The Study Of Influence Pattern For Floods On The Outbreak Of Hemorrhagic Fever With Renal Syndrome

[Background]In recent years, under the background of global climatic change, the frequent occurrence of flood disaster has been one of the significant environmental problems for human beings, which is also the major factor resulting for the change in epidemic characteristic of natural endemic infectious disease. Hemorrhagic Fever with Renal Syndrome (HFRS) is a typical natural endemic disease, which spreads by rodents infected by Hantavirus. In China, Apodemus inhabited in field and Rattus inhabited in residential area are the major source of infection. The ebb and flow of rodent population with the seasons change can cause the regularity of HFRS outbreak, which leads to dominant peak type in different seasons. The flood disaster could affect the number of rodents by affecting the source of food and habitat environment of rodents, and then change the epidemic characteristics of HFRS. The previous studies indicated that the effect of floods on the epidemic characteristic of HFRS was closely related to the degree of floods, the climatic regularity of epidemic focus and the feature of landscape. However, there were some disadvantages in these studies, such as short study time, limited study area and not considering the different influencing mechanism of floods on the epidemic of HFRS with different type of rodents (Apodemus type and Rattus type). Shandong Province is an HFRS epidemic area with high incidence. The reported cases account to 1/3 of the whole cases in China and the type of epidemic focus go through the evolution process of "Apodemus type-mixed type-Rattus type-Apodemus type". Moreover, due to the vast geography, the climatic feature and geomorphology type are geographically different in Shandong Province, which providing advantaged condition for conducting horizontal comparative study of environmental epidemiology among different regions.[Objective]We choose Shandong Province as the study area and county (county-level city, district) as the study units. We plan to perform the method of case-crossover design to quantify and discuss the effect of floods on the outbreak of HFRS with different type in different geographic environmental regions based on the data of HFRS reported cases, meteorological monitoring data for long time and the fact of different climatic characteristic and the type of rodents.[Methods]The data information concludes monthly reported HFRS morbidity data and meteorological data in county (county-level city, district) level from 1974 to 2012 in Shandong Province. Firstly, we divide the monthly HFRS morbidity data into two parts-Rattus type and Apodemus type according to the epidemic season to calculate the morbidity data of the two types by year, and define the outbreak years of HFRS. Then, we choose one HFRS outbreak year and the next two no-outbreak years to combine a matched group of one to two using the method of bidirectional symmetric case-crossover design. We quantify the effect of floods on the outbreak of HFRS by comparing the difference of floods between the risk period before the outbreak of HFRS and the control period before the no-outbreak of HFRS. Using the statistics method of conditional logistic regression, we estimate the OR (odds ratio) and CI (confidence interval) of floods of spring, summer, autumn and winter in the one-year risk period on the outbreak of HFRS with Apodemus type and Rattus type respectively, after controlling the effect of relative humid and sunshine hours on the outbreak of HFRS. Meanwhile, we should analyze and compare the effect of floods on the outbreak of HFRS of Rattus type and Apodemus type in different areas with different geographical environment.[Results]1. Floods in autumn could decrease the risk of the outbreak of HFRS of Rattus type in the next year by 51.4%(OR=0.486,95%CI:0.309-0.765). Floods in spring could decrease the risk of the outbreak of HFRS of Rattus type the same year by 46.5% (OR=1.465,95%CI:1.018-2.107). There was no significant influence of floods in summer on the outbreak of HFRS of Rattus type the same year.2. Severe floods in summer could decrease the risk of the outbreak of HFRS of Apodemus type the same year by 49.6%(OR=0.504,95%CI:0.358-0.712), while no significant effect for light floods (P>0.05). Light and severe floods in autumn could decrease the risk of the outbreak of HFRS of Apodemus type the same year by 53.7% (OR=0.463,95%CI:0.293-0.730) and 60.0%(OR=0.400,95%CI:0.201-0.793), respectively.3. The effect of floods on the outbreak of HFRS is not coincident in different geographical areas. Mountain area in the middle and south of Shandong province: Floods in spring could increase the risk of the outbreak of HFRS of Rattus type the same year by 1.347 times (OR=2.347,95%CI:1.244-4.428). Floods in autumn could decrease the risk of the outbreak of HFRS of Apodemus type the same year by 54.2%(OR=0.458,95%CI:0.274-0.766). Plain area in the west and north of Shandong province:Floods in autumn could decrease the risk of the outbreak of HFRS of Rattus type in the next year by 48.5%(OR=0.515,95%CI:0.309-0.858). Severe floods in summer and floods in autumn could decrease the risk of the outbreak of HFRS of Apodemus type the same year by 73.4%(OR=0.266,95%CI:0.079-0.895) and 62.8%(OR=0.372,95%CI:0.147-0.941). Hills area in east of Shandong province: Floods in spring could increase the risk of the outbreak of HFRS of Rattus type the same year by 32.8%(OR=1.328,95%CI:1.081～1.631), while there was no significant influence of floods on the outbreak of HFRS of Apodemus type(P>0.05).[Conclusions]1. Light floods in autumn could decrease the risk of the outbreak of HFRS of Rattus type the next year, while light floods in spring could increase the risk of the outbreak of HFRS of Rattus type the same year.2. Severe floods in summer, light and severe floods in autumn could decrease the risk of the outbreak of HFRS of Apodemus type the same year.3. The effect of floods on the outbreak of HFRS is not coincident in different geographical areas.4. The effect of floods on the outbreak of HFRS is related to the type of epidemic focus, climatic characteristics and the type of rodents.[Innovations and limitations]The study generalized the idea of case-crossover design from individual-level to population-level and performed stratified analysis by the type of geographical environment and rodents, making the results more reliable. However, there were some limitations in this study:the definition of flooding disaster only depending on monthly precipitation, not considering the duration of floods and damage area, the selection of control period while the years of HFRS outbreak were continuous and the bias of only selecting Shandong province as study area, which needs a more comprehensive study in the future.