"Low Level
Radiation Health Effects: Compiling  the Data"

Revision 1
March 19, 1998
by Radiation, Science, and Health, Inc.,
Edited by J. Muckerheide

1.2.2
Ocupational

1.2.2.1
Weapons Plant Workers

References

"The International Agency for Research on Cancer (IARC) Study Group on Cancer Risk among Nuclear Industry Workers performed an independent study of the combined data... from the United Kingdom, United States, and Canada. This study, involving more than 95,000 individuals, is the most extensive study to date for cancer mortality risk associated with protracted exposure to low levels of radiation. The distribution of cumulative doses received by the study population... was rather skewed in that 60 per cent of the cohort received doses of 1 rem or less and only about 1 per cent received doses of 50 rem or more. All doses are assumed to be at low dose rates. Excluded from the study were 19 workers who received greater than 25 rem in a single year.

"The excess relative risk (ERR) for all cancers, excluding leukemia, was reported to be negative at -7 x 104 per rem, with a 90-per-cent confidence interval from -39 x 104 to +30 x 104 per rem, which is consistent with zero risk. For leukemia, excluding chronic lymphocytic (CL) leukemia, which is thought not to be induced by radiation, the excess relative risk (ERR) was reported to be positive at 2.2 x 102 per rem, with a 90-per-cent confidence interval from 0.1 x 102 to 5.7 x 102 per rem, which is barely significant (the 95-per-cent confidence interval overlaps zero risk). Taking into account the range of uncertainties, the quoted results for non-CL leukemia are consistent with those obtained from a linear extrapolation of the high-dose, high-dose-rate data from the atomic-bomb survivors, and with a low-dose, low-dose-rate effectiveness multiplier of one-half, though the range of uncertainty of this multiplier is quite large (0.027-1.7)."

"The authors of this study give the relative risk (RR) for all leukemias except CL leukemia for I0-rem exposure as 1.22, which means that a person exposed to 10 rem of low-LET radiation over a working lifespan is 22 per cent more likely to die from non-CL leukemia than a similar, but unexposed worker. This statement would lead the casual reader to infer that the data at dose levels around 10 rem actually show an effect. However, an examination of the data presented for all non-CL leukemia mortality in 7 dose intervals, the last being greater than 40 rem, shows that for only the last dose interval is a positive effect observed (Figure 7).

Figure 7

"The risk factors quoted above are found by forcing a linear fit to all of the data; however, if the one data point for doses above 40 rem is excluded, the remaining 6 data points for doses below 40 rem show a flat response with dose (that is, no increasing risk with dose). The range of uncertainties in the final results would also seem to allow either a sublinear or superlinear dose response at low doses, in addition to the assumed linear response. This very large and careful study of nuclear workers does not provide a definitive resolution of the problem of determining the dose response at low doses (less than 20 rem). However, this study does provide valuable new information at low dose rates."