Search tips
Search criteria 


Logo of tropmedLink to Publisher's site
Am J Trop Med Hyg. 2016 April 6; 94(4): 741–749.
PMCID: PMC4824213

Identifying Flood-Related Infectious Diseases in Anhui Province, China: A Spatial and Temporal Analysis


The aim of this study was to explore infectious diseases related to the 2007 Huai River flood in Anhui Province, China. The study was based on the notified incidences of infectious diseases between June 29 and July 25 from 2004 to 2011. Daily incidences of notified diseases in 2007 were compared with the corresponding daily incidences during the same period in the other years (from 2004 to 2011, except 2007) by Poisson regression analysis. Spatial autocorrelation analysis was used to test the distribution pattern of the diseases. Spatial regression models were then performed to examine the association between the incidence of each disease and flood, considering lag effects and other confounders. After controlling the other meteorological and socioeconomic factors, malaria (odds ratio [OR] = 3.67, 95% confidence interval [CI] = 1.77–7.61), diarrhea (OR = 2.16, 95% CI = 1.24–3.78), and hepatitis A virus (HAV) infection (OR = 6.11, 95% CI = 1.04–35.84) were significantly related to the 2007 Huai River flood both from the spatial and temporal analyses. Special attention should be given to develop public health preparation and interventions with a focus on malaria, diarrhea, and HAV infection, in the study region.


Over the last decade, flooding has been the most common natural disaster globally,1 which has been responsible for almost half of all victims of natural disasters with economic losses of nearly US$185 billion (EM-DAT, 2011, China is one of the most flood-prone countries in the world. Large population, complicated topography and climate conditions, and rapid urbanization promote a high risk of exposure to flood.

From June 29 to July 25 in 2007, Huai River basin experienced heavy rainfall, which resulted in long-term severe flood.3 The flood event examined in this study was the worst flood event of Anhui Province in last 50 years, leading to a total cost of US$1.9 billion.4 There were more than 7,000 displaced persons with a total affected population of approximately 15 million.5

The floods had serious implications on public health through contamination of drinking water and accumulation of waste. Active proliferation of pathogen and disease vectors around stagnant waters can be induced by flood events.6,7 In addition, flood events often cause population displacement.8 Insufficient public health condition, overcrowding, and intimate contact among refugees may lead to outbreaks of infectious diseases.9,10 Floods may particularly assist the transmission of the following infectious diseases, including waterborne diseases, vector-borne diseases, and rodent-borne diseases.11

Although, some literatures have analyzed the relationship between flood event and a specific infectious disease in China, no previous studies have systematically detected sensitive diseases from arrange of infectious diseases with different pathways of transmission. Our study aimed to examine the impact of flood events to detect sensitive infectious diseases in Huai River basin. Results will be helpful to explore flood-related sensitive infectious diseases so as to assist in developing policies to prevent and reduce the risk of infectious diseases associated with floods.

Materials and Methods

Study area and floods.

Anhui Province is located between longitude 114°54′ and 119°37′E and latitude 29°41′ and 34°38′N. The total population of Anhui Province is approximately 61.2 million, of which the proportion of rural population is 61.30%.12 Huai River flows through the north part of Anhui Province, forming the Huai River basin. Many branches of Huai River cross north Anhui Province, leading the adjacent regions (including seven cities Fuyang, Bozhou, Huaibei, Suzhou, Luan, Huainan, and Bengbu, which can be further divided into 29 counties), the worst affected area in the 2007 Huai River basin flood event. Our study area covers all these 29 counties (Figure 1C ). The study area locates in a subhumid warm temperate continental monsoon climate zone, with a rainy weather in summer seasons. During the flood period, the average temperature was 27°C. Figure 2 showed meteorological conditions in the study area from 2004 to 2011. The daily accumulated rainfall peaked during the flood period in 2007, with the highest value of 5,500 mm. The average precipitation of Huai River basin was 370 mm during this period, which equaled to two to three times of precipitation during the same period of previous years. Other meteorological variables in flood period were similar with the corresponding periods in other years.5,13

Figure 1.
(A) China and the location of Anhui Province, (B) meteorological stations in Anhui Province, and (C) Huai River basin.
Figure 2.
Daily meteorological conditions in Huai River basin, 2004–2011.

Data collection.

On the basis of the transmission mechanism of infectious diseases and findings from previous research, floods mainly affect vector-borne,14,15 waterborne,14 and rodent-borne diseases.16 This study examined these three types of infectious diseases regarding their relationship to the flood event, including malaria,1719 dengue, diarrhea,20 bacillary dysentery,21 hepatitis A virus (HAV) infection, hepatitis E virus (HEV) infection, typhoid and paratyphoid,22 schistosomiasis, cholera, leptospirosis,2325 Japanese encephalitis,26 pestis, and hemorrhagic fever with renal syndrome (HFRS).19 According to the law of the People's Republic of China on Prevention and Treatment of Infectious Diseases (2013 Amendment),27 diarrhea is infectious diarrhea other than cholera, dysentery, typhoid, and paratyphoid.27 Disease surveillance data on these selected diseases and the annual population from 2004 to 2011 were obtained from the National Notifiable Disease Surveillance System (NDSS). All infectious disease cases were diagnosed by National Notifiable Disease diagnostic criteria reported in the law of the People's Republic of China on the Prevention and Treatment of Infectious Diseases. In China, statutory notifiable infectious diseases were divided into three categories: A, B, and C. The statutory notifiable category A infectious disease should be reported within 2 hours, while statutory notifiable category B and C infectious disease should be reported within 24 hours.28 In our study, both clinical and laboratory cases were included. The laboratory-confirmed frequency of each disease has been included in Table 1.

Table 1
Incidence rates and number of cases during flood period in 2007 in flood-affected areas with lag effects

Daily meteorological data from 2004 to 2011 were collected from the China Meteorological Data Sharing Service System.13 The flood event were defined by Yearbook of Meteorological Disasters in China and Bulletin of Flood and Drought Disaster in China 2007.4,5 All the 17 stations in the whole Anhui Province were used to record the climatic data. If there was a meteorological station in the city, the meteorological data from this station were used in the study for this city. If no meteorological station was in the city, we chose the meteorological data reported by the nearest station for this city (Figure 1B). The meteorological data retrieved from each meteorological station represented the condition of all the counties in this city.

The meteorological variables were selected based on findings from literature review and biological possibility. The proliferation of parasites, such as flies and mosquitoes, and pathogens would boost during hot and humid days.2931 There were some evidence of the effect of rainfall on infectious diseases, while the association between infectious diseases and rainfall is far complex than that with temperature.32 Rainfall may also affect the frequency and level of contamination of drinking water.32 The results of some studies also indicated that air pressure may affect the transmission of waterborne diseases.33 Remarkable increase was observed in risks of vector-borne diseases associated with increased sunshine hours in previous studies.34 All the meteorological factors that have a potential biological impact on the diseases were initially included in the spatial regression model. Therefore, the accumulated rainfall data from June 29 to July 25 in 2007 for each county were included only in the spatial regression models as control factors. The mean measurements from June 29 to July 25 in 2007 of the following variables were also included in the multivariate spatial regression models as control factors, including daily mean temperature (MeanT), daily maximum temperature (MaxT), relative humidity (RH), mean air pressure (MeanP), mean speed of wind (MeanSW), daily maximum speed of wind (Max SW), and daily sunshine hours (SH).

To control the potential confounding caused by socioeconomic status and health-care conditions in different cities, Gross Domestic Product/person (GDP/person), health facilities/100,000 persons, and health faculties/100,000 persons of each county were also included in the spatial regression analysis as control factors. These data were collected from Statistical Yearbook of Anhui Province 2007.12

Temporal analysis.

In temporal analysis, the study area included all 29 counties in Huai River basin (Figure 1C). June 29 to July 25 in 2007 was chosen to be the exposed period. To alleviate the seasonal variation and secular trend of diseases, the same periods from 2004 to 2006 and 2008 to 2011 were defined as the control period, which were all in summers but not affected by floods. Lag effects were detected by using daily odds ratios (ORs) of the flood event for each disease. We compared daily incidences between exposed periods and control periods (from June 29 to July 25, 2004 to 2011 except 2007). The lag value with the maximum significant OR for each disease was selected for inclusion in subsequent temporal and spatial analysis. Considering the results from previous studies and biological plausibility of infectious diseases, up to 30-day lagged effects were examined in our study.14,21

We defined flood event as a binary independent variable. The exposed period coded to 1 and the control period coded to 0. We used daily accumulated incidence for each disease of all 29 counties in Huai River basin as a dependent variable. The lag effect adjusted incidence during June 29 and July 25 from 2004 to 2011was included in the Poisson regression model. We assumed that the socioeconomic conditions, environment, human immunity, and other control programs in the study areas were same during these years. The Poisson regression analysis was performed to examine the association between flood event and each disease. All the temporal analysis was performed in SAS 9.1.3 (SAS Institute Inc., Cary, NC).

Spatial analysis.

In recent years, spatial analysis had been widely implemented to describe the geographic distribution of diseases and to identify the factors associated with changing incidences.3538 It could adjust the mutual effect among the spatial neighbors, which may be ignored by traditional regression analysis.39 In this study, spatial autocorrelation and regression analysis were attached to detect disease aggregation and association between diseases and flood event, respectively. Lagged effect was controlled in spatial analyzes. In the spatial analysis, the study area included all 79 counties in whole Anhui Province (Figure 1B). The sample size was sufficient to perform the spatial analysis to detect any patterns in the spatial distribution.37,40 We defined flood event as a binary variable. All the counties were defined into two categories based on the Huai River basin flood event: flood-affected counties (Figure 1B without color), coded to 1 and non-affected areas (Figure 1B in colors), coded to 0. This binary variable was regarded as flood event factor. All the independent and dependent variables were included in county level.

For each of the 79 counties in Anhui Province, daily incidence for each disease from June 29 to July 25 in 2007 with lag phase was accumulated. Accumulated incidences, binary flood event variable, and meteorological and socioeconomic variables were included in a shapefile map of Anhui Province according to the county ID by ArcGIS software version 10.0 (ESRI Inc., Redlands, CA). County-level polygon map of Anhui Province was gained from Geographic Sciences and Natural Resources Research Data Center.

Spatial autocorrelation analysis.

Global spatial autocorrelation analysis was performed by Open GeoDa 1.2.0 (Luc Anselin, Phoenix, AZ). Global Moran's I statistic was used to measure the correlation among neighboring observations, to detect whether the diseases were randomly distributed in the whole Anhui Province.36,41 Queen criterion was attached to create the contiguity spatial weight file from the polygon shapefile constructed by ArcGIS software version 10.0 (ESRI Inc.). The null hypothesis assumed that the disease distributed randomly in Anhui Province.39 The range of Moran's I is from −1 to 1. There is a higher regional aggregation of incidence between counties, if the absolute value of Moran's I is more close to 1, otherwise, there is a lower correlation, which indicates the incidence may distribute randomly.39 P < 0.05 indicates the disease had a character of regional aggregating.42 Some factors may result in the regional aggregation. Therefore, spatial autocorrelation is the prerequisite of spatial regression analysis.39

Spatial regression analysis.

The exploration of the relationship between disease spatial distribution and flood event was adjusted by other meteorological variables and socioeconomic factors from a geographical aspect. Traditional regression models assume that the observation are mutually independent, which is not valid due to the spatial structure. Spatial regression models do not make such an assumption. Random effects were included to explain the potential impact caused by the spatial correlation. Randomly distributed diseases tested by a spatial autocorrelation analysis were not included in the regression models. The spatial regression model is expressed as follows:

equation image

where ui is spatial unstructured effect, which represents spatial heterogeneity, and ei is spatial structured effect, which reflects spatial dependence, for example, spatial autocorrelation.39

With Open GeoDa 1.2.0, the distributions of incidences in whole Anhui Province were obtained. For the incidences of diseases that were not normally distributed, incidences were log transformed. Each incidence then was included for the spatial regression modeling as a dependent variable based on the significant result from the spatial autocorrelated analysis (for those spatial autocorrelated diseases). Flood was input to the models as an explanatory variable. Control variables, including all the other meteorological variables (MeanT, MaxT, RH, RF, MeanP, MeanSW, Max SW, and SH) and socioeconomic and health-care conditions variables (GDP/person, health facilities/100,000 person, and health faculties/100,000 person of each county) were input into the model as well. Diseases, which reached a significance level of 0.05 or less in the global autocorrelation analysis, were selected for inclusion in multiple spatial regression models.43

Three regression models, including classic ordinary least squares (OLS) regression model, spatial lag model (SLM), and spatial error model (SEM) were used in our analysis. The OLS models were first performed on each disease separately to fit the data, and spatial dependence test was then conducted to compare these three kinds of models. If test shows no significant dependence, the OLS model was adopted, otherwise, P values of SLM and SEM were used to decide which model was appropriate for spatial regression.44

Ethical review.

Disease surveillance data used in this study were obtained from the NDSS with an ethical approval by the Chinese Center for Disease Control and Prevention. This study was fully reviewed and approved by the human research ethical committee of Shandong University.


Descriptive analysis.

During the flood event in 2007, the most commonly notified diseases were malaria (incidence rate = 17.867/100,000), diarrhea (incidence rate = 8.113/100,000), and bacillary dysentery (incidence rate = 3.474/100,000). Incidences of HAV infection, HFRS, Japanese encephalitis, typhoid and paratyphoid, and HEV infection were less than 1/100,000. No cases of leptospirosis, schistosomiasis, pestis, cholera, and dengue were notified in the study areas over the study period (Table 1). Given a very low number of notified cases may cover the real association, HFRS was excluded with only four cases occurred.

Lagged effects.

Detected lagged effects with the maximum significant OR of the seven diseases notified in the study region are presented in Table 2. The lag days with maximum significant OR for typhoid and paratyphoid (OR = 6.00, 95% confidence interval [CI] = 3.54–8.46), bacillary dysentery (OR = 1.64, 95% CI = 1.12–2.16), Japanese encephalitis (OR = 4.50, 95% CI = 1.40–7.60), diarrhea (OR = 1.70, 95% CI = 1.29–2.10), HAV infection (OR = 3.75, 95% CI = 1.78–5.72), HEV infection (OR = 4.50, 95% CI = 1.86–7.14), and malaria (OR = 4.62, 95% CI = 2.83–6.41) were varying from 9 to 27 days.

Table 2
The lag days and maximum relative risks based on incidence comparison between exposed period and control periods

Temporal analysis.

The results of Poisson regression analysis were shown in Table 3. It suggested that, compared with control periods, the ORs of malaria (OR = 3.44, 95% CI = 3.28–3.62), diarrhea (OR = 1.10, 95% CI = 1.05–1.15), HAV infection (OR = 1.40, 95% CI = 1.11–1.77), and Japanese encephalitis (OR = 0.58, 95% CI 0.47–0.71) were significant. Bacillary dysentery (OR = 1.04, 95% CI = 0.97–1.12), typhoid and paratyphoid (OR = 0.40, 95% CI = 0.14–1.14), and HEV infection (OR = 0.94, 95% CI = 0.67–1.31) were not significantly associated with flood event.

Table 3
Poisson regression analysis between flood and diseases

Spatial autocorrelation analysis.

The global spatial autocorrelation analysis for each disease showed that bacillary dysentery (Moran's I = 0.336, P = 0.001), diarrhea (Moran's I = 0.311, P = 0.001), Japanese encephalitis (Moran's I = 0.154, P = 0.017), HAV infection (Moran's I = 0.341, P = 0.001), HEV infection (Moran's I = 0.278, P = 0.001), malaria (Moran's I = 0.475, P = 0.001), and typhoid and paratyphoid (Moran's I = 0.165, P = 0.020) were all spatially autocorrelated in the study region (Table 4).

Table 4
Global spatial autocorrelation analysis

Spatial regression analysis.

As shown in Table 5, malaria and bacillary dysentery were included in the SLM, typhoid and paratyphoid was analyzed by SEM. Classic OLS regression model was applied to HAV infection, HEV infection, diarrhea, and Japanese encephalitis.

Table 5
Spatial regression model selection

Table 6 shows the ORs derived from the spatial regression models. Compared with the non-flood-affected areas, malaria (OR = 3.67, 95% CI = 1.77–7.61), HAV infection (OR = 6.11, 95% CI = 1.04–35.84), bacillary dysentery (OR = 2.46, 95% CI = 1.43–4.23), and diarrhea (OR = 2.16, 95% CI = 1.24–3.78) were significantly increased in Huai River basin in 2007. Japanese encephalitis (OR = 0.70, 95% CI = 0.35–1.39), typhoid and paratyphoid (OR = 0.66, 95% CI = 0.32–1.36), and HEV infection (OR = 1.24, 95% CI = 0.59–2.59) were not significantly impacted by flood. The R squares, which represent the goodness of fit for models in Table 6, indicated the proportion of disease variation that could be explained by all the variables in the model.

Table 6
Parameters estimated by spatial regression analysis


Our study has identified flood-related infectious diseases in Anhui Province, using both spatial and temporal analyses. Flood events may induce differential health impacts among different demographic groups. Therefore, our study has raised valuable epidemiology information for flood-related infectious disease control in a flood-prone area, Huai River basin. Results indicate that the increased incidences of malaria, diarrhea, and HAV infection are associated with the flood event in Huai River basin.

It is well known that risk of malaria outbreaks increases in the wake of flood events. The results from both the temporal analysis and the spatial analysis show that malaria is sensitive to flood. The changed living condition, environmental ecosystem, and human behaviors during flood events may increase the incidence of vector-borne infectious diseases.8,17,18,23 Natural systems and environmental health are damaged during the flood.45 In that mosquitoes and pathogens are sensitive to external moisture levels and temperature. Expanded waterlog, stagnant water, and high temperature can conduce to proliferation of pathogens, mosquitoes, and other arthropods.46 Large rainfall also impacts the breeding of mosquitoes. Heavy rainfall may increase larval habitat and vector population size by creating new habitat. Epic rainfall events can synchronize vector host seeking and virus transmission. Humid environment has been found to be one of the most critical determinants in vector-borne diseases.23 Our study shows that even after controlling these factors, flood has independent impacts on malaria.

The results suggest that flood may increase the incidence of waterborne diseases including diarrhea and HAV infection. There is some evidence that the increment in waterborne disease incidences during flood may be due partly to direct exposure to floodwater.23,46,47 During the flood, people living in flood-hit houses are at the risk of waterborne diseases.46,48 Flood event may lead to the transmission of waterborne diseases, by affecting pathogens, contaminating the drinking water47 and then influence the incidences of diarrhea and HAV infection.30 Flood may cause disruption of water purification and sewage disposal systems, rupture of underground pipelines and storage tanks, and overflowing of toxic waste sites. These events can result in increased exposure to contaminated waste, food and more pathogens.49 The damp and hot living condition in disaster areas may assist the growth and reproductive of pathogenic bacteria of gastrointestinal disease.50 Meanwhile, in the humid and heat environment, pathogenic factors are easily formed in the human body. The combination of terrible meteorological condition, flood contaminated water and food, substantial refugees and emergency workers displacement following spoiled houses, and lower degree of health care facilitate the transmission of communicable diseases in the Huai River basin.8,23,32,51 The result from spatial analysis indicated that compared with the non-affected areas, the increment of bacillary dysentery in Huai River basin was statistically significant. Temporal analysis showed that the OR of bacillary dysentery was also larger than one, although it was not statistically significant. The temporal results indicated that there was an increment of bacillary dysentery incidence during flood period while the increment was not significant compared with the control periods. Both analyses suggested that bacillary dysentery could be sensitive to flood events. In terms of typhoid and paratyphoid, there were only 19 cases observed during the flood event in Huai River basin. The ORs from both temporal and spatial were all less than one and not statistically significant. The results indicated that typhoid and paratyphoid might not be associated with flood event. In addition, a very low number of notified cases may cover the real associations. There was a low basic incidence of typhoid and paratyphoid in Huai River basin. During June 29 and July 25, from 2004 to 2006 and 2008 to 2011, the average number of cases of typhoid and paratyphoid was 17. This may not be sufficient to detect any significant trend over the period.

No significant associations have been detected between rodent-borne diseases and flood.

The results of the current work provide useful framework to identify infectious diseases related to flood events. Understanding the associations between diseases and flood events would be beneficial to examine population vulnerability to flood events. The capacity of people to have a better resilience to the event in specific region is also essential. More locally driven actions can be effectively taken to deal with the disease epidemic, once the flood-related diseases could be identified. In addition, with limited health resources, our quantified results of the association between diseases and flood events could help to control the flood-related diseases in public health preparation. It can assist local government to intensify their surveillance of sensitive diseases to achieve better resilience. The design of early warning system should take the lag effects of flood for different diseases into consideration. The combination of temporal and spatial analysis could provide more robust results. The Poisson regression analysis compares disease incidences in different years in the same region, which has overcome any secular trend and seasonal variation. Spatial analysis could adjust the mutual effects among counties, which completes and confirms the results from the temporal analysis. Factors that vary over different cities in Anhui Province, such as daily meteorological data and health facilities, have been controlled as potential confounders.

Local demand-driven governmental responses had been taken to deal with the flood events. For example, affected population were transferred and relocated to safe areas as quickly as possible.52 Finance support from the government, donation from home, and abroad and rescue workers were all allocated to the flood-affected area as emergency. Buildings, transportations, medical, and educational facility reconstruction were all incorporated in post flood actions.52 However, the governmental actions were not very effective to protect people from impacts of flood events.53 A lot of defects still exist in flood disaster emergency management.54 First, the measures mentioned were not specifically for the response to floods but rather general preparation for all natural disasters. No flood alerts system linked to special disease surveillance data were developed. Evidence shows that in regions where flood events are common, early warning systems to prevent and control potential disease outbreaks may contribute to a more rapid return to baseline levels of incidence after flood.55 Second, health emergency management related to flood disaster lacks suitable indicators and evaluation.54 Extreme precipitation during flood events that overwhelm sewer systems and treatment plant can induce untreated water to be released directly into receiving waters. Therefore, monitoring environmental sanitation (particularly disinfection and water potability) and providing continuity of medical and social care by training health professionals should be integrated into flood-related action plans.49 Third, a very low proportion of population in the flood-affected area would know appropriate actions to respond to flood, for example, self-rescue behaviors and disease prevention measures during flood events, even though Huai River basin was prone to flood events. The propagations of post flood health prevention knowledge, including increased public awareness of environmental hygiene, safe and clean food, should be carried out by mass media and health-care institutions.56 In addition, clearly describing flood impact on health and identifying flood-related diseases are necessary for developing strategies to respond to future flood event and to reduce associated health burden.

Results indicate that more work need to be done to reduce the risk of infectious diseases related to flood events. Our study on the association between flood events and diseases is useful for developing models to predict future health impacts from floods, which could be useful to prevent and reduce disease outbreaks. With the advanced notice of potential risk of flood event to related diseases, those in charge of water quality can take proactive actions to reduce disease risks.57

Limitations of our study should be acknowledged. First, underreport is an issue in any communicable disease surveillance systems in China and other countries as well.5860 Because of the underreporting, there might be some underestimations in our results regarding the impact of flood events on diseases. However, the underreporting rate was assumed stable over the study period, and the impact on our analysis would be minimal. Second, we just selected 1 year flood exposure. To obtain a more general result, more flood events in the Huai River basin and other areas should be further evaluated in the future to have a better understanding of the health impact of flood events.


The study shows that malaria, diarrhea, and HAV infection are sensitive to flood in the study region. Given the increasing frequency of flood events and the considerable health burden caused by infectious diseases, prevention of these flood-associated infectious diseases is paramount.


We acknowledge Chinese Center for Disease Control and Prevention, National Meteorological Information Center of China, and Data Center for Institute of Geographic Sciences and Natural Resources Research of China for sharing with us the data needed for this study. The American Society of Tropical Medicine and Hygiene (ASTMH) assisted with publication expenses.


Financial support: This study was supported by the National Basic Research Program of China (973 Program) (grant no. 2012CB955502).

Authors' addresses: Lu Gao and Baofa Jiang, Department of Epidemiology, School of Public Health, Shandong University, Jinan City, Shandong Province, People's Republic of China, E-mails: moc.anis@udsuloag and moc.liamg@3102gnaijb. Ying Zhang, School of Public Health, China Studies Centre, The University of Sydney, New South Wales, Australia, and Center for Climate Change and Health, School of Public Health, Shandong University, Shandong Province, People's Republic of China, E-mail: ua.ude.yendys@gnahz.gniy. Guoyong Ding, Department of Occupational and Environmental Health, School of Public Health, Taishan Medical College, Shandong Province, People's Republic of China, E-mail: moc.621@351ygd. Qiyong Liu, State Key Laboratory for Infectious Diseases Prevention and Control, National Institute for Communicable Disease Control and Prevention, China Centers for Disease Control and Prevention, Beijing City, People's Republic of China, E-mail: nc.cdci@gnoyiquil.


1. Howard MJ, Brillman JC, Burkle FM., Jr Infectious disease emergencies in disasters. Emerg Med Clin North Am. 1996;14:413–428. [PubMed]
2. Guha-Sapir D, Below R, Hoyois Ph. The International Disaster Database. 2011. Available at. Accessed December 1, 2013.
3. Cangsheng Xue JJ, Deng Feng, Li Shubao. Characteristic analysis on rainstorm flood of 2007 in the Huai River drainage area in Anhui [in Chinese] J Catastrophology. 2009;24:86–91.
4. The Ministry of Water Resources of the People's Republic of China . Bulletin of Flood and Drought Disasters in China 2007 [in Chinese] Beijing, China: WaterPower Press; 2009. pp. 22–47.
5. China Meteorological Administration . Yearbook of Meteorological Disasters in China 2007 [in Chinese] Beijing, China: Meteorological Press; 2008.
6. Massad E, Forattini OP. Modelling the temperature sensitivity of some physiological parameters of epidemiologic significance. Ecosyst Health. 1998;4:119–129.
7. Martens WJ. Health impacts of climate change and ozone depletion: an ecoepidemiologic modeling approach. Environ Health Perspect. 1998;106((Suppl 1)):241–251. [PMC free article] [PubMed]
8. Watson JT, Gayer M, Connolly MA. Epidemics after natural disasters. Emerg Infect Dis. 2007;13:1–5. [PMC free article] [PubMed]
9. Kouadio IK, Aljunid S, Kamigaki T, Hammad K, Oshitani H. Infectious diseases following natural disasters: prevention and control measures. Expert Rev Anti Infect Ther. 2012;10:95–104. [PubMed]
10. Wangdi K, Singhasivanon P, Silawan T, Lawpoolsri S, White NJ, Kaewkungwal J. Development of temporal modelling for forecasting and prediction of malaria infections using time-series and ARIMAX analyses: a case study in endemic districts of Bhutan. Malar J. 2010;9:251. [PMC free article] [PubMed]
11. World Health Organization Flooding and communicable diseases fact sheet. Wkly Epidemiol Rec. 2005;80:21–28. [PubMed]
12. Buerau of Statistics of Anhui Province Statistical Yearbook of Anhui Province 2007. 2008. Available at. Accessed May 13, 2014.
13. National Meteorological Information Center Daily Surface Meteorological Data. 2011. Available at. Accessed June 16, 2013.
14. Kondo H, Seo N, Yasuda T, Hasizume M, Koido Y, Ninomiya N, Yamamoto Y. Post-flood—infectious diseases in Mozambique. Prehosp Disaster Med. 2002;17:126–133. [PubMed]
15. Lifson AR. Mosquitoes, models, and dengue. Lancet. 1996;347:1201–1202. [PubMed]
16. Zhou J, Zhang X, Chen M, Huang X, Liu A, Yang T, Tan H. Epidemiological study on hemorrhagic fever with renal syndrome in flood areas. Zhong Nan Da Xue Xue Bao Yi Xue Ban. 2011;36:223–228. [PubMed]
17. Saenz R, Bissell RA, Paniagua F. Post-disaster malaria in Costa Rica. Prehosp Disaster Med. 1995;10:154–160. [PubMed]
18. Gagnon AS, Smoyer-Tomic KE, Bush AB. The El Nino southern oscillation and malaria epidemics in South America. Int J Biometeorol. 2002;46:81–89. [PubMed]
19. Warraich H, Zaidi AK, Patel K. Floods in Pakistan: a public health crisis. Bull World Health Organ. 2011;89:236–237. [PubMed]
20. Marx MA, Rodriguez CV, Greenko J, Das D, Heffernan R, Karpati AM, Mostashari F, Balter S, Layton M, Weiss D. Diarrheal illness detected through syndromic surveillance after a massive power outage: New York City, August 2003. Am J Public Health. 2006;96:547–553. [PubMed]
21. Gao L, Zhang Y, Ding G, Liu Q, Zhou M, Li X, Jiang B. Meteorological variables and bacillary dysentery cases in Changsha City, China. Am J Trop Med Hyg. 2014;90:697–704. [PMC free article] [PubMed]
22. Vollaard AM, Ali S, van Asten HA, Widjaja S, Visser LG, Surjadi C, van Dissel JT. Risk factors for typhoid and paratyphoid fever in Jakarta, Indonesia. JAMA. 2004;291:2607–2615. [PubMed]
23. McMichael AJ, Campbell-Lendrum DH, Corvalán CF, Ebi KL, Githeko AK, Scheraga JD, Woodward A. Climate Change and Human Health Risks and Responses. Geneva, Switzerland: World Health Organization; 2003.
24. Ahern M, Kovats RS, Wilkinson P, Few R, Matthies F. Global health impacts of floods: epidemiologic evidence. Epidemiol Rev. 2005;27:36–46. [PubMed]
25. Trevejo RT, Rigau-Perez JG, Ashford DA, McClure EM, Jarquin-Gonzalez C, Amador JJ, de los Reyes JO, Gonzalez A, Zaki SR, Shieh WJ, McLean RG, Nasci RS, Weyant RS, Bolin CA, Bragg SL, Perkins BA, Spiegel RA. Epidemic leptospirosis associated with pulmonary hemorrhage—Nicaragua, 1995. J Infect Dis. 1998;178:1457–1463. [PubMed]
26. Hubalek Z, Zeman P, Halouzka J, Juricova Z, St'ovickova E, Balkova H, Sikutova S, Rudolf I. Antibodies against mosquito-born viruses in human population of an area of Central Bohemia affected by the flood of 2002. Epidemiol Mikrobiol Imunol. 2004;53:112–120. [PubMed]
27. Standing Committee of the National People's Congress . Beijing, China: China legal publishing house; 2013. Law of the People's Republic of China on Prevention and Treatment of Infectious Diseases (2013 Amendment) [in Chinese]
28. Ministry of Health of the People's Republic of China Emergency Events and Regulation of the Notifiable Disease Surveillance System. 2006. [in Chinese] Available at. Accessed December 22, 2014.
29. D'Souza RM, Becker NG, Hall G, Moodie KB. Does ambient temperature affect foodborne disease? Epidemiology. 2004;15:86–92. [PubMed]
30. Zhang Y, Bi P, Hiller JE, Sun Y, Ryan P. Climate variations and bacillary dysentery in northern and southern cities of China. J Infect. 2007;55:194–200. [PubMed]
31. Liu X, Jiang B, Gu W, Liu Q. Temporal trend and climate factors of hemorrhagic fever with renal syndrome epidemic in Shenyang City, China. BMC Infect Dis. 2011;11:331. [PMC free article] [PubMed]
32. Campbell-Lendrum D, Woodruff R. In: Climate Change—Quantifying the Health Impact at National and Local Levels. Prüss-Üstün A, Corvalán C, editors. Geneva, Switzerland: World Health Organization (WHO Environmental Burden of Disease Series, No. 14).; 2007.
33. Guan P, Huang D, Guo J, Wang P, Zhou B. Bacillary dysentery and meteorological factors in northeastern China: a historical review based on classification and regression trees. Jpn J Infect Dis. 2008;61:356–360. [PubMed]
34. Guo C, Yang L, Ou CQ, Li L, Zhuang Y, Yang J, Zhou YX, Qian J, Chen PY, Liu QY. Malaria incidence from 2005–2013 and its associations with meteorological factors in Guangdong, China. Malar J. 2015;14:116. [PMC free article] [PubMed]
35. Kelly-Hope LA, Alonso WJ, Thiem VD, Anh DD, Canh do G, Lee H, Smith DL, Miller MA. Geographical distribution and risk factors associated with enteric diseases in Vietnam. Am J Trop Med Hyg. 2007;76:706–712. [PubMed]
36. Lee J, Wong DWS. Statistical Analysis with ArcView GIS. New York, NY: John Wiley and Sons, Inc; 2000.
37. Alonso WJ, Acuna-Soto R, Giglio R, Nuckols J, Leyk S, Schuck-Paim C, Viboud C, Miller MA, McCormick BJ. Spatio-temporal patterns of diarrhoeal mortality in Mexico. Epidemiol Infect. 2012;140:91–99. [PubMed]
38. Wang T, Xue F, Chen Y, Ma Y, Liu Y. The spatial epidemiology of tuberculosis in Linyi City, China, 2005–2010. BMC Public Health. 2012;12:885. [PMC free article] [PubMed]
39. Zhou X. Spatial Epidemiology. Beijing, China: Science Press; 2009.
40. McCormick BJ, Alonso WJ, Miller MA. An exploration of spatial patterns of seasonal diarrhoeal morbidity in Thailand. Epidemiol Infect. 2012;140:1236–1243. [PubMed]
41. Tsai PJ, Lin ML, Chu CM, Perng CH. Spatial autocorrelation analysis of health care hotspots in Taiwan in 2006. BMC Public Health. 2009;9:464. [PMC free article] [PubMed]
42. M P The interpretation of statistial maps. J R Stat Soc, B. 1948;10:243–251.
43. Neimann J, Engberg J, Molbak K, Wegener HC. A case-control study of risk factors for sporadic Campylobacter infections in Denmark. Epidemiol Infect. 2003;130:353–366. [PubMed]
44. Anselin L, Syabri I, Kho Y. GeoDa: an introduction to spatial data analysis. Geogr Anal. 2006;38:5–22.
45. Asian Development Bank, Government of Pakistan, World Bank . Pakistan Floods 2010: Preliminary Damage and Needs Assessment. Washington, DC: World Bank; 2010.
46. Wade TJ, Sandhu SK, Levy D, Lee S, LeChevallier MW, Katz L, Colford JM., Jr Did a severe flood in the Midwest cause an increase in the incidence of gastrointestinal symptoms? Am J Epidemiol. 2004;159:398–405. [PubMed]
47. Heller L, Colosimo EA, Antunes CM. Environmental sanitation conditions and health impact: a case-control study. Rev Soc Bras Med Trop. 2003;36:41–50. [PubMed]
48. Centers for Disease Control and Prevention (CDC) Tropical Storm Allison rapid needs assessment—Houston, Texas, June 2001. MMWR. 2002;51:365–369. [PubMed]
49. Duclos P, Vidonne O, Beuf P, Perray P, Stoebner A. Flash flood disaster—Nimes, France, 1988. Eur J Epidemiol. 1991;7:365–371. [PubMed]
50. Ma SL, Tang QL, Liu HW, He J, Gao SH. Correlation analysis for the attack of bacillary dysentery and meteorological factors based on the Chinese medicine theory of Yunqi and the medical-meteorological forecast model. Chin J Integr Med. 2013;19:182–186. [PubMed]
51. Patrick ME, Christiansen LE, Waino M, Ethelberg S, Madsen H, Wegener HC. Effects of climate on incidence of Campylobacter spp. in humans and prevalence in broiler flocks in Denmark. Appl Environ Microbiol. 2004;70:7474–7480. [PMC free article] [PubMed]
52. Xinhua News Agency Special Report of 2007 Huai River Flood Control. 2007. [in Chinese] Available at. Accessed July 2, 2015.
53. China Ministry of Water Resources Notice of the Ministry of Water Resources on the Strengthening of the Construction of Flood Control Suggestions for Huaihe Basin from 2001 to 2010. 2002. [in Chinese] Available at.
54. Yuan X. The Study of Health Emergency Management in Floods Events in China—Take Hubei Province as an Example. Wuhan, China: Southern Medical University; 2012. [in Chinese]
55. Milojevic A, Armstrong B, Hashizume M, McAllister K, Faruque A, Yunus M, Kim Streatfield P, Moji K, Wilkinson P. Health effects of flooding in rural Bangladesh. Epidemiology. 2012;23:107–115. [PubMed]
56. Chen Xiaoxuan FZ. Evaluation for the effect of public health education on disease prevention in flood-hit areas [in Chinese] Chinese J Public Health Manage. 2012;28:654–655.
57. Auld H, MacIver D, Klaassen J. Heavy rainfall and waterborne disease outbreaks: the Walkerton example. J Toxicol Environ Health A. 2004;67:1879–1887. [PubMed]
58. Zhang Y, Bi P, Hiller JE. Projected burden of disease for Salmonella infection due to increased temperature in Australian temperate and subtropical regions. Environ Int. 2012;44:26–30. [PubMed]
59. Ding G, Zhang Y, Gao L, Ma W, Li X, Liu J, Liu Q, Jiang B. Quantitative analysis of burden of infectious diarrhea associated with floods in northwest of Anhui Province, China: a mixed method evaluation. PLoS One. 2013;8:e65112. [PMC free article] [PubMed]
60. Zhang Y, Bi P, Hiller J. Climate variations and salmonellosis transmission in Adelaide, South Australia: a comparison between regression models. Int J Biometeorol. 2008;52:179–187. [PubMed]

Articles from The American Journal of Tropical Medicine and Hygiene are provided here courtesy of The American Society of Tropical Medicine and Hygiene