0311 Development of a predictive model for estimating gamma radiation exposures among Ontario uranium miners0311 Development of a predictive model for estimating gamma radiation exposures among Ontario uranium miners
基于1981-1985年安大略省铀矿工人的数据，用多元线性回归开发了一个预测伽马辐射暴露的模型，发现辐射暴露与氡暴露、工作年限、暴露年份和矿石产量显著相关，但模型解释力中等。
<h3>Objectives</h3> The objective of this study is to develop and validate a predictive model for estimating gamma radiation exposure for miners working in uranium mines between 1981 and 1985. <h3>Method</h3> The dose prediction model was developed and validated using multiple linear regression. To aid in model development, 70% random sample of workers were used in the model development (i.e., Training Sample) while the remainder 30% (i.e., Test Sample) was used to determine model performance. A stepwise approach was used to select variables into the model. Criteria for retaining the variables in the model was based on a p-value of <=0.15. Model fit was assessed using adjusted R-square. Co-linearity was determined by the magnitude of the variance inflation factor (VIF). Variables with VIF greater than 3.0 were removed from the model. In addition, SAS procedure ROBUSTREG was used to minimise the effects of outliers and high leverage in order to provide resistant (stable) results in the presence of outliers and high leverage. <h3>Results</h3> Based on 8949 employments records, dosimetric measurements of gamma radiation were significantly correlated with radon exposure (r = 0.499), duration of employment (r = 0.429), year of exposure (r = 0.239), and ore production (r = 0.230). Age was inversely related to gamma dose. Regression analysis showed that individual dosimetric readings can be modestly predicted by individual work history and geological characteristics of Ontario uranium mines (p < 0.001, R2 = 0.374). Additional sources of variation are likely related to individual variability that could not be accounted for in this ecological assessment. <h3>Conclusions</h3> Reconstructed gamma dose provides modest agreement with individual dosimetric readings.