Epidemiological studies of US nuclear weapons workers allow evaluation of the effects of low dose, low dose rate radiation exposures accrued over years or decades of employment. In the proposed study, we will examine a cohort of badge-monitored workers at the Oak Ridge National Laboratory (ORNL). The last mortality follow-up of this epidemiologically important cohort was conducted 18 years ago and considered workers hired between 1943 and 1972. Analyses of these data revealed a significant association between occupational radiation dose and all cancer mortality. We propose to expand this cohort to include workers hired between 1943 and 1984 and update mortality follow-up through 2007;based on life table projections, we expect 2.4 times the number deaths observed in the last mortality follow-up of the ORNL cohort. Extending follow-up of this cohort is important not just for statistical power, but also because many important questions in radiation research concern the long term effects of irradiation, particularly on slow-progressing malignant diseases, such as non-Hodgkin's lymphoma and chronic lymphocytic leukemia. The proposed analyses will commence with evaluation of time-since- exposure as a temporal modifier of radiation effects. In addition to using methods of exposure lagging and time-window analysis, we will apply innovative methods for kernel estimation of distributed lag models. Prior research on ORNL and other cohorts suggests that age-at-exposure is another important modifier of radiation- cancer associations. Extending follow-up of this cohort, and expanding the roster to include more recent hires, will allow us to better assess age-at-exposure as a temporal modifier of radiation effects. In addition to empirical models of age-at-exposure effects we will fit biologically-based multistage cancer models to these data. Non-radiological agents may also modify the effects of ionizing radiation. We propose two approaches to assessing non-radiological effect modifiers: one considering the full cohort;the other applying a nested case- control approach to assess modifiers of radiation dose- lung cancer mortality associations. In the full cohort, focusing on asbestos, benzene, and beryllium exposures, we will develop job-exposure-matrices for these hazards, informed by historical lists of asbestos and beryllium workers available at ORNL and recent work by NIOSH to assess benzene among ORNL workers. We will apply innovative methods to examine whether the temporal ordering of these non-radiological exposures impacts their joint effects. In the nested case-control study of lung cancer mortality, we will utilize medical records to assess modification or confounding by cigarette smoking as well as detailed industrial hygiene methods to assess asbestos and beryllium exposure potential. Study results for this USDOE cohort will be evaluated in relation to observations from studies of other DOE facilities;and, study data will be compiled in a manner that will facilitate future pooled analyses. In this way, the proposed work will substantially strengthen the available epidemiological information about low level radiation effects in USDOE cohorts. Public Health Relevance: The proposed study directly evaluates the long term effects of effects of low dose, low dose rate occupational radiation exposures, the subject of long debate and clear policy implications. This work is innovative in its questioning of an established paradigm regarding the absence of radiation effects on lymphatic cancers. We propose to novel models of time-related exposure effects to consider issues of latency and age at exposure, apply biologically-based carcinogenesis models, and investigate non-radiological exposures in work that will improve our understanding of the consequences of exposure to multiple chemical and physical agents in the DOE complex.
The proposed study directly evaluates the long term effects of effects of low dose, low dose rate occupational radiation exposures, the subject of long debate and clear policy implications. This work is innovative in its questioning of an established paradigm regarding the absence of radiation effects on lymphatic cancers. We propose to novel models of time-related exposure effects to consider issues of latency and age at exposure, apply biologically-based carcinogenesis models, and investigate non-radiological exposures in work that will improve our understanding of the consequences of exposure to multiple chemical and physical agents in the DOE complex.
| Richardson, David B; Keil, Alexander P; Tchetgen Tchetgen, Eric J et al. (2017) The Authors Respond. Epidemiology 28:e30-e31 |
| Richardson, David B; Keil, Alexander P; Tchetgen Tchetgen, Eric et al. (2015) Negative Control Outcomes and the Analysis of Standardized Mortality Ratios. Epidemiology 26:727-32 |
| Richardson, David B; Wing, Steve; Keil, Alexander et al. (2013) Mortality among workers at Oak Ridge National Laboratory. Am J Ind Med 56:725-32 |
| Richardson, David B; Cole, Stephen R; Chu, Haitao (2013) Random effects regression models for trends in standardised mortality ratios. Occup Environ Med 70:133-9 |
