Ecological relationship between disease-specific mortality risks and estimated exposure levels of UV-B in Japan

首先，为了在生态学研究中应用，开发了一个从气象数据估算地面UV-B的公式。估计值是从一个方程中得到的，该方程包含了日本各地不同地区的太阳辐射、臭氧总量、纬度和月份，与气象站的位置和气象数据相对应。基于UV-B的估计值和观测值之间的高度相关性，认为估算是有效的。其次，以相应地区的标准化死亡率(SMR)表示，评估了过去30年的区域UV-B估计值与所有癌症、白血病和肺癌的近期死亡风险之间的生态关联。1960年十年期间的估计UV-B值与1983-87年男性和女性所有死因死亡率的SMR之间存在微弱但显著的负相关(表明保护作用)。相比之下，男性和女性在1960年十年期间的UV-B估计值和1983-87年白血病的SMR之间存在微弱但显著的正相关(提示不良反应)。这样的关系被维持，而不考虑区域点的选择或仅针对女性所有癌症死亡率的UV-B估计值的周期。当纬度作为UV-B暴露的替代指标时，这种相关性在大小和显著水平上都更高。因此，区域UV-B水平可能在一定程度上解释了日本特定疾病发病率或死亡率的区域差异。这项研究将进一步考虑到区域点的选择和其他特定疾病发病率或死亡率的区域指数之间的关系等因素。

First, a formula to estimate ground level UV-B from meteorological data was developed for the purpose of utilization in the ecological study. Estimated values were obtained from an equation incorporating global solar radiation, total ozone, latitude and month for various regional points throughout Japan corresponding to the location of weather stations with the meteorological data. Estimation was deemed valid based on the high correlation between estimated and observed values of UV-B.Second, the ecological association was assessed between estimated regional UV-B values for the past 30 years and recent mortality risks of all cancer, leukemia and lung cancer, expressed in terms of the standardized mortality ratio (SMR) of the corresponding region. There was a weak but significant negative correlation (suggesting protective effect) between estimated UV-B values of the 1960 decennial period and 1983-87 SMR of all cause mortality in both men and women. In contrast, there was a weak but significant positive correlation (suggesting adverse effect) between estimated UV-B values of the 1960 decennial period and 1983-87 SMR of leukemia in both men and women. Such relationships were maintained regardless of the selection of regional points or period of estimated UV-B values only for all cancer mortality in women. The correlations were higher in magnitude and significance level than when latitude was used as a surrogate of UV-B exposure. It is thus possible that regional UV-B levels partly account for the regional differences in disease-specific morbidity or mortality in Japan. The study will further be developed taking into consideration such factors as the selection of regional points and relationship between other regional indices of disease-specific morbidity or mortality.