Adjusting Lung Cancer Risks for Temporal and Spatial Variations in Radon Concentration in Dwellings in Gansu Province, China

Abstract

Our recent study in Gansu Province, China reported an increasing risk of lung cancer with increasing residential radon concentration that was consistent with previous pooled analyses and with meta-analyses of other residential studies (Wang et al., Am. J. Epidemiol. 155, 554-564, 2002). Dosimetry used current radon measurements (1-year track-etch detectors) in homes to characterize concentrations for the previous 30 years, resulting in uncertainties in exposure and possibly reduced estimates of disease risk. We conducted a 3-year substudy in 55 houses to model the temporal and spatial variability in radon levels and to adjust estimates of radon risk. Temporal variation represented the single largest source of uncertainty, suggesting the usefulness of multi-year measurements to assess this variation; however, substantial residual variation remained unexplained. The uncertainty adjustment increased estimates of the excess odds ratio by 50-100%, suggesting that residential radon studies using similar dosimetry may also underestimate radon effects. These results have important implications for risk assessment.

Genetic variant in TP63 on locus 3q28 is associated with risk of lung adenocarcinoma among never-smoking females in Asia

“…Extensive research has been conducted on the environmental and genetic determinants of nonsmoking lung cancer throughout Asia (Lan et al 2000, 2002, 2004; Hosgood III et al 2007, 2008; Yang et al 2004; Xu et al 1989; Lubin et al 2005), where lung cancer rates among nonsmoking women in some regions are among the highest in the world (Mumford et al 1987; Lam 2005). The relatively high rates of lung cancer among nonsmoking females in Asia are partially attributed to known environmental risk factors, such as exposure to combustion products of indoor heating and cooking solid fuel and cooking oil fumes (Lubin et al 2005; Lan et al 2002; Hosgood III et al 2011). In-home coal used for heating and cooking has been classified as carcinogenic in humans (Straif et al 2006), and genetic variation has been shown to influence this association (Hosgood III et al 2007, 2008; Lan et al 2000).…”

Bayesian dose–response analysis for epidemiological studies with complex uncertainty in dose estimation

“…This assumes that the unknown true values of individual dose will vary at random about unbiased mean doses assigned to individual members of cohort subgroups and that the remaining error structure is completely Berkson . Some investigators have used the individual median rather than the individual mean dose in their regression model , while others have estimated the dose–response by treating dose uncertainties as mixtures of Berkson and random classical errors .…”

Nonparametric Methods for Solving the Berkson Errors-in-Variables Problem

“…(2001), Stram et al . (2002) or Lubin et al . (2005) for epidemiologic applications, where X represents the unobservable personal exposure to (or random dose of) a toxic substance and W represents the concentration of the substance that is measured at certain fixed locations.…”

“…Of course, there is a close analogy here with better-understood problems in nonparametric regression, where deterministic, equally spaced design produces estimators with properties that are similar to those when the design is random and uniformly distributed. However, there are many examples where the W s are genuine random variables, rather than deterministic quantities; see, for example Reeves et al (1998), Thomas et al (1999), Raaschou-Nielsen et al (2001), Stram et al (2002) or Lubin et al (2005) for epidemiologic applications, where X represents the unobservable personal exposure to (or random dose of) a toxic substance and W represents the concentration of the substance that is measured at certain fixed locations. See also Huwang and Huang (2000).…”