Radiation risk estimation.

Estimation of the risk of cancer following exposure to ionizing radiation remains largely empirical, and the models used to adduce risk incorporate few, if any, of the advances in molecular biology of the past decade or so. These facts compromise the estimation of risk where the epidemiological data are weakest, namely, at low doses and dose rates. Without a better understanding of the molecular and cellular events ionizing radiation initiates or promotes, it seems unlikely that this situation will improve. Nor will the situation improve without further attention to the identification and quantitative estimation of the effects of those host and environmental factors that enhance or attenuate risk.