Methods for analyzing combined data from studies of workers exposed to low doses of radiation

Epidemiologic studies of workers exposed occupationally to protracted low doses of radiation provide a direct assessment of health effects resulting from such exposure and thus supplement information provided by studies of populations exposed at high doses of radiation and high dose rates. Analyses based on combined data from several studies can be expected to provide a more thorough assessment of low dose occupational studies and more precise risk estimates than can be obtained from any single study. Statistical methods for conducting such combined analyses are discussed, and different approaches, such as basing analyses on various levels of aggregation of exposure data, are compared and evaluated. Emphasis is given to methods for obtaining risk estimates and confidence limits that can be appropriately compared with estimates that form the basis for current radiation protection standards; these estimates have been obtained through extrapolation from high dose data. Methods are illustrated using combined data on workers at three US Department of Energy facilities: the Hanford Site, Richland, Washington; the Oak Ridge National Laboratory, Oak Ridge, Tennessee; and the Rocky Flats Nuclear Weapons Plant, Denver, Colorado.