It is assumed that cancer risk at doses < 100 mGy (< 100 mSv) are also linear with dose, and that this risk declines as the dose reaches zero, without a threshold (ie, the linear, no threshold [LNT] model). Controversy exists regarding cancer risk at < 100 mGy.2 Results of laboratory studies often show minimal or no biologic effects at a dose of < 100 mGy. Moreover, the LNT model for stochastic effects contrasts with the dose-effect relationship for deterministic effects in humans in which a threshold dose must be reached for deterministic injuries such as bone marrow depression, desquamation, mucositis, pneumonitis, pulmonary fibrosis, pericardial effusion and tamponade, esophagitis and stricture, hepatic fibrosis, venoocclussive disease, proctitis, cystitis, nephritis, and so forth. Therefore, the LNT hypothesis has not been validated by either experimental data generated in laboratory studies or observational data generated in epidemiologic studies. In fact, a dose threshold model (with a threshold value of 40 mGy, 95% CI, < 0-85 mGy) for cancer incidence may fit the data as well as but no better than the LNT model.3,4 If this threshold is applied, exposure to doses of < 40 mGy would not result in a higher cancer risk.