Main Objectives

The main objectives of this second phase are:

1. to extend the evaluation of the health effects of employment in the nuclear industry in Oak Ridge to include Oak Ridge workers who were omitted from earlier studies;
2. to compare the mortality experience among the facilities;
3. to carry out dose-response analyses for those individuals who were potentially exposed to external radiation; and
4. to address methodological problems that occur when individuals employed at more than one facility are included in the analyses.

Three types of analyses were done for this report. First, cause-specific mortality for the Oak Ridge cohort was compared with the U. S. population by race and gender groups.

In the second type of analyses, Poisson regression methods were used to evaluate the joint influence of several risk factors on cause-specific mortality. Observed and expected deaths were cross-classified by levels of facility of employment, duration of employment, socioeconomic status, and period of follow-up and a main- effects model was used to describe the joint effects of the four risk factors.

In the third type of analysis the goal was to evaluate the potential adverse health effects of low level occupational exposure to ionizing radiation. These analyses were limited to white males who were ever employed at X-10 or Y-12. A uniform approach to radiation exposure assessment (based on ordered exposure categories for internal and external radiation for each facility) was used, and cause-specific mortality patterns were evaluated for potential association with radiation exposure. Length of employment, socioeconomic status, birth year, and age are considered as covariates.

During the course of this study it was determined that external radiation doses at X-10 and Y-12 were underestimated for some individuals---before 1957 at X-10 and before 1961 at Y-12 when not all workers were monitored. A preliminary procedure for computing adjusted doses was developed by Watkins et al [2,5,3] and used to evaluate the bias that may result in dose-response coefficients as the result of this dose underestimation.