Questionnaire study of the lung cancer risk from radon in homes

First time measurements of radon levels in homes, self-selected by occupants, were accompanied by questionnaires asking about people who had lived in the house and died of cancer. Responses were received before measurement results were available. The ratio of lung cancer deaths to deaths from other cancers was plotted as a function of radon level in the house. The results seem to indicate essentially no dependence of this ratio on radon exposure, in sharp contrast to the strong increase with increasing radon exposure predicted by the theory which should have been easily observable. Possible explanations of the discrepancy are discussed. Use of two alternative control groups increases the discrepancy with theory. This work has no connection with the recent test of the linear nonthreshold theory in the low dose region by the same author.