The role of confounding factors in a radon epidemiological study

Abstract

Objective: A simulation of a large-scale epidemiological case-control study to identify the relationship between exposure to radon and lung cancer in the presence of factors that distort the results of the assessment of exposure to radon in homes. Materials and Methods: Analysis of sources of uncertainties arising during radon epidemiologic case-control studies. Evaluation of the uncertainties caused by the errors of the measurements of the long-term variations in the radon concentration, exposure to radon in other places of the human habitat, except dwellings, etc. Simulation by Monte Carlo technique of radon epidemiologic study, comparable to the combined European radon study, and assessment of uncertainties, which affect the evaluation of dose-effect dependence. Results: The multiplicative error in the assessment of individual exposure based on the radon concentration is shown generally caused by the combined effect of long-term variations of the radon concentration and the differences in the levels of the radon concentration in living houses and other places of the human habitat. The logarithmic standard deviation of this errors σerr is from 0,70 to 0,90. The estimated value of this error is 2,0 times higher than the value used for correction of the results of the combined European radon study. It is shown that for the σerr <0,9 regression calibration technique, there is a possibility to make a full correction of uncertainty. Conclusion: Errors in the assessment of uncertainties of the radon exposure based on the radon concentration in the combine European radon case-control study has led to an underestimation of the relative risk of lung cancer incidence at least with a factor of 1,5.

Keywords

• radon
• lung cancer
• case-control studies
• uncertainties
• modeling