[Skip to Content]
[Skip to Content Landing]
September 10, 1982

Radiation and Human Health

JAMA. 1982;248(10):1179-1180. doi:10.1001/jama.1982.03330100019015

To the Editor.—  Because of the favorable review in a recent issue of The Journal (1982;247:1637) of John Gofman's book Radiation and Human Health,1 I purchased a copy, only to discover that it is an antinuclear diatribe rather than a serious scientific work. Its innuendos about the integrity and competence of scientists who disagree with the author serve as flags to stimulate closer scrutiny. The "innovative" method proposed for calculating the risk of cancer induced by low-dose radiation begins with doubtful assumptions and multiplies manyfold the error of values observed in epidemiologic studies. The results are then tabulated to three significant figures for use by compensation lawyers and policymakers. Such statistical esoterica as confounding variables, confidence limits, and goodness-of-fit tests are introduced, if at all, only to explain why actual observations do not confirm Gofman's predictions. While the reviewer commented on the large number of references, he did not