EN中文

Journal of Geo-information Science >

2019 , Vol. 21 >Issue 3: 398 - 406

DOI: https://doi.org/10.12082/dqxxkx.2019.180517

Spatiotemporal Variation Analysis and Risk Determinants of Hand, Foot and Mouth Disease in Beijing-Tianjin-Tangshan, China

  • ZHANG Xiangxue , 1, 2 ,
  • WANG Li , 3, * ,
  • YIN Lichang 1 ,
  • XU Chengdong 2 ,
  • LI Xia 1 ,
  • LIU Yang 4, 5
Expand
  • 1. The School of Earth Science and Resources, Chang'an University, Xi’an 710054, China;
  • 2. State Key Laboratory of Resources and Environmental Information System, Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of Sciences, Beijing 100101, China
  • 3. The College of Environment and Planning of Henan University, Kaifeng 475004, China
  • 4. Guizhou Academy of Sciences,Guiyang 550001, China
  • 5. China National Center for Food Safety Risk Assessment, Beijing 100022, China
*Corresponding author: WANG Li, E-mail:

Received date: 2018-10-14

  Request revised date: 2018-12-20

  Online published: 2019-03-15

Supported by

Innovation Project of LREIS (O88RA205YA、O88RA200YA)

Special Scientific Research Fund of Public Welfare Profession of China, No.GYHY20140616.

Copyright

《地球信息科学学报》编辑部 所有
Fold

Abstract

：Hand, foot and mouth disease (HFMD) is a common infectious childhood disease. In recent years, the number of cases of HFMD in China has increased rapidly, and has received increasing attention. Although there are many related studies, only a few studies focus on the spatiotemporal heterogeneity of HFMD incidence and quantify the association between meteorological factors, socioeconomic variables, and HFMD incidence. Geodetector and Bayesian space-time hierarchical models were applied to analyze the spatiotemporal heterogen-eity of the HFMD incidence from 2009 to 2013 within the Beijing-Tianjin-Tangshan region. These were used to quantify the determinant power of meteorological factors, socioeconomic variables, and the interactions between two of these factors. The Geodetector method has the axiom that if an explanatory variable (x) determined an explained variable (y), the explained variable would exhibit a spatial distribution similar to that of the explanatory variable. This method has been widely used to measure the determinant power of potential explanatory variables. The Bayesian space-time hierarchical model has the potential to show the spatiotemporal variation of a geographic phenomenon. The results showed that: (1) the highest incidence of HFMD occurred in late spring and summer (May to July), and the lowest incidence occurred in winter (December to February). (2) Spatial heterogeneity existed. In particular high risks areas were mainly concentrated in areas of high economic development. The population density and proportion of the tertiary industry determinants, play a lead role in contributing to the spatial heterogeneity of HFMD incidence (q values of 0.35 and 0.28, respectively, as calculated by GeoDetector). (3) The main meteorological factors affecting the temporal heterogeneity of HFMD incidence were average temperature, cumulative precipitation, and relative humidity (with a determinant powers calculated by GeoDetector of 0.38, 0.27 and 0.13, respectively). Additionally, the interactions were greater than the independent effects between socioeconomic variables or meteorological factors. For example, the interaction of average temperature and relative humidity, average temperature and precipitation, average temperature and wind speed were 0.43, 0.40 and 0.42, respectively. The interaction of population density and proportion of the tertiary industry was 0.55. This result presented the strongest correlation with HFMD incidence. Temperature and relative humidity were also dominant factors influencing the spatiotemporal transmission of HFMD, along with areas of high economic development with high population density. This study provides a theoretical basis for the prevention and control of HFMD by detecting the spatiotemporal heterogeneity of the HFMD incidence and quantifying the impact factors within the study region.

Key words： Hand; foot; and mouth disease; GeoDetector; spatiotemporal heterogeneity; determinant power; meteorological factors

Cite this article

ZHANG Xiangxue , WANG Li , YIN Lichang , XU Chengdong , LI Xia , LIU Yang . Spatiotemporal Variation Analysis and Risk Determinants of Hand, Foot and Mouth Disease in Beijing-Tianjin-Tangshan, China[J]. Journal of Geo-information Science, 2019 , 21(3) : 398 -406 . DOI: 10.12082/dqxxkx.2019.180517

1 引言

HFMD 大多是由肠道病毒Enterovirus 71 （EV71）、柯萨奇病毒A16 （CV-A16）引起的常见的儿童传染病[1]。发病症状多见：发热;多部位现皮疹、溃疡,如手、脚、口部;少数病例可能会迅速发展为严重的神经系统疾病或引起致命性的并发症[2,3]。
迄今为止,HFMD已在全球各地多次爆发,其中亚太地区是主要流行地区之一,如新加坡[4]、越南[5]、泰国[6]、日本[7]、澳大利亚[8]和中国[1]都曾多次经历疫情爆发的情况。2007年和2008年初,中国经历了几次严重疫情后,强化了国家监测系统,并于2008年5月将其纳入丙类传染病[9]。考虑到它对儿童健康的威胁及其在继脊髓灰质炎后,成为肠道病毒相关疾病的主要诱因,因此加强HFMD的规范化预防控制刻不容缓。
目前大多数研究证实了HFMD风险具有时间异质性,并将其归因于气象因子。然而,在不同的研究中,气象因子对手足口病有不同的影响。例如,在中国北方,HFMD发病率在6月达到峰值（单峰）,而在中国南方,则在5月和9-10月各出现一次峰值（双峰）[1]。在新加坡,研究表明3月或5月可观测到HFMD发病率的峰值（双峰）[4]。此外,一项越南的研究表明,手足口病发病率在温暖的春季（3-5月）和冬季（9-12月）各出现一次年度峰值[5]。近年来,人们越来越关注于探索气象因子对HFMD发病率时间异质性的影响,如平均气温,相对湿度,累积降水量和平均风速[4,10-11]。而且不同地理区域的气候条件,人口特征和卫生条件的多样性也可能是导致不同区域内手足口病发病风险出现峰值的月份不一致的原因。
与此同时,一些研究证实了HFMD风险具有空间异质性。研究表明,其与社会经济因子密切相关,如人口密度,地理环境,社会经济状况和基础设施等。例如,Yan等[12]的研究结果表明城市地区的HFMD发病率要明显高于农村地区,并证明其与距离高速公路的远近和人均GDP等风险因素有密切关系;Hu等[13]采用地理加权回归方法得出儿童人口密度可以解释中国2912个县HFMD月累计发病率变异的56%; Cao等[14]的研究结果表明地理环境要素如人口密度、NDVI、医院密度与HFMD发病风险有显著关系。且不同地区的社会经济条件、基础设施的完备性、医疗设施与资源等存在差异,这些差异也会影响HFMD的传播和发病[9]。
近年来,关于HFMD的研究主要使用的方法有地理加权回归[13]、空间扫描统计[15]、广义相加模型[16]等。这些方法主要侧重在时间或空间上分别探测潜在驱动因子对HFMD发病率的影响,但对于量化各影响因子及其两两交互作用对HFMD的解释力及时空的分层异质性尚未深入研究。而地理探测器和贝叶斯时空层次模型在分析疾病的时空异质性和量化影响因子解释力方面有着不可替代的优势。因此,本研究采用贝叶斯时空层次模型对京津唐地区HFMD发病率的时空异质性进行分析,并由地理探测器模型量化影响因子的解释力及其交互作用。

2 研究区概况、数据来源与研究方法

2.1 研究区概况

京津唐地区位于华北平原东北部,包括北京市、天津市、唐山市和廊坊市（图1）,是中国人口基数最多、人口流动最多的地区之一。该地区地貌类型多样,以山地、平原为主,属于大陆性季风气候,年均温为12 ℃。且降水主要集中在夏季（6-8月）,占全年降水量的68.7%。京津唐地区是中国HFMD发病率最高的地区之一[15,17]。
Fig. 1 Spatial differences in HFMD and number of cases per month in the Beijing-Tianjin-Tang area from 2009 to 2013

图1 2009-2013年京津唐地区HFMD月均疾病数及空间差异

2.2 基础数据

研究区包括京津唐地区的49个区县,采用的数据是2009年1月至2013年12月京津唐地区各区县的HFMD发病数据。另外,用于探测影响HFMD季节性的气象因子,包括平均温度/（℃）、相对湿度/%、风速/（m/s）和降水量/mm（图2）。用于探测影响HFMD空间异质性的社会经济因子数据涵盖研究区的经济发展状况,如人口密度、人均生产总值（GDP）、第二产业比重和大三产业比重。本文基于以上所选取的气象因子和社会经济因子对京津唐地区的手足口病的时空异质性进行研究（表1）。
Fig. 2 Temporal change in meteorological factors in the Beijing-Tianjin-Tang area from 2009 to 2013

图2 2009-2013年京津唐地区气象因子的时间变化

Tab. 1 Descriptive characteristics for various data

表1 数据描述

最小值 25%值 均值 标准差 中位数 75%值 最大值
发病率/（10-4） 0.00 2.23 24.26 35.89 11.48 31.88 338.49
平均温度/°C -9.20 0.70 11.93 11.37 13.50 22.60 28.50
相对湿度/% 26.00 49.00 57.56 11.71 57.00 67.00 82.00
累积降水/mm 0.00 5.80 49.13 55.54 22.00 72.03 439.90
平均风速/（m/s） 1.11 1.97 2.37 0.53 2.30 2.69 4.26
人口密度/（人/km2） 103.00 490.00 6058.00 10 366.00 934.00 5487.00 35 627.00
人均GDP/（104元） 1.72 4.10 7.13 4.55 5.88 8.90 28.78
第二产业比重/% 4.00 25.00 42.29 20.60 47.00 59.00 71.00
第三产业比重/% 23.00 33.00 51.47 22.33 42.00 75.00 95.00

2.3 贝叶斯时空层次模型

贝叶斯时空层次模型（BSTHM）用于分析疾病的时间趋势和空间分布格局,解决可能出现在时间和空间上的方差非齐次问题[18]。此外,其充分利用样本信息、先验信息来估计时空参数的后验分布,使结果更稳健,置信度更高。
考虑到HFMD发病数属于计数数据,且各区县发病率完全不同,会存在过度散布的情况,因此本文采用参数为nit和uit的泊松分布来模拟数据,即yit为i（i=1,2,…,49）区县在t（t =1,2,…,60）月的病例数,nit为i区县t月的风险人群数,公式如下：
y it ~ Pois son ( n it , u it ) （1）
log ( u it ) = α + s i + ( b 0 t * + v t ) + b 1 i t * + ε it （2）
式中：uit表示i区县和t月的HFMD潜在风险。α表示研究期内京津唐地区总体疾病风险的固定效应。si描述i区县研究期内HFMD发病率的空间相对风险。（b0t*+vt）由线性趋势b0t*和随机效应vt组成,表示研究区的总体时间变化趋势。t*=t-tmid表示相对于研究中期tmid的时间跨度。b1it*表示研究时期内各区县偏离总体的局部变化趋势[3]。具体而言,若b1i>0,则局部变化强度高于总体变化趋势,反之亦然。假设高斯噪声随机变量εit服从正态分布N（ σ ε 2 ）[18]。
Besag York Mollie （BYM）模型通过卷积运算实现空间结构随机效应和空间非结构随机效应的相互作用,以确定参数si和b1i的先验分布[19]。另外,条件自回归（CAR）先验分布和空间邻接矩阵W用以增强空间结构随机效应。即如果区域i和j共享一个共同边界,则wij=1,否则wij=0。空间随机效应的CAR先验分布意味着相邻区域往往具有相似的疾病风险,而时间随机效应参数先验分布也采用条件自回归先验形式,邻接权重为1,否则为0[20]。另外,模型中所有随机变量均方差（如 σ v , σ ε ）的先验分布都被确定为正值半高斯分布N+∞（0,10） [21]。
此外,根据Richardson等[22]提出的分类原则将研究区按照HFMD相对风险的高低分为热点、冷点、非冷非热区,即：若某区县的空间相对风险大于1的后验概率 p ( exp ( s i ) > 1 | data ) 大于0.99,则该区县属于热点区;若后验概率 p ( exp ( s i ) > 1 | data ) 小于0.01,则将该区县定义为冷点区。若 p ( exp ( s i ) > 1 | d ata ) 介于0.01~0.99之间,则该区县为非冷非热区。
本文计算由WinBUGS实现[23],通过马尔可夫链蒙特卡罗（MCMC）模拟获得模型中所有参数的后验分布。本文采用50万次迭代的MCMC链,其中20万次预烧期,30万次估算迭代。

2.4 地理探测器

地理探测器（GeoDetector）不仅可以量化单变量的空间分层异质性,也可探测不同解释变量对因变量的解释力（Power of Determinant）。即若因变量受某解释变量影响,则二者的时间和空间分布会趋于一致[24,25,26]。
在本研究中,该模型用于量化气象因子各自及两两交互作用对HFMD发病率的解释力。每个因子的解释力及其交互作用可通过因子探测器中的q值来确定,其输入数据包括一个因变量及各因子的分层信息[26]。q值计算公式如下：
q = 1 - ∑ h = 1 L N h σ h 2 N σ 2 （3）
SSW = ∑ h = 1 L N h σ h 2 , SST = N σ 2
式中：q表示风险因子的解释力,范围从0~1,即所选因子解释疾病发病率变化的q×100%。q值越大,因子的解释力就越大。若q值为0,则意味着所选因子与疾病完全无关。相反,若q值为1,则意味着所选因子与疾病完全相关。此外,h=1,2,…,L为变量Y或因子X的分层（Strata）或分类; N h 和N分别为层h和全区的单元数; σ 2 表示全区Y的方差; σ h 2 是层h内疾病发病率的方差;SSW和SST分别为层内方差之和、全区总方差。
不同因子（Xs）的相互作用由交互作用探测器检验,其可揭示因子X1和X2的相互作用是否增强或削弱对Y的影响或它们是否独立地影响Y[26]。因子X1和X2的q值计算在因子探测器中记为q（X1）和q（X2）,交互作用探测器计算它们交互时的q（X1∩X2）值。通过比较q（X1）、q（X2）和q（X1∩X2）,用以下5个类别以解释交互关系 [27],如表2如示。
Tab. 2 Types of interaction

表2 不同因子交互作用的类型

类型 交互作用
q(X1∩X2)<Min(q(X1,X2)) 非线性减弱
Min(q(X1,X2))<q(X1∩X2)<Max(q(X1,X2)) 单因子非线性减弱
q(X1∩X2)>Max(q(X1,X2)) 双因子增强
q(X1∩X2)=q(X1)+q(X2) 独立
q(X1∩X2)>q(X1)+q(X2) 非线性增强

3 结果及分析

3.1 时空异质性

在空间上,不同区县间HFMD发病风险 （exp（si））存在较大差异,其中朝阳区、大兴区、海淀区等地区的发病风险较高;遵化县、丰润县、迁安县等区域的发病风险较低（图3）。且由GeoDetector计算的空间分层异质性统计q值为0.79。
Fig. 3 Means of spatial relative risks (exp(si)) of HFMD for each region in the study area from 2009 to 2013

图3 2009-2013年京津唐地区各区县HFMD空间相对风险（exp（si））分布

将研究区按照Richardson分类规则进行分类[22]：其中,10个（20.41%）和9个（18.37%）区县分别被认为是热点区和冷点区。另外30个（61.22%）区县被认为非热非冷区,其中热点区主要分布在经济较发达的区域,冷点区主要分布在研究区的东部一带（图4）。
Fig. 4 Spatial distribution of hot and cold areas from 2009 to 2013

图4 2009-2013年HFMD发病相对风险热点、冷点区域的空间分布

图5表示2009-2013年京津唐地区HFMD发病风险的总体时间变化趋势。由图5可知,HFMD发病风险存在很强的季节性,高发期主要集中在春末和夏季（5-7月）,月均发病率为68.10/万人;而在发病风险较低的冬季（12-次年2月）,月均发病率为2.91/万人（图5）。
Fig. 5 Overall temporal trend (exp(b0t*+vt)) of HFMD at a 97.5% confidence interval

图5 HFMD发病风险的时间变化趋势（exp（b0t*+vt））及97.5% 的置信区间

3.2 气象因子解释力

GeoDetector的结果表明,气象因子与HFMD发病率的季节性变化密切相关,且不同的气象因子对HFMD发病率的影响存在差异。其中,平均温度的解释力最强,q值为0.38（p<0.01）,结合皮尔逊相关系数分析,可以得出：在一定的温度范围内,平均温度越高,HFMD发病率越高（表3）。
Tab. 3 Interaction of meteorological factors

表3 气象因子交互作用结果

气象因子 平均温度 相对湿度 累积降水 平均风速
平均温度 0.38**
相对湿度 0.43 0.13**
累积降水 0.40 0.30 0.27**
平均风速 0.42 0.17 0.30 0.03**

注：**表示显著水平P<0.01。

累积降水、相对湿度与HFMD发病率也呈正相关,q值分别为0.27和0.13（且p<0.01）。而平均风速在这些选定因子中对HFMD发病率的影响最小,q值为0.03,且与HFMD发病率呈负相关（表3）。
交互探测器分析了两因子对HFMD发病率的交互作用。即两因子对HFMD发病率的共同作用大于、等于或小于二者各自对HFMD发病率的影响之和[27]（表1）。本研究所选的指标中,以平均温度和相对湿度两个因子为例,它们各自对HFMD发病率的解释力是0.38,0.13。通过交互作用分析,二者共同对HFMD发病率的解释力达到0.43,即平均温度和相对湿度的交互作用要大于它们各自作用。结果表明所选因子之间的交互作用都呈现加强的趋势。
与此同时,社会经济因子与HFMD发病率的空间变化密切相关,且不同的社会因子对HFMD发病率的影响存在差异。其中,人口密度的解释力最强,q值为0.35（p<0.01）,结合皮尔逊相关系数分析,可以得出：人口密度与HFMD发病率呈正相关;第三产业比重与HFMD发病率也呈正相关,q值为0.28（p<0.01）（表4）。且人口密度和第三产业比重二者共同对HFMD发病率的解释力达到0.55,即人口密度和第三产业比重的交互作用要大于它们各自作用。结果表明所选因子之间的交互作用都呈现加强的趋势。
Tab. 4 Interaction of socioeconomic factors

表4 社会经济因子交互作用结果

社会经济
因子		 人口
密度		 人均
GDP		 第二产业
比重		 第三产业
比重
人口密度 0.35**
人均GDP 0.53 0.17
第二产业比重 0.41 0.34 0.18
第三产业比重 0.55 0.34 0.40 0.28**

注：**表示显著水平P<0.01。

另外,人均GDP、第二产业比重与HFMD发病率的关系不显著,q值分别为0.17和0.18（且p>0.05）（表4）。

4 讨论

HFMD已成为近些年中国大陆儿童死亡的主要原因之一,严重威胁儿童健康[1]。京津唐地区作为中国人口基数大、流动人口最多的地区之一,近些年在人口和经济的共同影响下,经历了数次大规模的HFMD爆发[15,17]。本研究利用BSTHM从时空角度分析京津唐地区HFMD的流行病学特征,且利用GeoDetector量化了各影响因子及其两两交互作用对HFMD发病率的解释力[28]。结果表明,HFMD发病风险最高的地区主要集中在经济较为发达的地区,即人口密度和第三产业比重二者对HFMD发病率具有显著影响,同时温度和湿度的交互作用对HFMD传播的影响最为显著。
HFMD风险具有明显的时间异质性,春夏季（5-7月）发病率最高,冬季（12-次年2月）发病率最低。许多研究结果表明,气象因子会影响HFMD病毒的传播和存活,从而影响HFMD的季节性变 化[15,17],一些研究报道温度和相对湿度在HFMD的季节变化中起着非常重要的作用[29,30,31]。
本研究发现平均温度和相对湿度与HFMD发病率呈正相关,可能是因为HFMD病毒在这种湿热的环境情况下更加活跃,与之前的结果相似。相关表明,温度和湿度对手足口病发病率的影响尤为突出[7]。温度和湿度与手足口病发病率呈正相关[11]。温度和相对湿度与HFMD发病率显著相关[32]。也有相关研究表明,随着温度和相对湿度的升高,HFMD的风险会增加[33]。
降水也与HFMD发病率呈正相关,与之前的研究结果一致。例如,在新加坡,研究发现降水与HFMD发病率呈正相关[34]。在香港,研究结果表明,降水也与HFMD的发病率有显着关系[35]。潜在的机制可能是炎热和潮湿的环境更有利于病毒的滋生和繁殖,从而增加HFMD的患病风险。
风速也是影响HFMD发病率的重要因子,且与HFMD发病率呈负相关,与之前的研究结果一致。如,一项研究发现,在香港,HFMD发病率与风速呈负相关[36]。原因可能是较高的风速会加速蒸发,使得病毒很难在干燥的食物和环境中生存和繁殖。此外,高风速也会影响人们的户外活动,减少人与人之间的接触,从而降低HFMD的传播[37]。这些结果表明,气象因子可通过影响病原体的生长环境,暴露概率和宿主易感性而在促进肠道传染病传播中起不同作用,从而导致疾病的发生,且气象因子两两之间的交互作用要比它们独自影响的作用更为显著。
此外,HFMD的发病率也呈现明显的空间异质性。HFMD的发病风险在不同地区往往存在较大差异。从冷热点区县的分布可知,HFMD发病风险较高的区县主要集中在经济和城市化水平较高的地区,且人口密度和第三产业比重二者共同对HFMD发病率的解释力高达55%,表明社会经济因子对HFMD发病率有显著的影响,这些结论与之前的研究结论一致。如Xu发现,第三产业比重与HFMD发病率呈正相关[15]。Zhu等[17]发现,经济发达地区（如北京,天津,上海和浙江）的发病率高于欠发达地区。Huang等[38]发现人口密度对HFMD的传播影响最大,解释力为42%。此外,丘文洋等[39]发现相对湿度和GDP等因子与手足口病发病风险密切相关。潜在的原因可能是,位于北京市和天津市的大多区县属于发达地区,城市经济水平较高,且人口基数大,又由于近年来经济的快速发展和城市化水平的提高,使得大城市及其周围的区县流动人口日益增加,这无疑加速了手足口病的传播;同时过多的人口涌入,其有限的生活和工作空间会加大人与人相互接触的机会。因此,在经济发达和人口集聚相互作用下,人口密度增高和经济社会平的提高会导致手足口病的传播频率加快。本研究仍存在一些局限性：本研究所采用的数据是以区县为空间研究单元,时间研究单元为月,而较小的研究单元,如乡镇、社区或周数据可能会更加凸显手足口病的流行病学特征,且结果会更有利于为相关部门制定详尽地防控方针提供更加全面的信息;因此可能会带来一些生态谬误[40]。另外,本研究中使用的气象数据来自研究区及周围地区有限的监测站点,采用反距离加权方法对气象数据进行插值,可能会引入一些不可避免的误差。

5 结论

本研究利用地理探测器和贝叶斯时空层次模型分析了2009-2013年京津唐地区HFMD发病率的时空动态及其与气象因子的关系。研究结果表明,温度和相对湿度的相互作用与手足口病的发病率显着相关,即在湿热环境下,HFMD 发病风险较高;且检测到高风险区域 （热点）主要集中在经济水平较高的地区。这些发现能为预防和控制京津唐地区HFMD提供参考和依据,合理配置有限的医疗资源;也可为其他区域的HFMD的时空异质性分析提供借鉴作用。

The authors have declared that no competing interests exist.

References

Publishing order | Descend order by publishing year | Descend order by cited within
[1]
Xing W J, Liao Q H, Viboud C.Hand, foot, and mouth disease in China, 2008-12: An epidemiological study[J]. Lancet Infectious Diseases, 2014,14(4):308-318.Hand, foot, and mouth disease is a common childhood illness caused by enteroviruses. Increasingly, the disease has a substantial burden throughout east and southeast Asia. To better inform vaccine and other interventions, we characterised the epidemiology of hand, foot, and mouth disease in China on the basis of enhanced surveillance. We extracted epidemiological, clinical, and laboratory data from cases of hand, foot, and mouth disease reported to the Chinese Center for Disease Control and Prevention between Jan 1, 2008, and Dec 31, 2012. We then compiled climatic, geographical, and demographic information. All analyses were stratified by age, disease severity, laboratory confirmation status, and enterovirus serotype. The surveillance registry included 76420064092 probable cases of hand, foot, and mouth disease (annual incidence, 1·2 per 1000 person-years from 2010–12), of which 26764942 (3·7%) were laboratory confirmed and 2457 (0·03%) were fatal. Incidence and mortality were highest in children aged 12–23 months (38·2 cases per 1000 person-years and 1·5 deaths per 10064000 person-years in 2012). Median duration from onset to diagnosis was 1·5 days (IQR 0·5–2·5) and median duration from onset to death was 3·5 days (2·5–4·5). The absolute number of patients with cardiopulmonary or neurological complications was 826464486 (case-severity rate 1·1%), and 2457 of 82486 patients with severe disease died (fatality rate 3·0%); 1617 of 1737 laboratory confirmed deaths (93%) were associated with enterovirus 71. Every year in June, hand, foot, and mouth disease peaked in north China, whereas southern China had semiannual outbreaks in May and September–October. Geographical differences in seasonal patterns were weakly associated with climate and demographic factors (variance explained 8–23% and 3–19%, respectively). This is the largest population-based study up to now of the epidemiology of hand, foot, and mouth disease. Future mitigation policies should take into account the heterogeneities of disease burden identified. Additional epidemiological and serological studies are warranted to elucidate the dynamics and immunity patterns of local hand, foot, and mouth disease and to optimise interventions. China–US Collaborative Program on Emerging and Re-emerging Infectious Diseases, WHO, The Li Ka Shing Oxford Global Health Programme and Wellcome Trust, Harvard Center for Communicable Disease Dynamics, and Health and Medical Research Fund, Government of Hong Kong Special Administrative Region.

DOI PMID

[2]
赵成松,赵顺英.HFMD的流行概况和应对策略[J].中国实用儿科杂志,2009,24(6):419-421.手足口病（hand—Lot—mouth disease）是由肠道病毒引起的传染病，多发生于5岁以下的婴幼儿，可引起发热和手、足、口腔等部位的皮疹、溃疡，在个别患儿可引起心肌炎、肺水肿、无菌性脑膜脑炎等并发症。引发手足口病的肠道病毒有20多种，其中（Cox A16）和最常见。

[ Zhao C S, Zhao S Y.Epidemic situation and coping strategies of hand, foot and mouth disease[J]. Chinese Journal of Practical Pediatrics, 2009,24(6):419-421. ]

[3]
吴北平,杨典,王劲峰,等.利用贝叶斯时空模型分析山东省手足口病时空变化及影响因素[J].地球信息科学学报,2016,18(12):1645-1652.手足口病是一种常见的传染病,多见于5岁以下儿童。近年来,中国手足口病发病人数逐年上升,疾病疫情也越来越受到公共卫生部门与社会大众的关注。虽然已有不少手足口病相关的研究,但对其时空变化及影响因素驱动效应的研究仍然较少。本文采用贝叶斯时空模型,对2008年山东省手足口病高发时间段（5-8月）的发病时空演变特征进行系统分析,并探究影响手足口病发病风险的气象因素。结果表明：1空间上不同区县的手足口病发病风险存在一定差异,且区县间的发病风险随时间变化趋势也各不相同;2 5月和6月手足口病发病风险明显高于整个研究阶段（5-8月）平均发病风险;3对手足口病发病风险影响较大的气象因素依次是：周平均温度、平均风速和平均气压。本文针对山东省手足口病时空演化特征及气象影响因素的研究,能为高发时间段内手足口病的区域化防控提供科学依据。

DOI

[ Wu B P, Yang D, Wang J F, et al.Space-time variability and determinants of hand, foot and mouth in Shandong province: A Bayesian spatio-temporal modeling approach[J]. Journal of Geo-Information Science, 2016,18(12):1645-1652. ]

[4]
Ang L W, Koh B K W, Chan K P. Epidemiology and control of hand, foot and mouth disease in Singapore, 2001-2007[J]. Annals Academy of Medicine Singapore, 2009,38(2):106-112.We reviewed the epidemiology of hand, foot and mouth disease (HFMD) in Singapore after the 2000 epidemic caused by Enterovirus 71 (EV71), with particular reference to the cyclical pattern, predominant circulating enteroviruses and impact of prevention and control measures in preschool centres.We analysed the epidemiological data from all clinical cases and deaths of HFMD diagnosed by medical practitioners and notified to the Ministry of Health, as well as laboratory data on enteroviruses detected among HFMD patients maintained by the Department of Pathology, Singapore General Hospital, and the Microbiology Laboratory, KK Women's and Children's Hospital from 2001 to 2007.The incidence rate was highest in the 0 to 4 years old age group, with males being predominant. Three deaths were reported between January and February 2001. Nationwide epidemics occurred periodically; the predominating circulating virus was Coxsackievirus A16 (CA16) in the 2002, 2005 and 2007 epidemics, and EV71 in the 2006 epidemic. During the epidemic years between 2005 and 2007, 2 peaks were observed. The number of institutional outbreaks had increased 10-fold from 167 in 2001 to 1723 in 2007, although most of these outbreaks were rapidly brought under control with an attack rate of less than 10%.HFMD remains an important public health problem in Singapore with the annual incidence rate per 100,000 population increasing from 125.5 in 2001 to 435.9 in 2007, despite stringent measures taken in preschool centres to prevent the transmission of infection. A high degree of vigilance should be maintained over the disease situation, in particular, surveillance of EV 71 which continues to cause severe complications and deaths in the region.

PMID

[5]
Van Tu Phan, Thao, Nguyen Thi Thanh, et al. Epidemiologic and virologic investigation of hand, foot, and mouth disease, southern Vietnam, 2005[J]. Emerging infectious diseases, 2007,13(11):1733-1741.During 2005, 764 children were brought to a large children's hospital in Ho Chi Minh City, Vietnam, with a diagnosis of hand, foot, and mouth disease. All enrolled children had specimens (vesicle fluid, stool, throat swab) collected for enterovirus isolation by cell culture. An enterovirus was isolated from 411 (53.8%) of the specimens: 173 (42.1%) isolates were identified as human enterovirus 71 (HEV71) and 214 (52.1%) as coxsackievirus A16. Of the identified HEV71 infections, 51 (29.5%) were complicated by acute neurologic disease and 3 (1.7%) were fatal. HEV71 was isolated throughout the year, with a period of higher prevalence in October-November. Phylogenetic analysis of 23 HEV71 isolates showed that during the first half of 2005, viruses belonging to 3 subgenogroups, C1, C4, and a previously undescribed subgenogroup, C5, cocirculated in southern Vietnam. In the second half of the year, viruses belonging to subgenogroup C5 predominated during a period of higher HEV71 activity.

DOI PMID

[6]
Puenpa J, Theamboonlers A, Korkong S.Molecular characterization and complete genome analysis of human enterovirus 71 and coxsackievirus A16 from children with hand, foot and mouth disease in Thailand during 2008-2011[J]. Archives of Virology, 2011,156(11):2007-2013.Hand, foot and mouth disease (HFMD) has mostly been caused by enterovirus 71 (EV71) and coxsackievirus A16 (CA16). CA 16 was the most common cause of HFMD in 2010. EV71 had a high prevalence in 2008-2009 and has been identified with a higher frequency since 2011. Nearly complete genome sequences of three EV71 strains (2008-2009 strains) and two CA16 strains (2010 strains) obtained from outbreaks in Thailand in 2008 to 2010 were characterized. Based on a phylogenetic tree of the complete VP1 region, three EV71 strains grouped into the B5, C1 and C4 genotypes, and two CA16 strains grouped into the C genotype. Based on sequence analysis, nucleotide changes were found to cluster in the internal ribosome entry site (IRES) element of the 5′-untranslated region (5′-UTR). Amino acid differences identified in all strains were located in the non-structural protein. These data also provide the molecular epidemiology of EV71 and CA16 outbreaks in Thailand.

DOI PMID

[7]
Onozuka D, Hashizume M.The influence of temperature and humidity on the incidence of hand, foot, and mouth disease in Japan[J]. Science of the Total Environment, 2011,410(1):19-25.78 We estimated the relationship of weather variability with pediatric HFMD cases. 78 HFMD cases increased with increasing average temperature and relative humidity. 78 This increase was most remarkable in children aged less than 10 years. 78 Weather factors have a significant influence on the incidence of HFMD infections.

DOI PMID

[8]
Burry J N, Moore B, Mattner C.Hand, foot and mouth disease in South Australia[J]. The Medical journal of Australia, 1968,2(18):812.The authors give a full clinical description of 3 cases of hand, foot and mouth disease in South Australia, not the first to be described in that continent [e.g. Bull. Hyg., 1962, v. 37, 397]. By using suckling mice they isolated Coxsackie A10 virus from both the cases investigated virologically. D. G. D avies.

DOI PMID

[9]
别芹芹,邱冬生,胡辉.我国HFMD时空分布特征的GIS分析[J].地球信息科学学报,2010,12(3):380-384.手足口病是一种常见传染病。近几年在我国多次暴发且发病人数显著增加,引起了我国政府和社会各界的广泛关注。目前,对手足口病的研究主要集中在医学领域,而在宏观尺度上的时空分布特征研究及其重点地区分布研究等方面均较少。本文探索应用地理信息系统(GIS)的工具和方法,对2008-2009年中国疾病预防控制信息系统收集的手足口病监测数据进行统计计算、空间可视化和空间分析,得到我国手足口病疫情的时空分布与动态变化特征。研究表明:(1)尽管全国均有手足口病的报告病例,但各省之间发病情况差异较大,且区域内的发病情况也存在较显著差异,一般在人口密度和人口流动性均较大的城市疫情较严重;(2)手足口病在我国的流行高峰期为4-7月,比国外相关研究中的描述提前了一个月;(3)5-6月,我国手足口病的高发区分布明显由南向北移动。(4)2008-2009年,我国手足口病患者98%以上为托幼儿童、散居儿童和学生。鉴此分析,本文提出了具有时间、空间和人群针对性的防控手足口病暴发流行的科学建议。

[ Bie Q Q, Qiu D S, Hu H.Spatial and temporal distribution characteristics of hand-foot-mouth disease in China[J]. Joural of Geo-Information Science, 2010,12(3):380-384. ]

[10]
Chang L Y, King C C, Hsu K H.Risk factors of enterovirus 71 infection and associated hand, foot, and mouth disease/herpangina in children during an epidemic in Taiwan[J]. Pediatrics, 2002,109(6):e88.Abstract OBJECTIVE: In 1998, an enterovirus 71 (EV71) epidemic in Taiwan was associated with hand, foot, and mouth disease (HFMD)/herpangina and involved 78 fatal cases. We measured EV71 seroprevalence rates before and after the epidemic and investigated risk factors associated with EV71 infection and illness. METHODS: Neutralizing antibodies to EV71 were assayed for 539 people before the epidemic and 4619 people of similar ages after the epidemic. Questionnaires, which were completed during household interviews after the epidemic, solicited demographic variables, exposure history, and clinical manifestations. RESULTS: A total of 129 106 cases of HFMD were reported during the epidemic. Age-specific pre-epidemic EV71 seroprevalence rates were inversely related to age-specific periepidemic mortality rates (r = -0.82) or severe case rates (r = -0.93). Higher postepidemic EV71 seropositive rates among children who were younger than 3 years positively correlated with higher mortality rates in different areas (r = 0.88). After the epidemic, 51 (56%) of 91 younger siblings of elder siblings who were EV71-seropositive were EV71-seropositive; otherwise, 2.2% (4 of 186) of younger siblings were EV71-seropositive (matched odds ratio [OR]: 10; 95% confidence interval [CI]: 3.4-29). Stepwise multiple logistic regression revealed other factors associated with EV71 infection to be older age (adjusted OR: 2.5; 95% CI: 1.9-3.4), attendance at kindergartens/child care centers (adjusted OR: 1.8; 95% CI: 1.3-2.5), contact with HFMD/herpangina (adjusted OR: 1.6; 95% CI: 1.2-2.1), greater number of children in a family (adjusted OR: 1.4; 95% CI: 1.1-1.7), and rural residence (adjusted OR: 1.4; 95% CI: 1.2-1.6). Twenty-nine percent of preschool children who were infected with EV71 developed HFMD/herpangina. Younger age and contact with HFMD/herpangina were significant factors for the development of EV71-related HFMD/herpangina in these children. CONCLUSIONS: An increased incidence of EV71 infection in young children occurred more often in geographic areas with increased mortality rates. Intrafamilial and kindergarten transmissions among preschool children were major modes of disease transmission during the widespread EV71 epidemic in Taiwan in 1998.

DOI PMID

[11]
Li T, Yang Z, Di B.Hand-foot-and-mouth disease and weather factors in Guangzhou, southern China[J]. Epidemiology and Infection, 2014,142(8):1741-1750.Hand-foot-and-mouth disease (HFMD) is becoming one of the common airborne and contact transmission diseases in Guangzhou, southern China, leading public health authorities to be concerned about its increased incidence. In this study, we aimed to examine the effect of weather patterns on the incidence of HFMD in the subtropical city of Guangzhou for the period 2009–2012, and assist public health prevention and control measures. A negative binomial multivariable regression was used to identify the relationship between meteorological variables and HFMD. During the study period, a total of 16602770 HFMD-confirmed cases were reported, of which 11 died, yielding a fatality rate of 0·66/1002000. Annual incidence rates from 2009 to 2012 were 132·44, 311·40, 402·76, and 468·59/1020000200 respectively. Each 1°C rise in temperature corresponded to an increase of 9·38% (95% CI 8·17–10·51) in the weekly number of HFMD cases, while a 102hPa rise in atmospheric pressure corresponded to a decrease in the number of cases by 6·80% (95% CI 616·99 to 616·65), having an opposite effect. Similarly, a 1% rise in relative humidity corresponded to an increase of 0·67% or 0·51%, a 102m/h rise in wind velocity corresponded to an increase of 4·01% or 2·65%, and a 1 day addition in the number of windy days corresponded to an increase of 24·73% or 25·87%, in the weekly number of HFMD cases, depending on the variables considered in the model. Our findings revealed that the epidemic status of HFMD in Guangzhou is characterized by high morbidity but low fatality. Weather factors had a significant influence on occurrence and transmission of HFMD.

DOI PMID

[12]
Yan L, Li X L, Yu Y Q.Distribution and risk factors of hand, foot, and mouth disease in Changchun, northeastern China[J]. Chinese Science Bulletin, 2014,59(5-6):533-538.Hand, foot, and mouth disease (HFMD) is a public health problem, and there have been increasing numbers of outbreaks in mainland China since 2008. Over 17,000 HFMD cases have been reported in Changchun between 2008 and 2011. This study characterized the temporal and spatial distribution of the disease and identified the risk factors for HFMD. The main findings were as follows: (i) there were significant differences in HFMD incidence among age groups, with 86.8% of reported cases in children younger than 5 years old, and boys showed a higher incidence than girls (

DOI

[13]
Hu M G, Li Z J, Wang J F.Determinants of the incidence of hand, foot and mouth disease in china using geographically weighted regression models[J]. Plos One, 2012,7(6):e38978.Over the past two decades, major epidemics of hand, foot, and mouth disease (HFMD) have occurred throughout most of the West-Pacific Region countries, causing thousands of deaths among children. However, few studies have examined potential determinants of the incidence of HFMD. Reported HFMD cases from 2912 counties in China were obtained for May 2008. The monthly HFMD cumulative incidence was calculated for children aged 9 years and younger. Child population density (CPD) and six climate factors (average-temperature [AT], average-minimum-temperature [ATmin], average-maximum-temperature [ATmax], average-temperature-difference [ATdiff], average-relative-humidity [ARH], and monthly precipitation [MP]) were selected as potential explanatory variables for the study. Geographically weighted regression (GWR) models were used to explore the associations between the selected factors and HFMD incidence at county level. There were 176,111 HFMD cases reported in the studied counties. The adjusted monthly cumulative incidence by county ranged from 0.26 cases per 100,000 children to 2549.00 per 100,000 children. For local univariate GWR models, the percentage of counties with statistical significance (p<0.05) between HFMD incidence and each of the seven factors were: CPD 84.3%, ATmax 54.9%, AT 57.8%, ATmin 61.2%, ARH 54.4%, MP 50.3%, and ATdiff 51.6%. The R2 for the seven factors univariate GWR models are CPD 0.56, ATmax 0.53, AT 0.52, MP 0.51, ATmin 0.52, ARH 0.51, and ATdiff 0.51, respectively. CPD, MP, AT, ARH and ATdiff were further included in the multivariate GWR model, with R2 0.62, and all counties show statistically significant relationship. Child population density and climate factors are potential determinants of the HFMD incidence in most areas in China. The strength and direction of association between these factors and the incidence of HFDM is spatially heterogeneous at the local geographic level, and child population density has a greater influence on the incidence of HFMD than the climate factors.

DOI PMID

[14]
Cao C, Li G, Zheng S, et al.editors. Research on the environmental impact factors of Hand-Foot-Mouth disease in Shenzhen, China using RS and GIS technologies[C]. Shenzhen: 2012 IEEE International Geoscience and Remote Sensing Symposium, 2012.

[15]
Xu C D.Spatio-temporal pattern and risk factor analysis of hand, foot and mouth disease associated with under-five morbidity in the Beijing-Tianjin-Hebei region of China[J]. International Journal of Environmental Research and Public Health, 2017,14(4):416.Hand, foot and mouth disease (HFMD) in children under the age of five is a major public health issue in China. Beijing–Tianjin–Hebei is the largest urban agglomeration in northern China. The present study aimed to analyze the epidemiological features of HFMD, reveal spatial clusters, and detect risk factors in this region. Reports of HFMD cases in Beijing–Tianjin–Hebei from 1 January 2013 to 31 December 2013 were collected from 211 counties or municipal districts. First, the epidemiological features were explored, and then SaTScan analysis was carried out to detect spatial clusters of HFMD. Finally, GeoDetector and spatial paneled model were used to identify potential risk factors among the socioeconomic and meteorological variables. There were a total of 90,527 HFMD cases in the year 2013. The highest rate was in individuals aged one year, with an incidence of 24.76/103. Boys (55,168) outnumbered girls (35,359). Temporally, the incidence rose rapidly from April, peaking in June (4.08/103). Temperature, relative humidity and wind speed were positively associated with the incidence rate, while precipitation and sunshine hours had a negative association. The explanatory powers of these factors were 57%, 13%, 2%, 21% and 12%, respectively. Spatially, the highest-risk regions were located in Beijing and neighboring areas, with a relative risk (RR) value of 3.04. The proportion of primary industry was negatively associated with HFMD transmission, with an explanatory power of 32%. Gross domestic product (GDP) per capita, proportion of tertiary industry, and population density were positively associated with disease incidence, with explanatory powers of 22%, 17% and 15%, respectively. These findings may be helpful in the risk assessment of HFMD transmission and for implementing effective interventions to reduce the burden of this disease.

DOI PMID

[16]
Huang Y, Deng T, Yu S C.Effect of meteorological variables on the incidence of hand, foot, and mouth disease in children: A time-series analysis in Guangzhou, China[J]. Bmc Infectious Diseases, 2013,13:134.Over the last decade, major outbreaks of hand, foot, and mouth disease (HFMD) have been reported in Asian countries, resulting in thousands of deaths among children. However, less is known regarding the effect of meteorological variables on the incidence of HFMD in children. This study aims at quantifying the relationship between meteorological variables and the incidence of HFMD among children in Guangzhou, China. The association between weekly HFMD cases in children aged <1502years and meteorological variables in Guangzhou from 2008 to 2011 were analyzed using the generalized additive model (GAM) and time-series method, after controlling for long-term trend and seasonality, holiday effects, influenza period and delayed effects. Temperature and relative humidity with one week lag were significantly associated with HFMD infection among children. We found that a 1°C increase in temperature led to an increase of 1.86% (95% CI: 0.92, 2.81%) in the weekly number of cases in the 0–1402years age group. A one percent increase in relative humidity may lead to an increase of 1.42% (95% CI: 0.97, 1.87%) in the weekly number of cases in the 0–1402years age group. This study provides quantitative evidence that the incidence of HFMD in children was associated with high average temperature and high relative humidity. The one-week delay in the effects of temperature and relative humidity on HFMD is consistent with the enterovirus incubation period and the potential time lag between onset of children’s sickness and parental awareness and response.

DOI PMID

[17]
Zhu Q, Hao Y T, Ma J Q.Surveillance of hand, foot, and mouth disease in mainland China (2008-2009)[J]. Biomedical and Environmental Sciences, 2011,24(4):349-356.Since HFMD was designated as a class C communicable disease in May 2008, 18 months surveillance data have been accumulated to December 2009. This article was to describe the distribution of HFMD for age, sex, area, and time between 2008 and 2009, to reveal the characteristics of the epidemic. We analyzed weekly reported cases of HFMD from May 2008 to December 2009, and presented data on the distribution of age, sex, area and time. A discrete Poisson model was used to detect spatial–temporal clusters of HFMD. More than 1 065 000 cases of HFMD were reported in Mainland China from May 2008 to December 2009 (total incidence: 12.47 per 10 000). Male incidence was higher than female for all ages and 91.9% of patients were <5 years old. The incidence was highest in Beijing, Shanghai, Zhejiang and Hainan. The highest peak of HFMD cases was in April and the number of cases remained high from April to August. The spatial–temporal distribution detected four clusters. Children <5 years old were susceptible to HFMD and we should be aware of their vulnerability. The incidence was higher in urban than rural areas, and an annual pandemic usually starts in April.

DOI PMID

[18]
Li G, Haining R, Richardson S.Space-time variability in burglary risk: A Bayesian spatio-temporal modelling approach[J]. Spatial Statistics, 2014,9(1):80-91.Modelling spatio-temporal offence data contributes to our understanding of the spatio-temporal characteristics of the risk of becoming a victim of crime and has implications for policing. Space–time interactions are deeply embedded both empirically and theoretically into many areas of criminology. In this paper, we apply a familiar Bayesian spatio-temporal model to explore the space–time variation in burglary risk in Peterborough, England, between 2005 and 2008. However, we extend earlier work with this model by presenting a novel two-stage method for classifying areas into crime hotspots, coldspots or neither and studying the temporal dynamics of areas within each risk category. A further contribution of this paper is the inclusion of covariates into the model in order to explain the space–time classification of areas. We discuss the advantages of, and identify future directions for, this form of modelling for analysing offence patterns in space and time. Implications for crime research and policing are also discussed.

DOI

[19]
Besag J, York J, Mollie A.Bayesian image-restoration, with 2 applications in spatial statistics[J]. Annals of the Institute of Statistical Mathematics, 1991,43(1):1-20.

DOI

[20]
李俊明. 基于Bayesian层次时空模型的我国老龄化分析与预测[J].统计研究,2016,33(8):89-94.本文首次利用Bayesian层次时空模型,以1995—2014年全国省级人口统计数据为基础,分析了近20年来我国老龄化在空间和时间上的变化规律。研究发现:1我国高老龄化地区分布已形成X型地理空间分布结构,东部地区为主,西部地区为辅,总体老龄化率呈上升趋势;2四川、重庆、辽宁、安徽、湖北和湖南等6个地区不仅是老龄化热点区域,而且老龄化增速也快于全国平均水平,特别是四川和重庆,老龄化程度和增速都是全国最高;3中西部地区老龄化程度虽然低于全国平均水平,但增加速度却高于全国平均水平;4北京、天津、上海、江苏、浙江和广东等6个高老龄化地区的老龄化率趋于平稳或增速放缓;5预测"全面二孩"政策情境下我国2030年老龄化率为13.19%(11.10%,20.94%)。

DOI

[ Li J M.Space-time variation of chinese aging based on bayesian hierarchy spatio-temporal model[J]. Statistical Research, 2016,33(8):89-94. ]

[21]
Gelman A.Prior distributions for variance parameters in hierarchical models (Comment on an Article by Browne and Draper)[J]. Bayesian Analysis, 2006,1(3):515-533.

DOI

[22]
Richardson S, Thomson A, Best N.Interpreting posterior relative risk estimates in disease-mapping studies[J]. Environmental Health Perspectives, 2004,112(9):1016-1025.There is currently much interest in conducting spatial analyses of health outcomes at the small-area scale. This requires sophisticated statistical techniques, usually involving Bayesian models, to smooth the underlying risk estimates because the data are typically sparse. However, questions have been raised about the performance of these models for recovering the "true" risk surface, about the influence of the prior structure specified, and about the amount of smoothing of the risks that is actually performed. We describe a comprehensive simulation study designed to address these questions. Our results show that Bayesian disease-mapping models are essentially conservative, with high specificity even in situations with very sparse data but low sensitivity if the raised-risk areas have only a moderate (< 2-fold) excess or are not based on substantial expected counts (> 50 per area). Semiparametric spatial mixture models typically produce less smoothing than their conditional autoregressive counterpart when there is sufficient information in the data (moderate-size expected count and/or high true excess risk). Sensitivity may be improved by exploiting the whole posterior distribution to try to detect true raisedrisk areas rather than just reporting and mapping the mean posterior relative risk. For the widely used conditional autoregressive model, we show that a decision rule based on computing the probability that the relative risk is above 1 with a cutoff between 70 and 80% gives a specific rule with reasonable sensitivity for a range of scenarios having moderate expected counts (~ 20) and excess risks (~1.5- to 2-fold). Larger (3-fold) excess risks are detected almost certainly using this rule, even when based on small expected counts, although the mean of the posterior distribution is typically smoothed to about half the true value.

DOI PMID

[23]
Lunn D J, Thomas A, Best N.WinBUGS-A Bayesian modelling framework: Concepts, structure, and extensibility[J]. Statistics and Computing, 2000,10(4):325-337.

DOI

[24]
Wang J F, Li X H, Christakos G.Geographical detectors-based health risk assessment and its application in the neural tube defects study of the Heshun region, China[J]. International Journal of Geographical Information Science, 2010,24(1):107-127.Physical environment, man‐made pollution, nutrition and their mutual interactions can be major causes of human diseases. These disease determinants have distinct spatial distributions across geographical units, so that their adequate study involves the investigation of the associated geographical strata. We propose four geographical detectors based on spatial variation analysis of the geographical strata to assess the environmental risks of health: the risk detector indicates where the risk areas are; the factor detector identifies factors that are responsible for the risk; the ecological detector discloses relative importance between the factors; and the interaction detector reveals whether the risk factors interact or lead to disease independently. In a real‐world study, the primary physical environment (watershed, lithozone and soil) was found to strongly control the neural tube defects (NTD) occurrences in the Heshun region (China). Basic nutrition (food) was found to be more important than man‐made pollution (chemical fertilizer) in the control of the spatial NTD pattern. Ancient materials released from geological faults and subsequently spread along slopes dramatically increase the NTD risk. These findings constitute valuable input to disease intervention strategies in the region of interest.

DOI

[25]
Wang J F, Zhang T L, Fu B J.A measure of spatial stratified heterogeneity[J]. Ecological Indicators, 2016,67(2):50-56.Spatial stratified heterogeneity, referring to the within-strata variance less than the between strata-variance, is ubiquitous in ecological phenomena, such as ecological zones and many ecological variables. Spatial stratified heterogeneity reflects the essence of nature, implies potential distinct mechanisms by strata, suggests possible determinants of the observed process, allows the representativeness of observations of the earth, and enforces the applicability of statistical inferences. In this paper, we propose aq-statistic method to measure the degree of spatial stratified heterogeneity and to test its significance. Theqvalue is within [0,1] (0 if a spatial stratification of heterogeneity is not significant, and 1 if there is a perfect spatial stratification of heterogeneity). The exact probability density function is derived. Theq-statistic is illustrated by two examples, wherein we assess the spatial stratified heterogeneities of a hand map and the distribution of the annual NDVI in China.

DOI

[26]
王劲峰,徐成东.地理探测器:原理与展望[J].地理学报,2017,72(1):116-134.空间分异是自然和社会经济过程的空间表现,也是自亚里士多德以来人类认识自然的重要途径.地理探测器是探测空间分异性,以及揭示其背后驱动因子的一种新的统计学方法,此方法无线性假设,具有优雅的形式和明确的物理含义.基本思想是:假设研究区分为若干子区域,如果子区域的方差之和小于区域总方差,则存在空间分异性;如果两变量的空间分布趋于一致,则两者存在统计关联性.地理探测器q统计量,可用以度量空间分异性、探测解释因子、分析变量之间交互关系,已经在自然和社会科学多领域应用.本文阐述地理探测器的原理,并对其特点及应用进行了归纳总结,以利于读者方便灵活地使用地理探测器来认识、挖掘和利用空间分异性.

DOI

[ Wang J F, Xu C D.Geodetector: principle and prospective[J]. Acta Geographica Sinica, 2017,72(1):116-134. ]

[27]
Wang J F, Hu Y.Environmental health risk detection with GeogDetector[J]. Environmental Modelling & Software, 2012,33(1):14-15.Human health is affected by many environmental factors. Geographical detector is software based on spatial variation analysis of the geographical strata of variables to assess the environmental risks to human health: the risk detector indicates where the risk areas are; the factor detector identifies which factors are responsible for the risk; the ecological detector discloses the relative importance of the factors; and the interaction detector reveals whether the risk factors interact or lead independently to disease. [All rights reserved Elsevier].

DOI

[28]
李媛媛,徐成东,肖革新.京津唐地区细菌性痢疾社会经济影响时空分析[J].地球信息科学学报,2016,18(12):1615-1623.细菌性痢疾是常见疾病,也是备受关注的公共健康问题。近年来,京津唐地区的细菌性痢疾发病率相对较高。本文首先分析了2012年京津唐地区细菌性痢疾的季节性和人群特征;其次,使用热点分析模型,探索了京津唐地区细菌性痢疾发病率的时空聚集性;最后,运用地理探测器模型研究了细菌性痢疾的发生和社会经济因素之间的量化关系。结果表明：1细菌性痢疾发病的峰值时间是8月;发病率最高的年龄段是0-9岁,其次是80岁以上;农民群体发病率最高,其次是散居儿童。2京津唐地区细菌性痢疾在空间和时间上都存在聚集性。空间上,细菌性痢疾发病率的高聚集区主要分布于北京市的房山区及门头沟区和天津市的滨海新区,低聚集区主要分布于唐山市的滦县,时间上,细菌性痢疾发病率的高聚集区在12个月均有发生,低聚集区主要发生在1-4月以及6月。3影响细菌性痢疾发病率空间分布的主要社会经济因素为农村人口占总人口的比例、人口密度和各区县的人均GDP,它们的解释力分别为61%,37%和20%,并且发现它们的交互作用都大于独自影响的作用。本研究通过对京津唐地区细菌性痢疾发病情况的人群特征、时空特征以及影响因素的分析,为本地区细菌性痢疾的预防和控制提供理论依据。

DOI

[ Li Y Y, Xu C D, Xiao G X.Spatial-temporal analysis of social-economic factors of bacillary dysentery in Beijing-Tianjin-Tang[J]. Journal of Geo-Information Science, 2016,18(12):1615-1623. ]

[29]
Basu R, Ostro B D.A multicounty analysis identifying the populations vulnerable to mortality associated with high ambient temperature in California[J]. American Journal of Epidemiology, 2008,168(6):632-637.The association between ambient temperature and mortality has been established worldwide, including the authors' prior study in California. Here, they examined cause-specific mortality, age, race/ethnicity, gender, and educational level to identify subgroups vulnerable to high ambient temperature. They obtained data on nine California counties from May through September of 1999-2003 from the National Climatic Data Center (countywide weather) and the California Department of Health Services (individual mortality). Using a time-stratified case-crossover approach, they obtained county-specific estimates of mortality, which were combined in meta-analyses. A total of 231,676 nonaccidental deaths were included. Each 10 degrees F (approximately 4.7 degrees C) increase in mean daily apparent temperature corresponded to a 2.6% (95% confidence interval (CI): 1.3, 3.9) increase for cardiovascular mortality, with the most significant risk found for ischemic heart disease. Elevated risks were also found for personsat least 65 years of age (2.2%, 95% CI: 0.04, 4.0), infants 1 year of age or less (4.9%, 95% CI: -1.8, 11.6), and the Black racial/ethnic group (4.9%, 95% CI: 2.0, 7.9). No differences were found by gender or educational level. To prevent the mortality associated with ambient temperature, persons with cardiovascular disease, the elderly, infants, and Blacks among others should be targeted.

DOI PMID

[30]
Zhuang D F, Hu W S, Ren H Y.The influences of temperature on spatiotemporal trends of hand-foot-and-mouth disease in mainland China[J]. International Journal of Environmental Health Research, 2014,24(1):1-10.Understanding the influence of temperature on hand-foot-and-mouth disease (HFMD) is an important public health concern as well as being a major climate-infection issue in mainland China. City-scale data of incidence rates (IRs) of HFMD and temperature from 2008 to 2009 in mainland China has been analyzed. There were two peak periods for HFMD prevalence from April to July and August to November. Regions with higher monthly IR of HFMD periodically shifted following the pattern of south090009north090009south from March to December. Monthly IR of HFMD at city scale were closely associated with both average monthly temperature and monthly temperature range. Our study shows that spatiotemporal trends of HFMD infection were sensitive to temperature variation, and suggest that preventive measures should be considered for limiting the epidemic of HFMD in the cities with higher monthly IR during the peak periods.

DOI PMID

[31]
Bell M L, O'Neill M S, Ranjit N. Vulnerability to heat-related mortality in Latin America: A case-crossover study in Sao Paulo, Brazil, Santiago, Chile and Mexico City, Mexico[J]. International Journal of Epidemiology, 2008,37(4):796-804.Abstract BACKGROUND: Factors affecting vulnerability to heat-related mortality are not well understood. Identifying susceptible populations is of particular importance given anticipated rising temperatures from climatic change. METHODS: We investigated heat-related mortality for three Latin American cities (Mexico City, Mexico; S o Paulo, Brazil; Santiago, Chile) using a case-crossover approach for 754 291 deaths from 1998 to 2002. We considered lagged exposures, confounding by air pollution, cause of death and susceptibilities by educational attainment, age and sex. RESULTS: Same and previous day apparent temperature were most strongly associated with mortality risk. Effect estimates remained positive though lowered after adjustment for ozone or PM(10). Susceptibility increased with age in all cities. The increase in mortality risk for those >or=65 comparing the 95th and 75th percentiles of same-day apparent temperature was 2.69% (95% CI: -2.06 to 7.88%) for Santiago, 6.51% (95% CI: 3.57-9.52%) for S o Paulo and 3.22% (95% CI: 0.93-5.57%) for Mexico City. Patterns of vulnerability by education and sex differed across communities. Effect estimates were higher for women than men in Mexico City, and higher for men elsewhere, although results by sex were not appreciably different for any city. In S o Paulo, those with less education were more susceptible, whereas no distinct patterns by education were observed in the other cities. CONCLUSIONS: Elevated temperatures are associated with mortality risk in these Latin American cities, with the strongest associations in S o Paulo, the hottest city. The elderly are an important population for targeted prevention measures, but vulnerability by sex and education differed by city.

DOI PMID

[32]
Wang H, Du Z H, Wang X J.Detecting the association between meteorological factors and hand, foot, and mouth disease using spatial panel data models[J]. International Journal of Infectious Diseases, 2015,34:66-70.The aim of this study was to quantify the relationship between meteorological factors and the occurrence of hand, foot, and mouth disease (HFMD) among children in Shandong Province, China, at a county level, using spatial panel data models. Descriptive analysis was applied to describe the epidemic characteristics of HFMD from January 2008 to December 2012, and then a global autocorrelation statistic (Moran's I) was used to detect the spatial autocorrelation of HFMD in each year. Finally, spatial panel data models were performed to explore the association between the incidence of HFMD and meteorological factors. Moran's I at the county level were high, from 0.30 to 0.45 (p<0.001), indicating the existence of a high spatial autocorrelation on HFMD. Spatial panel data models are more appropriate to describe the data. Results showed that the incidences of HFMD in Shandong Province, China were significantly associated with average temperature, relative humidity, vapor pressure, and wind speed. Spatial panel data models are useful when longitudinal data with multiple units are available and spatial autocorrelation exists. The association found between HFMD and meteorological factors makes a contribution towards advancing knowledge with respect to the causality of HFMD and has policy implications for HFMD prevention and control.

DOI PMID

[33]
Kim B I, Ki H, Park S.Effect of climatic factors on hand, foot, and mouth disease in south Korea, 2010-2013[J]. Plos One, 2016,11(6):e0157500.Hand, foot, and mouth disease (HFMD) causes characteristic blisters and sores mainly in infants and children, and has been monitored in South Korea through sentinel surveillance since 2009. We described the patterns of HFMD occurrence and analyzed the effect of climatic factors on national HFMD incidence. Weekly clinically diagnosed HFMD case rates (per 1,000 outpatients) in sentinel sites and weekly climatic factors, such as average temperature, relative humidity, duration of sunshine, precipitation, and wind speed from 2010 to 2013, were used in this study. A generalized additive model with smoothing splines and climatic variables with time lags of up to 2 weeks were considered in the modeling process. To account for long-term trends and seasonality, we controlled for each year and their corresponding weeks. The autocorrelation issue was also adjusted by using autocorrelation variables. At an average temperature below 18 C, the HFMD rate increased by 10.3% for every 1 C rise in average temperature (95% confidence interval (CI): 8.4, 12.3%). We also saw a 6.6% increase in HFMD rate (95% CI: 3.6, 9.7%) with every 1% increase in relative humidity under 65%, with a 1.5% decrease in HFMD rate observed (95% CI: 0.4, 2.7%) with each 1% humidity increase above 65%. Modeling results have shown that average temperature and relative humidity are related to HFMD rate. Additional research on the environmental risk factors of HFMD transmission is required to understand the underlying mechanism between climatic factors and HFMD incidence.

DOI PMID

[34]
Hii Y L, Rocklov J, Ng N.Short term effects of weather on hand, foot and mouth disease[J]. Plos One, 2011,6(2):S18-S18.Hand, foot, and mouth disease (HFMD) outbreaks leading to clinical and fatal complications have increased since late 1990s; especially in the Asia Pacific Region. Outbreaks of HFMD peaks in the warmer season of the year, but the underlying factors for this annual pattern and the reasons to the recent upsurge trend have not yet been established. This study analyzed the effect of short-term changes in weather on the incidence of HFMD in Singapore.The relative risks between weekly HFMD cases and temperature and rainfall were estimated for the period 2001-2008 using time series Poisson regression models allowing for over-dispersion. Smoothing was used to allow non-linear relationship between weather and weekly HFMD cases, and to adjust for seasonality and long-term time trend. Additionally, autocorrelation was controlled and weather was allowed to have a lagged effect on HFMD incidence up to 2 weeks.Weekly temperature and rainfall showed statistically significant association with HFMD incidence at time lag of 1-2 weeks. Every 1°C increases in maximum temperature above 32°C elevated the risk of HFMD incidence by 36% (95% CI66=661.341-1.389). Simultaneously, one mm increase of weekly cumulative rainfall below 75 mm increased the risk of HFMD by 0.3% (CI66=661.002-1.003). While above 75 mm the effect was opposite and each mm increases of rainfall decreased the incidence by 0.5% (CI66=660.995-0.996). We also found that a difference between minimum and maximum temperature greater than 7°C elevated the risk of HFMD by 41% (CI66=661.388-1.439).Our findings suggest a strong association between HFMD and weather. However, the exact reason for the association is yet to be studied. Information on maximum temperature above 32°C and moderate rainfall precede HFMD incidence could help to control and curb the up-surging trend of HFMD.

DOI PMID

[35]
Ma E, Lam T, Wong C.Is hand, foot and mouth disease associated with meteorological parameters?[J]. Epidemiology & Infection, 2010,138(12):1779-1788.We examined the relationship between meteorological parameters and hand, foot and mouth disease (HFMD) activity. Meteorological data collected from 2000 to 2004 were tested for correlation with HFMD consultation rates calculated through the sentinel surveillance system in Hong Kong. The regression model constructed was used to predict HFMD consultation rates for 2005-2009. After adjusting for the effect of collinearity, mean temperature, diurnal difference in temperature, relative humidity, and wind speed were positively associated with HFMD consultation rates, and explained HFMD consultation rates well with 2 weeks' lag time (R = 0 119, P = 0.010).The predicted HFMD consultation rates were also also well matched with the observed rates (Spearman's correlation coefficient = 0.276, P = 0.000) in 2005-2009. Sensitivity analysis showed that HFMD consultation rates were mostly affected by relative humidity and least affected by wind speed. Our model demonstrated that climate parameters help in predicting HFMD activity, which could assist in explaining the winter peak detected in recent years and in issuing early warning.

DOI PMID

[36]
Wang P, Goggins W B, Chan E Y Y. Hand, foot and mouth disease in Hong Kong: A time-series analysis on its relationship with weather[J]. Plos One, 2016,11(8):e0161006.Hand, foot and mouth disease (HFMD) is an emerging enterovirus-induced infectious disease for which the environmental risk factors promoting disease circulation remain inconclusive. This study aims to quantify the association of daily weather variation with hospitalizations for HFMD in Hong Kong, a subtropical city in China. A time series of daily counts of HFMD public hospital admissions from 2008 through 2011 in Hong Kong was regressed on daily mean temperature, relative humidity, wind speed, solar radiation and total rainfall, using a combination of negative binomial generalized additive models and distributed lag non-linear models, adjusting for trend, season, and day of week. There was a positive association between temperature and HFMD, with increasing trends from 8 to 20 C and above 25 C with a plateau in between. A hockey-stick relationship of relative humidity with HFMD was found, with markedly increasing risks over 80%. Moderate rainfall and stronger wind and solar radiation were also found to be associated with more admissions. The present study provides quantitative evidence that short-term meteorological variations could be used as early indicators for potential HFMD outbreaks. Climate change is likely to lead to a substantial increase in severe HFMD cases in this subtropical city in the absence of further interventions.

DOI PMID

[37]
Belanger M, Gray-Donald K, O'Loughlin J. Influence of weather conditions and season on physical activity in Adolescents[J]. Annals of Epidemiology, 2009,19(3):180-186.Little is known about how seasonal variation in physical activity relates to declining physical activity in adolescence. We quantified how each of daily weather conditions and season affect physical activity during adolescence. We followed 1293 students, initially aged 12 to 13 years over 5 years. Participants completed a 7-day physical activity recall checklist every 3 months. Data on daily weather conditions were obtained from Environment Canada. The association between the number of physical activity sessions per day, and each of season, and daily weather conditions was assessed in Poisson regressions. Adjusting for age, sex, and month, the average number of physical activity sessions per day was 2% to 4% lower for every 10mm of rainfall and 1% to 2% higher for every 10 C increase in temperature. Although every 10cm of snow accumulation was associated with 5% higher activity rates, days with snowfall had lower physical activity. Physical activity was lower during winter and increased during warmer months. However, the warm-month increases did not compensate for winter decreases so that activity decreased by 7% yearly. Declines in physical activity during adolescence may be partly explained by declines during winter. Increasing opportunities for physical activity during poor weather, in particular during winter, may mitigate declines in physical activity during adolescence.

DOI PMID

[38]
Huang J X, Wang J F, Bo Y C.Identification of health risks of hand, foot and mouth disease in China using the geographical detector technique[J]. International Journal of Environmental Research and Public Health, 2014,11(3):3407-3423.Hand, foot and mouth disease (HFMD) is a common infectious disease, causing thousands of deaths among children in China over the past two decades. Environmental risk factors such as meteorological factors, population factors and economic factors may affect the incidence of HFMD. In the current paper, we used a novel model—geographical detector technique to analyze the effect of these factors on the incidence of HFMD in China. We collected HFMD cases from 2,309 counties during May 2008 in China. The monthly cumulative incidence of HFMD was calculated for children aged 0–9 years. Potential risk factors included meteorological factors, economic factors, and population density factors. Four geographical detectors (risk detector, factor detector, ecological detector, and interaction detector) were used to analyze the effects of some potential risk factors on the incidence of HFMD in China. We found that tertiary industry and children exert more influence than first industry and middle school students on the incidence of HFMD. The interactive effect of any two risk factors increases the hazard for HFMD transmission.

DOI PMID

[39]
丘文洋,李连发,张杰昊,等.利用空间聚集的贝叶斯网络评估手足口病发病风险[J].地球信息科学学报,2017,19(8):1036-1048.手足口病是一种常见的传染病,以往的研究表明该疾病与气象、环境和社会经济等因素相关联,其影响关系复杂,而疾病本身体现出较强的区域聚集性,采用普通的线性风险建模方法无法捕捉影响因素的复杂性及空间聚集性。因此,本文以山东省为例,在前人研究的基础上,提出了采用贝叶斯网络综合风险建模方法研究手足口病的发病风险与气象、土地利用、社会经济及空气污染等要素间的关系,并通过引入空间扫描统计聚集结果,将空间聚集引入到贝叶斯网络模型加强其空间推理功能,减少模型的偏差,提高评估的精度。结果表明,本文建立的手足口病空间贝叶斯网络风险模型具有较高的估计效果,引入的空间聚集性较好地融入到贝叶斯概率推理模型中,合理建立预测因子同手足口病发病风险之间的关系。通过对建模结果的解译,分析了手足口病的发病风险影响因素,特别是气候、社会经济及空气污染的影响。本文的空间贝叶斯建模方法及研究结果对手足口病暴发的防控预警具有重要的意义。

DOI

[ Qiu W Y, Li L F, Zhang J H, et al.A Bayesian network method considering spatial cluster to evaluate health risk of hand, foot and mouth disease[J]. Journal of Geo-Information Science, 2017,19(8):1036-1048. ]

[40]
Jelinski D E, Wu J.The modifiable areal unit problem and implications for landscape ecology[J]. Landscape Ecology, 1996,11(3):129-140.Landscape ecologists often deal with aggregated data and multiscaled spatial phenomena. Recognizing the sensitivity of the results of spatial analyses to the definition of units for which data are collected is critical to characterizing landscapes with minimal bias and avoidance of spurious relationships. We introduce and examine the effect of data aggregation on analysis of landscape structure as exemplified through what has become known, in the statistical and geographical literature, as the Modifiable Areal Unit Problem (MAUP). The MAUP applies to two separate, but interrelated, problems with spatial data analysis. The first is the “scale problem”, where the same set of areal data is aggregated into several sets of larger areal units, with each combination leading to different data values and inferences. The second aspect of the MAUP is the “zoning problem”, where a given set of areal units is recombined into zones that are of the same size but located differently, again resulting in variation in data values and, consequently, different conclusions. We conduct a series of spatial autocorrelation analyses based on NDVI (Normalized Difference Vegetation Index) to demonstrate how the MAUP may affect the results of landscape analysis. We conclude with a discussion of the broader-scale implications for the MAUP in landscape ecology and suggest approaches for dealing with this issue.

DOI

Options
Outlines

/

〈 〉