疟疾（malaria）是一种经由携带有疟原虫的疟蚊传播给人类和其他动物的全球性急性寄生虫传染病。主要流行于热带亚热带地区。疟疾的发生是人、疟蚊和疟原虫、自然环境的交互作用过程，人的活动影响自然环境，自然环境的变化又影响疟蚊和疟原虫的生存，疟蚊和疟原虫的生存又影响着人的健康，三者相互影响，相互制约。人和疟蚊的作用不仅仅通过自然环境的作用而发生，也可以直接发生作用；人的活动会直接影响疟蚊的生存，与疟蚊的可接触性，进而影响疟疾的发生，如杀虫剂、蚊香、蚊帐的使用与否。人的作用可能通过社会、经济、政治、卫生、宗教等方式体现、影响，人类参与的社会向来是一个复杂的系统，疟疾发病率的影响因子也是错综复杂。教育、经济、医疗水平以及易感人群的比重、各行业从业人口的比重、主要经济作用方式等这些社会经济因素都或多或少的影响着疟蚊与人可接触性的强弱、疟疾发病率的高低。准确识别出各个因子的影响，与疟疾发病率的影响、贡献率，对人为干预有指导作用，以进行更有效的资源配置。由于其至今没有疫苗，其预防控制措施的是否有效、合理实施对疟疾发病率的高低起着举足轻重的作用。 目前，国际上应用多种方法、工具，探测变量间的相关性，如皮尔逊相关系数、互信息、神经网络、地理探测器等，但各有利弊。学科之间、领域之间也多存在方法借鉴和方法融合。很多其他领域（非地学\空间领域）工具、方法延伸发展进入地学领域，如皮尔逊相关系数引申发展的地学中的空间相关系数；地学领域的很多工具也在应用中扩展到其他领域（如社会科学、公共健康领域），如地理探测器应用在汶川地震五岁以下儿童死因探测中。MIC是当前领先的用以在海量数据中探测双变量间相关性的模型方法，远远优于其他探测双变量间相关性的方法。目前该方法主要应用于基因、蛋白质等这些基因、分子领域，还没有在公共健康领域大肆应用，也没有在地理信息领域涉入。本文意图应用这种方法分析可能对疟疾发病率产生影响的各种自然资源、社会经济因素，探测出到底哪些因素和疟疾发病率相关，相关程度有多大。当前已绘有最新的2010年全球主要疟蚊的分布图，为本研究的准确研究区域确定提供了依据，使之成为可能。自然环境存在空间分异性，疟蚊空间分布存在分异，精确量化各种自然环境要素、社会经济要素对疟蚊空间分布的影响，为有效配置资源、降低成本提供基本的决策支持知识。此外，意图起到抛装引玉的作用，以该方法在气候-疟疾敏感人群识别和脆弱区划中的应用研究，尝试、探索MIC在地理信息系统空间分析领域的应用，引来MIC在地理学界、公共健康界更深层次的领域应用和衍化发展。 研究发现：在全球主要疟蚊分布区统计意义上，抗疟资金的使用与疟疾发病率高度相关，年均温、月温度高于15℃的平均持续时间和月温高于25℃持续时间的变异性与疟疾发病率相关，月降水多于100毫升的平均持续时间较为相关。平均灯光指数在主要疟蚊分布区与疟疾发病率相关，但是在主要流行区，灯光指数的变化幅度与发病率的相关性更显著。 本研究在全球尺度上量化了各个因子对全球疟疾发病率的影响力，用同一把标尺度量了在全球尺度有争议的一些因子和其他一些对疟疾发病率有影响的因子的贡献率，弥补了国际上这一领域的空缺，为更进一步的敏感人群识别和脆弱区划研究奠定了理论基础，进而为人类如何去相应气候变化带来的健康威胁的响应、适应机制的制定提供了知识支持。

Exactly identifying malaria impact factors and each factor’s influence to malaria incidence would guide the preventive measure of malaria, and it was import to more efficiently allocating resources since the vaccine was absence. Malaria is a mosquito-borne infectious disease of humans and other animals caused by Plasmodium. It is presently endemic in a broad band around equator. The occurrence of malaria is the interaction of human being, anopheles and the natural environment. Human activity influences the natural environment, corresponding the change of natural environment affects the subsistence of anopheles, while the survival of anopheles threaten the health of human being. The three interact on and restricts each other. While the interaction of human and anopheles not limited to though the natural environment, but also can directly interaction. Human activity can directly affect the survival of anopheles, the probability of contact with anopheles, and then it influences the occurrence of malaria, such as the usage of pesticide, mosquito incense and mosquito net. It may represent the human being’s influence, such as society, economic, policy, health, religion, etc. The impact factors are intricate, for a society of human being participating is always a complex society. Factors of social economic more or less affect the possible of contact of human and anopheles and the incidence of malaria, such as the level of education, economic and health, the proportion of susceptible population, the population of each industry, the method of economic, etc. Presently, there are many methods and tools to detect the relationship of variables, such as Pearson correlation coefficient, manual information, artificial neural networks, geographical detector, etc. and each has its own adaptability condition and restriction. One subject always refers and applies another subject’s methods and tools, so as different fields. Some other field’s methods and tools are referred into spatial analysis field, such as Pearson correlation coefficient is referred to spatial correlation coefficient. Some spatial methods and tools are applied to analyze social science and public health problem, such as using geographical detector to detector the risk factors of the death of children under-five during Wenchuan earthquake. MIC is a newly international ahead method to detect the correlation of two variables in large dataset, which is superior to other methods to detect the correlation of two variables. This method presently applied in genus and protein field, and had not widely applied in public health. This article intends to detect/analysis those factors that relating to malaria incidence and rank the level of correlations among those economic and natural environment factors may relating to malaria incidence. The globally map of dominant malaria vectors in 2010 may it possible to determine more exactly research region. The study would provide basic knowledge for making decision to efficiently allocate resources and lower cost though quantifying the influence of each factor to malaria spatial distribution. In addition this paper’s research could be test about apply MIC in spatial analysis, and promote the method develop in the field of geosciences. We found that malaria incidence is highly associated to malaria funding, and related to mean annual temperature, mean duration month temperature higher than 15℃，standard deviation of duration of month temperature higher than 25℃，and mean of monthly precipitation more than 100mm, in global dominant malaria vector distribution region. Mean of nightlight index is associated with malaria incidence, which is not the same with malaria endemic region where the incidence is more associated with the standard deviation of nightlight index. This article quantified the influence of each factor to global malaria incidence, and measured the contribution of some factors in globally dispute and other effected malaria incidence by the same ruler. Which made up the absence of this field and established the theoretical foundation for further identifying sensitive population and dividing vulnerable region, more than this which can provide the necessary knowledge for human being to adapt the threaten of the climate change to public health.