Hypothesis testing and the choice of the dose–response model

The shape of the dose–response curve for exposure to ionising radiation is probably one of the most contentious issues in toxicology. The initial assumption was that there was a threshold to the appearance of a health detriment, including cancer, with the so-called linear-no-threshold (LNT) hypothesis first being introduced in the early 1960s. Since that time a number of models have been suggested, and present work in health physics, toxicology and epidemiology is concerned with questions about both the shape of the dose–response curve and whether or not a threshold exists. This paper presents an analysis of the robustness of the LNT hypothesis from a philosophy of science standpoint—arguing that claims about dose–response curve need to pay more attention to the assumptions and auxillary hypotheses behind choices, and that further mechanistic studies are required to unravel the effects of exposure to ionising radiation. It suggests that whereas LNT falls short of the requirements for a good scientific hypothesis, it is a reasonable model for regulating the carcinogenic and hereditary effects of radiation exposure.