Attributable risk for radiation in the presence of other risk factors

Two motivations for studying radiation risk are (1) to quantify the magnitude of risk as an aid to setting radiation protection standards and (2) to understand causality as an aid to assigning compensation for radiation exposed individuals whose disease or death may have been related to radiation exposure. Although it has long been known that radiation risk is modified by factors such as sex, age, and time, it is now apparent that radiation risk may also be modified by other risk factors, such as smoking, inflammation, genotype, and certain pathogens. Even apart from considerations of etiological interaction, the relative contribution of radiation to total burden of disease or death may depend on the level of background (spontaneous) risk of disease or death owing to those other factors if the joint effects do not multiply. Therefore, ignoring those other factors in assessing probability of causation for radiation (attributable fraction in epidemiological data) involves making a strong assumption about the joint effects. The concepts are discussed in detail and illustrated using results from studies on the Japanese atomic-bomb survivors.