Each of the proposed activities outlined below addresses one or more
of the NIOSH Energy-Related Health Research Agenda objectives and has
been approved by the DHHS Advisory Committee on Energy Related
Epidemiologic Research (ACERER). These research objectives are not a
comprehensive list of research opportunities or knowledge gaps for
occupational hazards at DOE sites. A new Advisory Committee to replace
ACERER is being considered, and NIOSH will work with the new committee
to refine and prioritize the research objectives that will guide future
energy-related research.
Establish a cohort from the more than 500,000 civilian nuclear
power workers who have been primarily exposed to external radiation
with estimated cumulative dose averages between 2.2 and 4.2 rem. This
cohort would also present opportunities to contrast neutron
exposures. Records have been collected on this cohort in support of
the cooperative agreement with IARC on the International
Collaborative Study of Nuclear Industry Workers.
Several epidemiologic studies have suggested that time-related
variables, including age at time of radiation exposure may have a
significant effect on risk of cancer. Methods for statistically
modeling this biological effect are not well defined. Following a
literature review, a statement of work was prepared to create
age-based analysis files and propose analytic approaches to
these issues.
Work environments with substantial neutron exposure records are
being evaluated to develop methods for neutron exposure estimation
and the appropriate incorporation of these estimates in epidemiologic
studies. Additionally, numerous reports indicate that substantial
uncertainties in estimating neutron exposure may have caused bias in
the neutron dose estimates. Currently, risk estimates for neutron
exposures are derived from animal studies.
Documents that describe historical dosimetry practices at DOE
facilities will be identified, collected, and summarized to better
describe uncertainties in exposure estimates in epidemiology studies
at those facilities. A list of key contacts with knowledge of
historical dosimetry practices at each site is being developed.
Decontamination and monitoring of a truck. Idaho National
Engineering and Environmental Laboratory, Idaho Falls, Idaho. 1975.
Photo courtesy of the U.S. Department of Energy.
Summary:
Phase I of this research project is described in Ongoing
Research. If the analysis of Phase I information indicates a need for
further evaluation, Phase II would include visits to collect
additional site information and develop hypotheses for further study
based on priorities of hazards and exposures. Phase III would include
studies to test the hypotheses.
This facility engaged in operations with potential exposures to
polonium-210, plutonium-238, and tritium. A mortality study through
1979 showed elevated lung cancer mortality in workers employed from
1943-1959 and a significant dose-response relationship between
plutonium-238 exposure and lymphopoietic/hematopoietic cancers and
leukemia. Because of these suggestive results, the health effects of
tritium, polonium, and external radiation along with potential
chemical exposures at the Mound plant should be further investigated.
An update would allow an additional 20 years of followup and use of
validated bioassay.
The precision of occupational epidemiology studies will be
improved by characterizing multiple exposure conditions. The
potential effect of medical and off-site exposures to ionizing
radiation in occupational epidemiologic studies has never been
evaluated. Risk estimates from studies without characterization of
these exposures could be biased. At K-25, an investigation has
found that routine fluoroscopic chest x-ray examinations in the 1940s
and 1950s may have resulted in substantial whole-body
equivalent radiation exposures. This study will determine the
feasibility of using medical and off-site records for workers and, if
feasible, whole-body doses would be estimated using information from
these additional sources of radiation.
Mortality studies at Los Alamos, Oak Ridge, Y-12, and Rocky
Flats have found nonsignificant excesses in brain cancer.
Individually, these studies lacked sufficient power to evaluate
exposure response associations because of the rarity of brain cancer
and the relatively small size of the cohorts. Preliminary feasibility
analysis indicates that sufficient brain cancer cases exist across
the DOE complex to support a multisite case-control study that would
evaluate any association of the disease with chemical and ionizing
radiation exposures.
Uranium and plutonium have well known tumorigenic potential.
Worker cohort studies at sites with potential uranium or plutonium
exposures have not demonstrated significant elevated risk for bone
cancer primarily because of low statistical power to detect an
excess. Bone cancer deaths from combined DOE cohorts should be
reviewed to determine whether a multisite study is warranted where
uranium or plutonium was present in the work environment. This study
would provide the most sensitive evaluation of excess bone cancer in
nuclear workers possible.
When actual monitoring data on individual workers are absent,
exposures may be estimated from facility, building, or job
information through exposure matrices. Because of the extensive
individual external radiation monitoring information at DOE sites,
the reliability of grouped estimates for chemicals or internal
sources may be evaluated when they are derived under various
conditions from exposure matrices. A sensitivity analysis will be
performed to determine "how much data is enough" to construct a
reasonably accurate job exposure matrix. This study will: (1)
identify a set of health physics monitoring data with known work
histories and tasks; (2) use programming to remove various portions
of the data (e.g., collapse job titles and reduce sample sizes within
jobs) according to preset criteria; (3) determine indicators of
instability or unreliability using various modeling techniques; and
(4) determine the level of uncertainty of the estimates at each level
of data completeness.