| BackgroundHemorrhagic fever with renal syndrome (HFRS), with characteristics of fever, hemorrhage, kidney damage and hypotension, is an important infectious disease caused by different species of hantaviruses. The disease mainly distributes in Eurasia, but the HFRS virus spread almost all over the continents of the world. The disease is of high mortality and widely distributed in China, and the country has been seriously infected in the last50years with an exception of Qinghai and Taiwan. In China, HFRS is caused mainly by two types of hantaviruses, Hantaan virus (HTNV) and Seoul virus (SEOV), each of which has co-evolved with a distinct rodent host. HTNV is associated with Apodemus agrarius, whereas SEOV, which causes a less severe form of HFRS, is associated with Rattus norvegicus. It is considered that the number of HFRS cases in China accounts for90%of the total cases worldwide. In mainland China, a total of1,585,435HFRS cases had been reported from1950to2010. During this period, HFRS was reported in29out of31provinces. The incidence and death rate of the disease in our country is the highest in the world, though the fatality rate has reduced from14.22/100,000in1969to1/100,000in1995-2007. HFRS is still one of the national infectious diseases that receive great attention in the country.The Chinese Center for Disease Control and Prevention (CDC) established the National Notifiable Disease Surveillance System in2004. HFRS is one of the disease that is under surveillance and reported in this system, which makes the surveillance data for HFRS more accurate and comprehensive. Rodent is the main source of infection for this disease, and the proof and deratization of rodents has been carried out in infected areas to control HFRS. However it is difficult to achieve satisfactory results due to the limitation of natural conditions, differences of the epidemic situations, and the migratory characteristics of rodents. Vaccination has become the major prevention measure in infected areas since1990s, but it has been only carried out for those seriously infected people in seriously infected areas because of the high cost. Scientific and reliable forecasting of HFRS epidemic based on disease surveillance is of great significance in the control of rodents and vaccination. Till now, the spatial distribution pattern of HFRS in China since2004is not clear, and the present research results are not so detailed. A systematic study, focusing on the HFRS epidemic situation in China since2004, is carried out in this study.China is the most seriously infected country with HFRS in the world. A large number of researches, focusing on time as well as the special distribution, have been taken placed to forecast the disease. It is the seriously infected areas such as Shandong Province and the northeast provinces that have received great attention in those studies. There is no quantitative prediction research of HFRS covering the whole area of China. ARIMA model will be used in this study to forecast the trend of HFRS in the next few years in China.Liaoning Province is one of the seriously infected areas in China. The incidence rate ranked the third in2002, the second in2003, and the first in2004and2005. The figure reached the highest (13.05/100,000) in2004, and it became an urgent public health problem. Therefore, to prevent HFRS in Liaoning Province effectively, it is important to get a comprehensive idea of the special distribution of HFRS, including the spatial diffusion trends, spatial correlation and aggregation, and their relationship with climate factors. The incidence rate has ranked the top in recent years in Shenyang, the capital of Liaoning Province, since the first case of HFRS in1958. It has become a serious public health problem as the13districts and counties have all been infected. Shenyang was set to be another study site where the research of the impact of climate factors on HFRS was carried out in a maller scale.As a rodent-borne disease with a seasonal distribution, external environmental factors including climate factors may play a significant role in its transmission. Studies in different areas of China and other countries have suggested that climate factors, such as temperature, precipitation, and relative humidity, may influence the incidence of HFRS. The role of climate factors in such transmission may vary, even leading to the opposite conclusions, given the large variations in climate types, ecological characteristics, population immunity, public health intervention measures, and socioeconomic status in the different regions.Epidemic monitoring data of31provinces, municipalities, and autonomous regions of China, obtained from the epidemic report system of CDC, has been used and studied to display the latest spatio-temporal distribution patterns of HFRS. The surveillance data of HFRS epidemic, climate factors and rodent density of Liaoning Province and Shenyang was collected to study the relationship between HFRS epidemic and climate factors.Objectives1. To establish the GIS database of HFRS epidemics, and to detect the latest spatio-temporal distribution patterns of HFRS and excessive hazard areas in China.2. To forecast the incidence trends in future three years in China.3. To detect the epidemic dynamics and spatio-temporal distribution patterns of HFRS and to expound the laws of spatial-temporal dynamic evolution of HFRS foci in Liaoning Province, so as to provide scientific evidence for working out practical control and prevention measures.4. To understand the impact mechanism of climate factors on HFRS epidemics in Liaoning and Shenyang, so as to provide analytica methods for further expounding the relationship between HFRS epidemic and climate factors.MethodsIn this study, the surveillance data of HFRS (national data:2004-2009; data in Liaoning Province:1985-2010; Shenyang City:1984-2010) were collected from the national surveillance system (China Information System for Disease Control and Prevention) and Liaoning Center for Disease Control and Prevention and Shenyang Center for Disease Control and Prevention and digital map at the prefecture-level (1:250000) of China and county-level (1:250000) of Liaoning Province were processed. The county-level data for the temperature, precipitations, relative humidity, air pressure, wind velocity, sunshine duration, and rodent density was extracted through the spatial analysis technology, and the corresponding GIS database was established. Descriptive analysis, correlation analysis, time-series analysis, spatial analysis, panel data analysis, and principal component regression were used to process data. Softwares used in this study include SAS9.1, ArcGIS9.2, GeoDa095i, STATA Version11.0, and SPSS13.0.Results1.100,886cases of HFRS had been reported from2004to2010in China with an annual incidence rate of1.11/100,000. There were HFRS cases reported every month, and the distribution was closely related to seasons, with a high incidence in winter and spring and a low incidence in summer and autumn.2. The cumulative incidence rate of HFRS from2004to2010in357cities ranged from0/100,000to180.0840/100,000.There ware53cities with0incidence rate,accounting for4.36%Of the country's total populationï¼›119cities with an incidence rate lower than1/100,000,accounting for35.34%of the country's total population and1.75%of the country's total incidenceï¼›127cities with an incidence rate ranged from1/100,000to10/100,000, accounting for44.03%of the country's total population and21.16%of the country's total incidenceï¼›43cities with an incidence rate ranged from10/100,000to50/100,000,accounting for13.52%of the country's total population and43%of the countyr's total incidenceï¼›12cities with an incidence rate ranged from50/100,000to100/100,000,accounting for2.32%of the country's total population and24.58%of the country's total incidenceï¼›and3cities with an incidence rate higher than100/100,000,accounting for0.43%of the country's total population and9.51%of the country's total incidence.3.An ARIMA(1,2,1)model was built based on the annual HFRS incidence data Of China from1950to2010.It was testined that the model had good fitting goodness. It was predicted by the model that the incidence of HFRS in the year2011,2012and2013would be0.7165/100,000,0.7155/100,000and0.7198/100,000,respectively, indicating that HFRS incidence would experierice a slight rise during the flext three years.4.There were44,793cases of HFRS reported in Liaoning Province from1985to2010.The incidence rate was much lower than the national average level before1996, and it was signincantly higher than the national average level since1997till now.The lowest incidence rate in Liaoning Province was0.8628/100,00in1990,and the highest was13.0747/100,000in2004,and it declined year aRer year from then on.5.The average incidellce rate of HFRS in Liaoning Province from1985to2010was uneven.Infected areas of high incidellce rate were mainly distributed in the mid-eastem and southwestem part of Liaoning Province with an expanding range of HFRS cases.Spatial alltocorrelation alialysis for annual incidence of HFRS and annualized average incidence in Liaoning Province from2004to2010showed that the Moran's I statistic was significant from2004to2010at a significance level of0.05,namely,global Moran's I was unequal to0in the interval(0,1),renecting positive spatial autocorrelation.According to LISA Visualization anaIysis,HFRS incidence of Benxi,Shenyang,and Huludao lied in high-high region constituting high incidence areas,but Kangping and Zhangwu lied in low-low region constituting low incidence areas. 6. The results of multifactor Poisson Panel Data Analysis showed that the monthly average temperature of the current month, the monthly average temperature with1month lag, the monthly precipitation with2months lag, the monthly average relative humidity, the average relative humidity with1month lag and the average relative humidity with2months lag were significantly negatively associated with the HFRS incidence in Liaoning Province (P<0.001), and IRR were less than1, indicating that the rise of these factors may reduce the HFRS incidence.7. Spearman correlation analysis showed that the monthly average pressure and monthly rodent density were positively correlated with the incidence rate of HFRS in Shenyang (P<0.000); the correlation between average wind velocity and HFRS incidence rate was not statistically significant (P=0.208); and other climate factors were negatively correlated with the incidence rate of HFRS (P<0.000). Cross-correlation analysis showed that climate factors had lag effect on the monthly incidence rate of HFRS in Shenyang. The autocorrelation coefficient and partial autocorrelation coefficient with first-order delay (AC=0.566, PAC=0.566) of HFRS incidence of Shenyang were significantly greater than the other coefficient with different orders delay, implicating that the monthly HFRS incidence in Shenyang from1984to2010existed first-order delay autocorrelation.8. Three principal components were extracted with a cumulative contribution rate of74.763%. Component1represented APO, MT1, MaxT1, MinT1, MAP1, SD1, SD2, RH5, RH6, and AP7; Component2represented AP0, SD0, MAP3, SD5, SD6, and RH7; and Component3represented SD0, SD2, and RH5.9. The model that included three principal components, rodent density, and first-order delay item was better with a bigger R2(R2=0.59). Residuals concentrated in the zero value of the upper and lower. The white noise test showed that P values of LB statistics with6-order delay,12-order delay, and18-order delay were all significantly less than0.05.Conclusions1. The incidence of HFRS in China declined from2004to2010and was distributed spatially uneven, both dispersedly and relatively concentrated. There were cases in every province and city, with an exception of Xijiang and Hainan Province. More cases occurred in the northern part, especially in the three northeastern provinces. Shandong, Shanxi and Hebei. Fewer cases were found in southern provinces and they were distributed mainly in Zhejiang, Hunan and Jiangxi.2. The distribution of highly risky areas was quite the same as that of the major infected areas, but was much larger, including some southern regions, such as some areas of Zhejiang and Sichuan. Compared with previous studies, the present-day risky areas of HFRS reduced to a certain extent, which partially showed the effectiveness of the control of HFRS in China.3. It was forecasted that there would be a slightly rising of HFRS incidence in China in the next three years. According to the published data in2011, the results were reliable, suggesting that relevant departments should take measures to monitor and control the disease.4. There were the following features of HFRS incidence in Liaoning Province; firstly, the occurrence of HFRS experienced an evolution of high-low-high-low from1985to2010; secondly, there was a change from Apodemus endemic areas to Mus musculus endemic areas; thirdly, the distribution was wide as well as relatively concentrated, and was expanding gradually.5. The annual incidence and annualized average incidence of HFRS in Liaoning Province from2004to2010showed a statistically significant positive spatial autocorrelation and aggregation of distribution. This positive spatial autocorrelation mainly reflected this kind of space contact form that regions with low observed values were surrounded by regions with low observed values (low-low). The spatial distribution of HFRS incidence in Benxi, Shenyang, and Huludao is not only a statistically significant cluster, but also a high incidence cluster. The results can make the allocation of health resources more rational and the control and prevention of HFRS in Liaoning Province more targeted.6. Excessive rise of the monthly average temperature of the current month, the monthly average temperature with1month lag, the monthly precipitation with2months lag, the monthly average relative humidity, the average relative humidity with1month lag and the average relative humidity with2months lag may reduce the HFRS incidence of Liaoning Province.7. Climate factors, which did influence the HFRS incidence rate, couldn't fully reflect the incidence information. Therefore, other factors, such as social demographic factors, vaccine factors, etc., should be taken into consideration.Significance and innovation 1. It is the first time that the incidence of HFRS in China is forecasted by mathematical models.2. The evolution of the epidemic area of Liaoning Province was clarified, the expansion pattern of HFRS cases was primarily worked out, highly infected areas of HFRS in Liaoning Province was identified, and therefore, a foundation was made for the identification of key control areas and rational allocation of health resources.3. A variety of methods were used to analyze the quantitively relationship between the climate factors and the incidence of HFRS in Liaoning Province and Shenyang City. A joint mechanism among various climate factors was revealed, and meanwhile, it was testified that climate factors were not the decisive factors for HFRS in Liaoning Province.
|
PDF Full Text Request