中国阳江高本底地区居民恶性肿瘤死亡研究(1979～1998)

目的本研究旨在探索平均年有效剂量为6.4 mSv的阳江天然放射性高本底辐射的致癌危险,为小剂量电离辐射致癌危害概率估计和预测提供对人类的直接观察资料.方法癌症死亡资料收集采用前瞻性调查方法,1979～1986年的资料从动态队列中收集,1987～1998年资料从固定队列获得.两组资料通过记录连接法合并.外照射个人剂量是根据辐射剂量仪测得的环境(室内、外)辐射水平,并考虑不同性别年龄的居留因子来估算.内照射个人剂量根据各类测量数据估算,不考虑性别年龄的差异.相对危险(RR)和超额相对危险系数(ERR/Sv)及95%置信区间(CI)采用Epicure中AMFIT的程序估算.结果通过对125 079人1979～1998年的观察,共累积观察了1992940人年,期间总共死亡12 444例,其中癌症死亡1202例.整个高本底地区全癌死亡的相对危险RR=1.00(95%CI,0.89～1.14),与对照地区相比无统计学差异.高本底地区高、中、低剂量组全癌死亡相对危险的计算结果,均未显示其与对照组有统计学差异,也未发现其与剂量呈一致性的变化趋势.调查还表明,不同地域、不同观察周期以及诊断水平等因素对全癌相对危险无统计学意义的影响.部位别癌症死亡相对危险分析表明,除食管癌高本底地区高于对照地区且具有统计学意义外,其他的癌两地区均无统计学意义差异.不同剂量组部位别癌症死亡相对危险的比较分析,以及部位别癌症死亡与估算的个人终生累积内外照射剂量的剂量-效应关系分析结果,均未发现任何癌症(含食管癌)死亡与剂量的一致性变化关系,趋势检验P值均＞0.05.但高本底地区剂量范围等于或大于400 mSv剂量组的居民中,肝癌死亡明显低于0～199 mSv剂量组的居民,RR=0.31(95%CI,0.13～0.66),差异有统计学意义.全部实体癌的超额相对危险系数(ERR/Sv),在整个高本底地区估算为-0.06(95%CI,-0.60～0.67).结论高本底地区全部癌症死亡与对照地区相比无统计学意义差异,也未发现高本底地区有辐射相关的部位别癌症死亡的增加.

Study of cancer mortality among inhabitants in the high background radiation area of Yangjiang, China (1979-1998)

Objective The objective of the study is to explore cancer risk associated with the low-level radiation exposure of average annual effective dose of 6. 4 mSv occurring in the high background radiation area(HBRA) in Yangjiang of Guangdong Province, China. Methods The data of cancer mortality for period of 1979-1986 were collected in prospective follow-up survey of dynamic populations and those of 1987-1998 were obtained from a fixed cohort in the same method. Record linkage was used to combine the two data sets. The external dose received by the inhabitants was estimated on the basis of hamlet-specific environmental doses and sex-age-specific occupancy factors, and the fixed values of internal dose regardless of sex and age was assumed. Relative risk (RR) and excess relative risk coefficient (ERR/Sv) and their 95% confidence interval (CI) were calculated using the AMFIT in Epicure. Results During the period of 1979-1998, the follow-up for 125 079 subjects accumulated 1 992 940 person-years and identified 12 444 deaths, including 1202 cancer deaths. The RR of mortality of all cancers in whole HBRA was 1. 00 (95% CI, 0. 89-1. 14) showing no statistical difference between the HBRA and the control area (CA). The RRs of all cancers in low, intermediate and high dose groups in the HBRA did not show statistical difference from the control group, either. The region (Yangdong and Yangxi) and the follow-up period(1979-1986 and 1987-1998) did not evidently modify the results, and the restriction of cancer cases to those with pathology information made no statistically significant change in RRs. In site-specific cancer analysis, only the RR of esophageal cancer showed statistical excess in the HBRA, the others did not show statistical difference between the HBRA and the CA. The comparison of RRs of site-specific cancer mortality among different dose groups and the dose-response analysis of site-specific cancer mortality related to cumulative individual lifetime dose did not give any monotonic trend, the P-values were larger than 0. 05 for all, including esophageal cancer. Interestingly to note that, the mortality of liver cancer, which is virus-related and the most common cancer in the studied areas, showed a statistical decrease among the HBRA inhabitants with the dose range of 400 mSv or higher compared to those of 0-199 mSv, RR=0. 31 (95%CI, 0. 13-0. 66). The ERR/Sv for all solid cancers associated with cumulative individual lifetime dose was estimated to be -0. 06 (95%CI, -0. 60-0. 67). Conclusions The mortality of all cancers showed no statistical difference between the HBRA and the CA;any radiation-related excess of site-specific cancer in HBRA could not be found.