近藤 宗平 近畿大学, 原子力研究所, 特別研究員
田ノ岡 宏 国立がんセンター研究所, 客員研究員
KONDO Sohei Atomic Energy Research Institute, Kinki University, Special Scientist
TANOOKA Hiroshi National Cancer Denter Research Institute, Visiting Scientist
|Budget Amount *help
¥3,800,000 (Direct Cost : ¥3,800,000)
Fiscal Year 1998 : ¥900,000 (Direct Cost : ¥900,000)
Fiscal Year 1997 : ¥2,900,000 (Direct Cost : ¥2,900,000)
To investigate the association between indoor radon exposure and lung cancer risk, we conducted a case-control study in Misasa Town, Tottori Prefecture, Japan. Source population was defined to be all residents in Misasa Town who aged 40 years or more in 1976 (N=4,331). Of these, 63 deceased lung cancer cases were identified by death certificate from Jan 1, 1976 to May 31, 1996. Family members were remained within the town for 51 cases and agreement for indoor radon measurement was obtained for 30 cases. Five controls per one case who were alive at the time of death of the case were selected from the source population, matched by sex and age, and finally agreement for the measurement was obtained for 36 controls, Indoor radon was measured by alpha track detector (Radtrack, Nagase-Landauer Co. Ltd.) for one year (6 months x 2). Adjusted odds ratios were estimated by Mantel-Haenszel method.
Measurements were completed for 28 cases and 36 controls (2 cases rejected after 1ィイD1stィエD1 measurement for 6 months), and mean indoor radon concentration was 45.6 and 50.6 Bq/mィイD13ィエD1, respectively. For males (26 cases and 32 controls). age and smoking adjusted odds ratio for lung cancer among those exposed to indoor radon level 25-49, 50-99 and 100 Bq/mィイD13ィエD1 or more was, 0.93 (95% confidence interval : 0.21-4.04), 1.33 (0.06-32.6) and 0.43 (0.04-4.82), respectively, when level 0-24 Bq/mィイD13ィエD1 was defined as a reference.
So far, several case-control studies have been conducted in Europe, North America and China. Meta-analysis summarizing these results estimated that indoor radon level 1 50 Bq/mィイD13ィエD1 so will result in odds ratio of 1.14, and this small magnitude of effect would be meaningful considering large impact for general population. In this study, because of the wide confidence interval due to small sample size, we could not conclude either the risk exists or does not exist.