skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: An assessment of ecological and case-control methods for estimating lung cancer risk due to indoor radon

Conference ·
OSTI ID:45723
;  [1]
  1. Univ. of New Mexico Medical Center, Albuquerque, NM (United States)

Studies of underground miners indicate that indoor radon is an important cause of lung cancer. This finding has raised concern that exposure to radon also causes lung cancer in the general population. Epidemiological studies, including both case-control and ecological approaches, have directly addressed the risks of indoor residential radon; many more case-control studies are in progress. Ecological studies that associate lung-cancer rates with typical indoor radon levels in various geographic areas have not consistently shown positive associations. The results of purportedly negative ecological studies have been used as a basis for questioning the hazards of indoor radon exposure. Because of potentially serious methodologic flaws for testing hypotheses, we examined the ecological method as a tool for assessing lung-cancer risk from indoor radon exposure. We developed a simulation approach that utilizes the Environmental Protection Agency (EPA) radon survey data to assign exposures to individuals within counties. Using the computer-generated data, we compared risk estimates obtained by ecological regression methods with those obtained from other regression methods and with the {open_quotes}true{close_quotes} risks used to generate the data. For many of these simulations, the ecological models, while fitting the summary data well, gave risk estimates that differed considerably from the true risks. For some models, the risk estimates were negatively correlated with exposure, although the assumed relationship was positive. Attempts to improve the ecological models by adding smoking variables, including interaction terms, did not always improve the estimates of risk, which are easily affected by model misspecification. Because exposure situations used in the simulations are realistic, our results show that ecological methods may not accurately estimate the lung-cancer risk associated with indoor radon exposure.

DOE Contract Number:
FG04-89ER60852
OSTI ID:
45723
Report Number(s):
CONF-901010-Pt.2; TRN: 95:003335-0028
Resource Relation:
Conference: 29. Hanford symposium on health and the environment: indoor radon and lung cancer--reality or myth, Richland, WA (United States), 15-19 Oct 1990; Other Information: PBD: 1992; Related Information: Is Part Of Indoor radon and lung cancer: Reality or myth? Part 2; Cross, F.T. [ed.]; PB: 599 p.
Country of Publication:
United States
Language:
English