高级检索

大气PM10浓度与健康暴露-反应关系门槛效应模型分析

Exposure-response relationship between ambient air PM10 pollution and health: a threshold effect model

  • 摘要: 目的 探讨大气可吸入颗粒物(PM10)污染浓度与健康的暴露-反应关系是否存在拐点及拐点的个数,为相关政策的制定提供参考。方法 收集2011年和2013年中国统计年鉴以及各城市环境公报中的PM10数据,利用雷达图作出2011和2013年各省份PM10平均浓度分布图;利用2011和2013年中国健康与养老追踪调查数据,通过门槛效应模型研究各城市PM10污染与自评健康、过去1个月是否患病、过去1年是否有住院需要等3个指标的暴露-反应曲线。结果 雷达图显示,2011和2013年全国27个省市的PM10浓度分布情况相似,中部和北部地区的污染物浓度高于东南沿海地区。门槛效应模型回归分析结果显示,自评健康、过去1个月是否患病、过去1年是否有住院需要3个健康指标的暴露-反应曲线均存在2个拐点(门槛效应的门槛值);当PM10浓度分别为<96、96~199、>199 μg/m3时,PM10浓度每升高1 μg/m3,被调查者自评健康较好的概率分别下降0.092%(P<0.01)、0.041%(P>0.05)、0.010%(P>0.05);当PM10浓度分别为<52、52~60、>60 μg/m3时,PM10浓度每升高1 μg/m3,被调查者报告过去1个月患病的概率分别升高0.131%(P<0.05)、下降0.070%(P>0.05)、升高0.023%(P<0.05);当PM10浓度分别为<45、45~56、>56 μg/m3时,PM10浓度每升高1 μg/m3,过去1年有住院需要的概率分别升高0.061%(P>0.05)、下降0.110%(P<0.05)、升高0.019%(P>0.05)。结论 PM10浓度与健康的暴露-反应关系曲线存在拐点,政府在制定卫生政策时应充分考虑到不同污染物浓度对居民健康影响的差异性,以减少不必要的卫生费用支出和卫生资源投入。

     

    Abstract: Objective To explore whether there are inflection points and the number of the points for the exposure-response relationship between ambient air particulate matter less than 10 μm in aerodynamic diameter (PM10) and health and to provide references for relevant policy-making.Methods PM10 data for cities in the year of 2012 and 2014 were obtained from China Statistical Yearbook and Environmental Bulletins of cities and radar chart was adopted to describe the distribution of PM10 level across provinces in China in 2011 and 2013.Based on the dataset of China Health and Retirement Longitudinal Study (2011 and 2013),we used threshold effect model to explore the exposure-response relationships between PM10 levels across cities and 3 indicators including self-rated health,illness experience during past one month,and history of required hospitalization during past one year.Results Radar chart analysis indicated that the distribution of PM10 across 27 provinces were similar between the two years and the pollution level in central China and north China was higher than that in southeast coastal regions.The threshold effect model analyses revealed two inflection points (threshold values) in the exposure-response curves between PM10 level and the 3 indicators;every 1 μg/m3 increase in PM10 was associated with the decreases of 0.092% (P<0.01),0.041% (P>0.05),and 0.01% (P>0.05) in the possibility of reporting good self-rated health when the PM10 level being<96,between 96 and 199,and >199 μg/m3;every 1 μg/m3 increase in PM10 was related to the 0.131% increase (P<0.05),0.070% decrease (P>0.05),and 0.023% increase (P<0.05) in the possibility of having illness during past one month when the PM10 level being<52,between 52 and 60,and >60 μg/m3;and every 1μg/m3 increase in PM10 was associated with the 0.061% increase (P>0.05),0.110% decrease (P<0.05),and 0.019% increase (P>0.05) in the probability of required hospitalization during past one year when the PM10 level being<45,between 45 and 56,and >56 μg/m3,respectively.Conclusion There are inflection points in the exposure-response curves between PM10 level and health outcomes and different impacts of various PM10 pollution levels on health should be concerned when making relevant policies by governmental agencies to conduct reasonable health resources allocation.

     

/

返回文章
返回