Survivorship Models For Lung Cancer Mortality In Uranium Miners – Is Cumulative Dose An Appropriate Measure Of Exposure?

A great amount of work in recent years has been devoted to refining older survivorship models, e.g. life tables, and developing newer ones, e.g. Cox regression. In this paper, these models are applied to data from an epidemiological study of uranium miners in four western states conducted by the U.S. Public Health Service. One of the principal problems addressed in this paper concerns the appropriateness of using cumulative dose as the measure of radiation exposure. An alternative is considered that incorporates dose rate into the model. This procedure does not implicity assume that equal cumulative doses produce the same lung cancer mortality, but instead considers the amount of time in which each miner's dose was accumulated. Results of the statistical analysis indicate that dose rate is a significant factor at higher cumulative dose levels, with lower dose rate apparently more damaging than higher dose rate. The shape of the dose response curve is curvilinear at higher dose levels. No interaction was found between smoking and radiation exposure on a multiplicative or relative scale.