The proportion of thyroid cancers in the Japanese atomic bomb survivorsassociated with natural background radiation

Abstract

Generalised absolute and relative risk models (with adjustment to the excessabsolute risk for time since exposure and age at exposure, and with adjustment tothe excess relative risk for age at exposure) are fitted to the Japanese atomicbomb survivor thyroid cancer incidence data followed up over the period1958–87, taking account of natural background radiation. Thyroid cancersassociated with natural background radiation and atomic bomb radiation areoverwhelmingly accounted for by exposure at young ages. Over 50% of theexcess cases associated with either the atomic bomb radiation or naturalbackground radiation are linked to exposures under the age of 20, irrespectiveof the assumed risk model or natural background dose rate. The excessrisk is overwhelmingly concentrated among females, again irrespectiveof the assumed model or natural background dose rate. Depending onthe assumed natural background dose rate (in the range 0.5–2.0 mSv/year) between 17.3 and 32.0% of thethyroid cancer in this cohort may be associated with naturalbackground radiation if an absolute risk model applies; between 4.2and 17.1% of the thyroid cancers may be associated withnatural background radiation if the relative risk model applies. Theproportion of the thyroid tumours attributed to the atomic bombradiation is between 21.1 and 22.0% for the absolute riskmodel, and is between 18.7 and 19.1% for the relative riskmodel, in both cases irrespective of the assumed background radiationdose. In particular, these proportions are not very different fromthe proportions calculated when fitting models that do not takeaccount of natural background radiation, namely 22.0% for theabsolute risk model and 18.6% for the relative risk model.The proportion of thyroid cancers accounted for by natural backgroundradiation progressively increases with attained age, from0.3% of cancers among those under the age of 15 to30.5% for those over the age of 60, assuming that theabsolute risk model applies. There is a similar increase in thispercentage, if to a rather lower level (from 0.2 to 10.2%),assuming that the relative risk model applies.