Skinner J, Mee T J, Blackwell R P, Maslanyj M P, Simpson J, Allen S G, Day N E, Cheng K K, Gilman E, Williams D, Cartwright R, Craft A, Birch J M, Eden O B, McKinney P A, Deacon J, Peto J, Beral V, Roman E, Elwood P, Alexander F E, Mott M, Chilvers C E D, Muir K, Doll R, Taylor C M, Greaves M, Goodhead D, Fry F A, Adams G, Law G
Strangeways Research Laboratory, University of Cambridge, Wort's Causeway, Cambridge CB1 8RN, UK.
Br J Cancer. 2002 Nov 18;87(11):1257-66. doi: 10.1038/sj.bjc.6600602.
The United Kingdom Childhood Cancer Study, a population-based case-control study covering the whole of Great Britain, incorporated a pilot study measuring electric fields. Measurements were made in the homes of 473 children who were diagnosed with a malignant neoplasm between 1992 and 1996 and who were aged 0-14 at diagnosis, together with 453 controls matched on age, sex and geographical location. Exposure assessments comprised resultant spot measurements in the child's bedroom and the family living-room. Temporal stability of bedroom fields was investigated through continuous logging of the 48-h vertical component at the child's bedside supported by repeat spot measurements. The principal exposure metric used was the mean of the pillow and bed centre measurements. For the 273 cases and 276 controls with fully validated measures, comparing those with a measured electric field exposure >/=20 V m(-1) to those in a reference category of exposure <10 V m(-1), odds ratios of 1.31 (95% confidence interval 0.68-2.54) for acute lymphoblastic leukaemia, 1.32 (95% confidence interval 0.73-2.39) for total leukaemia, 2.12 (95% confidence interval 0.78-5.78) for central nervous system cancers and 1.26 (95% confidence interval 0.77-2.07) for all malignancies were obtained. When considering the 426 cases and 419 controls with no invalid measures, the corresponding odds ratios were 0.86 (95% confidence interval 0.49-1.51) for acute lymphoblastic leukaemia, 0.93 (95% confidence interval 0.56-1.54) for total leukaemia, 1.43 (95% confidence interval 0.68-3.02) for central nervous system cancers and 0.90 (95% confidence interval 0.59-1.35) for all malignancies. With exposure modelled as a continuous variable, odds ratios for an increase in the principal metric of 10 V m(-1) were close to unity for all disease categories, never differing significantly from one.
英国儿童癌症研究是一项基于人群的病例对照研究，覆盖了整个大不列颠地区，其中纳入了一项测量电场的试点研究。对1992年至1996年间被诊断患有恶性肿瘤且诊断时年龄在0至14岁的473名儿童的家中进行了测量，同时选取了453名在年龄、性别和地理位置上匹配的对照。暴露评估包括在儿童卧室和家庭客厅进行的即时测量。通过在儿童床边连续记录48小时的垂直分量，并辅以重复的即时测量，对卧室电场的时间稳定性进行了研究。使用的主要暴露指标是枕头和床中心测量值的平均值。对于273例病例和276名对照的测量结果完全有效的情况，将电场暴露测量值≥20 V m⁻¹的人群与暴露参考类别<10 V m⁻¹的人群进行比较，急性淋巴细胞白血病的比值比为1.31（95%置信区间0.68 - 2.54），总白血病为1.32（95%置信区间0.73 - 2.39），中枢神经系统癌症为2.12（95%置信区间0.78 - 5.78），所有恶性肿瘤为1.26（95%置信区间0.77 - 2.07）。当考虑426例病例和419名对照且测量无无效值时，急性淋巴细胞白血病的相应比值比为0.86（95%置信区间0.49 - 1.51），总白血病为0.93（95%置信区间0.56 - 1.54），中枢神经系统癌症为1.43（95%置信区间0.68 - 3.02），所有恶性肿瘤为0.90（95%置信区间0.59 - 1.35）。将暴露作为连续变量进行建模时，主要指标每增加10 V m⁻¹，所有疾病类别的比值比都接近1，与1无显著差异。
