Radon, smoking and HPV as lung cancer risk factors in ecological studies

Cohen's ecological analyses revealed negative correlation between the lung cancer mortality and average indoor radon concentration in the US counties, that contradicts to linear non-threshold (LNT) model and is inconsistent with results of case-control studies. The aim of this study was to analyze dependence between radon exposure and lung cancer mortality rate taking into account more complete data on smoking and new findings on association of the lung cancer with human papillomavirus (HPV) infection.

Information on the cancer rates in the US counties and Russian oblasts, smoking prevalence and indoor radon concentration was found in literature. The cervix cancer incidence rate was used as surrogate of the HPV infection prevalence. The analysis included calculation of the coefficients of linear dependence between radon exposure and lung cancer mortality rate with adjustment to smoking and HPV infection prevalence.

After adjustment for the most relevant data on smoking and HPV infection, correlation between the lung cancer mortality and indoor radon was found to be consistent with results of the case control studies.

Analysis of geographically aggregated data on the lung cancer mortality and radon concentration in dwellings with adjustment to the significant risk factors confirms both the linear non-threshold dependency and results obtained in studies with individual accounting for the smoking and radon.