposure to other respiratory toxicants, the grouping of multiple diseases, and inaccuracies in the coding of death certificates for nonmalignant respiratory diseases. Some of the human studies are described briefly below.
A study of workers at the Naval Products Division of the United Nuclear Corporation, a nuclear-fuels fabricating company, determined standardized mortality ratios (SMRs) and incidence ratios for employees (Hadjimichael et al. 1983). The SMR for all causes in industrial male workers was significantly lower than expected, but there was an excess of deaths due to obstructive pulmonary disease. Of the six people who died from obstructive pulmonary disease, five had emphysema, but smoking information on four of the five was not available. Because emphysema can be caused by smoking, the incomplete information on smoking prevented adequate interpretation of excess deaths.
In another study, 1,484 men employed in uranium mills in the Colorado Plateau were evaluated (Pinkerton et al. 2004). The study determined a significant increase in mortality from nonmalignant respiratory disease but identified several limitations, including low cohort size, little power to detect a moderately increased risk of some outcomes, inability to estimate individual exposures, and lack of smoking data. Furthermore, positive trends with employment duration were not observed.
Other studies of workers at uranium facilities did not find an association between nonmalignant pulmonary diseases and mortality. For example, Dupree-Ellis et al. (2000) compared mortality in 2,514 workers employed during 1942-1966 at a uranium-processing plant with overall U.S. mortality. They reported an SMR of 0.90 for all causes of death and 1.05 for all cancers. The SMR for respiratory diseases was 0.80. A retrospective cohort mortality study of workers at a facility for production of nuclear fuel (Cragle et al. 1988) found significantly fewer deaths in many categories of disease, including all respiratory diseases. Polednak and Frome (1981) described mortality in a cohort of 18,869 men employed at a uranium conversion and enrichment plant and reported that the causes of particular interest, including respiratory diseases, did not exhibit high SMRs.
Lung-cancer mortality has been estimated in a number of cohort studies that included nearly 110,000 uranium-processing workers (see Chapter 6 for discussion); nearly all the studies had null results. A nested case-control study based on the four largest U.S. cohorts did not find an exposure-response relationship. The few positive results, when combined with uncertainties due to lack of smoking data in the studies, mean, however, that the possibility of associations cannot be dismissed.