The changing epidemiological characteristics of severe fever with thrombocytopenia syndrome in China, 2011–2016

Severe fever with thrombocytopenia syndrome (SFTS) is emerging and the number of SFTS cases increased year by year in China. In order to explore the epidemiology trend, we analyzed the changing epidemiological characteristics of SFTS cases in different years and compare characteristics in different provinces. From 2011 to 2016, a total of 5360 laboratory-confirmed SFTS cases were reported and annual case numbers increased year by year. Most SFTS cases occurred in individuals aged between 40 years and 80 years (91.57%), but age distributions of SFTS cases in different years were significantly different and the median ages increased slightly year by year. The numbers of affected counties from 2011 to 2016 increased sharply from 98 to 167. Of note, the seasonal distributions of SFTS cases in different provinces were significantly different (Fisher = 712.157, P = 0.000) and provinces in south regions showed earlier epidemic peak and longer epidemics durations. The median time from illness onset to confirmation of different years was significantly different (χ2 = 896.088, P = 0.000) and it peaked in 2014. Furthermore, case fatality rate was associated with province, year, and age of SFTS cases. These results may be helpful for authorities to better preventive strategy and improve interventions against SFTS.


Results
From 2011 to 2016, a total of 5360 laboratory-confirmed cases were reported to China Information System for Diseases Control and Prevention. Annual case numbers increased year by year, with the highest recorded in 2016 (1306 cases, Fig. 1). Of the total SFTS cases, 2509 cases were male and 2851 cases were female. There were more male cases than female cases in 2011, but there were less male cases than female cases from 2012 to 2016.The male-to-female ratio of different years was not significantly different (χ 2 = 9.526, P = 0.090).
The majority of SFTS cases were farmers (87.91%, 4712/5360) and the occupation distribution of SFTS cases in different years were similar (χ 2 = 15.552, P = 0.113). Most SFTS cases occurred in individuals aged between 40 years and 80 years (91.57%, Table 1). The median ages of SFTS cases from 2011 to 2016 were 61 years, 60 years, 62 years, 61 years, 63 years, and 63 years, respectively. However, age distributions of SFTS cases in different years were significantly different (Kruskal-Wallis statistic = 23.334, P = 0.000) and the median ages slightly increased from 2011 to 2016.
During 2011-2016, 98.00% (5253/5360) of SFTS cases were reported from April to October with a peak in May, June, and July (Fig. 3). The seasonal distributions of different years were significantly different (Fisher = 276.845, P = 0.000) and the peaks of reported SFTS cases from 2011 to 2016 occurred in July, May, June, May, May, and May, respectively. Of note, the seasonal distributions of SFTS cases in different provinces were significantly different (Fisher = 721.157, P = 0.000). SFTS cases in Liaoning Province showed shorter epidemic periods than other province (Table 2). Moreover, SFTS cases in Zhejiang Province, Hubei Province, Anhui Province, and Henan Province showed earlier peaks than those in Liaoning Province, Shandong Province and Jiangsu Province (April-October vs May-October, Table 2).
The median time from illness onset to confirmation of SFTS cases from 2011 to 2016 were 7 days, 6 days, 6 days, 27 days, 10 days and 12 days, respectively. Notably, the median time from illness onset to confirmation of different years was significantly different (χ 2 = 896.088, P = 0.000) and it peaked in 2014.
A total of 343 deaths were reported in China from 2011 to 2016 and the average case fatality rate was 6.40%. According to results of single variable logistic regression analysis, there were significant differences in year, province, age and gender. As shown in Table 3, the case fatality rates of SFTS cases in different years were significantly different and it decreased from 2014 to 2016. The case fatality rates of SFTS cases in different provinces were also significantly different and the lowest fatality rate recorded in Henan. The case fatality rate significantly increased with the increase of age of SFTS cases (P = 0.000).

Discussion
In this study, surveillance data on SFTS from 2011 to 2016 was used to investigate changing epidemiological characteristics of SFTS. We found that the number of SFTS increased year by year. Two factors may contribute the result. First, SFTSV might has been spread to more areas through humans, small mammals, ticks, or birds 14 . Therefore, people in more areas have opportunities to be infected with SFTSV and more SFTS cases occurred. Second, some physicians lacked experience of SFTS diagnosis when SFTSV was identified. Some SFTS cases were misdiagnosed and the reported cases were only the tip of the iceberg. Due to the high fatality rate, more attention have been paid by health departments and some measures have been conducted to improve the capacity of   diagnosis and response to SFTS. For example, all municipal center for disease control and prevention established detection methods for SFTSV in Zhejiang Province in 2014. The improvement of diagnosis and detection capacity may influence the number of reported SFTS cases. Nevertheless, the increase number of SFTS cases inform that more cases will be reported in future which prompt the research of prevention methods for SFTSV infection including the SFTSV vaccine. Occupation distributions of SFTS cases in different years were similar and the majority of cases were farmers who lived in rural areas. As SFTSV is believed to be transmitted by Haemaphysalis longicornis ticks 15 , higher tick density in rural areas and more outdoor activities of farmers may contribute to the result. Farmers are the key population for the prevention of SFTS.
Although most SFTS cases aged between 40 and 80, the age distributions of cases in different years were significantly different and the median age slightly increased. These high prevalence age group are in agreement with corresponding data from other studies [16][17][18] . Elderly people might have decreased immune function and  comorbidities with chronic diseases. If they are infected with SFTSV, they may get severe diseases. It is thereby that more elderly cases were identified. Ding et al. also reported that people at all age groups were susceptible to SFTSV, but only old people got severe disease and needed to be hospitalized or even died of SFTSV infection 19 . Besides this factor, demographic features of residents in areas where SFTS cases occurred may also influence the age distribution. Generally, young adults in rural areas go to cities to earn money and then return to their hometown for the spring festival. As a result, the majority of residents of rural areas between the months of March and November are elderly people. Furthermore, high prevalence of SFTSV among elderly people indicate that SFTSV is a new virus and the immunity rate is not very high among them.
SFTSV was firstly identified in Henan and Hubei, then it was identified in Anhui, Shandong, Jiangsu, Zhejiang, and Liaoning. Up to date, the overwhelming majority of SFTS cases were reported in these 7 provinces which all locate in the central part of China. We speculate that the weather in these 7 provinces is more suitable for the   Table 4. Multivariable analysis on factors associated with SFTS fatal outcome.
survival and breeding of ticks, and the transmission of SFTSV. But further study is needed to study the relation between meteorological factors and SFTSV incidence. Another factor may all influence the geographical distribution. More attentions have been paid, and more researches have been done in these 7 provinces where the capacity of SFTSV detection is developed.
Although the number of affected province in different years were similar, the number of affected counties increased sharply every year and SFTS cases were also reported in other provinces besides its endemic areas. These results highlight that the fact that the geographic range of SFTSV has apparently expanded in China, which is valuable information for consideration in national planning on SFTS control and prevention. More investigation should be conducted to confirm whether SFTS is endemic in the provinces where sporadic cases were reported.
Most SFTS cases occurred from April to October, but the seasonal distributions of different provinces were significantly different. Provinces in south regions showed earlier epidemic peak and longer epidemics durations. These are also related to the climate of these provinces. These results suggest that control measures should be conducted during different periods in different provinces.
To our great disappointment, the interval from illness onset to confirmation from 2011 to 2016 were not shortened. These were consistent with corresponding results of another study 20 . This may be due to SFTS patients' low incomes and poor access to medical care because most of them were from rural areas. Rural residents commonly visit local hospitals or outpatient clinics in villages or towns. These hospitals do not have the capability to identify and cure SFTSV infection. They have to go to urban areas for hospitals with better facilities for further care. The majority of hospitals don't have the capacity to detect SFTSV and most samples are transported to municipal centers for disease control and prevention (CDC) or provincial CDC for testing. These factors prolong the interval from illness onset to confirmation. However, the interval from illness onset to confirmation is very important for the cure of SFTS cases. Sun et al. reported that an increase in interval from illness onset to confirmation by 3 days was associated with fatality with an OR of 1.996 21 . These inform that detection capability of SFTSV should be established not only in centers for diseases control and prevention but also in different hospitals to decrease the case fatality rate.
The fatality rate decreased from 2011 to 2016, but it remained very high in 2016. In the analysis of factors associated with SFTS fatal outcome, fatality rates of cases in different years and different provinces were significantly different. Provinces with higher number of SFTS cases showed lower fatality rate. For example, Henan presented the highest number of SFTS cases and the lowest fatality rate. One reason is that physicians in these provinces are more experienced to cure SFTS, so cases are more probable to be cured in these provinces. Another reason is that the detection capability is developed in these provinces. Not only severe cases but also mild cases are identified in these provinces. As a result, more cases were identified and fatality rates were low in these provinces. The high fatality rate suggest that it is urgent to study effective treatments for SFTS.
Previously studies have reported that case fatality rate of SFTS increases with age 10, 22-24 . Our study also indicated that age was related to the fatality rate. Some factors associated with age including decreased immune function and comorbidities with chronic diseases may be relative to SFTS fatal outcome. Nevertheless, our findings further suggest that more intense treatment should be administered to elderly SFTS cases when they were hospitalized.
According to our results, we suggest that some measures should be done to prevent and control SFTS. First, health education should be enhanced. People should avoid go to hilly areas where ticks exist in epidemic seasons. If they have to go to these areas, they should dress light color clothes and check whether ticks bite them when they leave these areas. Second, hospitals in SFTS endemic areas should improve capacity of SFTSV detection. This is helpful for the early diagnosis and the decrease of fatality of SFTS patients. Finally, some measures should be conducted to decrease tick density in endemic areas during epidemic seasons.
There are several limitations in our study. First, the data used were collected from China Information System for Diseases Control and Prevention which is passive surveillance. Some factors including detection capability, reporting methods, availability of health facilities may influence the data quality. Second, there are too many floating population in some provinces of China. They go to urban areas and come back hometowns irregularly. So we can't get the real number of people who lived in rural areas during high risk seasons which is disadvantageous to analyze incidence of SFTS among people with different demographic characteristic.
To the best of our knowledge, this is the first report on the changing epidemiological characteristics of SFTS in China. We found that the number of SFTS cases increased, the affected areas expanded and some demographic characteristics changed year by year indicating that surveillance and control strategies should be adjusted to account for these changes. Unfortunately, surveillance and control strategies are largely inadequate in China. For example, there are no national sentinel vector surveillance project for ticks and no national control program for ticks in China. Therefore, it is essential for China to put more effort into ticks surveillance and control. The increased interval from illness onset to confirmation further suggests that it is urgent to improve the diagnosis capability of physicians for early diagnosis. More researches should be conducted and more attention should be paid to respond the changed characteristics of SFTS.