Since late 1992, over 100 building complexes in Taiwan, including both public and private schools, and 1,000 apartments have been identified as emitting elevated levels of gamma-radiation. These high levels of gamma- radiation have been traced to construction steel contaminated with 60Co. Accurate reconstruction of the radiation exposure dosage among residents is complicated by the discovery of multiple radioactive sources within the living spaces and by the lack of comprehensive information about resident life-style and occupancy patterns within these contaminated spaces. The objective of this study was to evaluate the sensitivity of current dose reconstruction approach employed in an epidemiological study for the health effects of these occupants. We apply a statistical method of local smoothing in dose rate estimation and examine factors that are closely associated with radiation exposure from multiple radioactive sources in the apartment. Two examples are used, a simulated measurement in a hypothetical room with three radioactive sources and a real apartment. Two examples are used, a simulated measurement in a hypothetical room with three radioactive sources and a real apartment in Ming-Shan Villa, one of the contaminated buildings. The simulated and estimated means are compared along 5-10 selected points of measurement: by local smoothing approach, with the furniture-adjusted space, and with the occupancy time-weighted mean. We found that the local smoothing approach came much closer to theoretical values. The local smoothing approach may serve as a refined method of radiation dose distribution modeling in exposure estimation. Before environmental exposure assessment, 'highly occupied zones' (HOZs) in the contaminated spaces must be identified. Estimates of the time spent in these HOZs are essential to obtain accurate dosage values. These results will facilitate a more accurate dose reconstruction in the assessment of residential exposure in apartments with elevated levels of radioactivity.
ASJC Scopus subject areas