Study On Spatiotemporal Distribution Of Tick-Borne Encephalitis In China

Background: Tick-borne encephalitis(TBE), also called forest encephalitis in China, is an acute infectious zoonotic disease caused by the TBE virus(TBEV), which mainly attacks the central nervous system of infected patients. The disease is widely endemic in Europe, and in north-eastern Asia including eastern Russia, northern Japan,northern Monglia and northeastern China. Despite the availability of effective vaccines,approximately 10,000–12,000 cases are reported in the world annually. In recent years,the number of human TBE cases reported in China has increased notably, and endemic areas are expanding, making the disease an important public health problem that would cause serious damage to human living, travelling or working in forests.Objectives:(1)To clarify the epidemiological characteristics of TBE in mainland China, including temporal, spatial and demographic distribution; to summarize the geographical distribution and habitat features on the main natural focus of TBE in mainland China; to predict the distribution of natural focus of TBE in China through establishing ecological model and to explore the environmental determinants associated with the occurrence of TBE.(2)Choosing the main natural focus of TBE(Heilongjiang Province, Jilin Province and Inner Mongolia Autonomous Region) to detect the spatiotemporal clusters, epidemiological features and habitat characteristics at the township level; to quantitatively evaluate the role of meteorological and land-cover factors on the spatial-temporal distribution of the disease.Methods: We collected and processed eight-year data of human TBE cases from2006 to 2013 in mainland China, meteorological data including monthly average temperature and relative humidity, and monthly accumulative rainfall and sunshine hours during 2006–2013, demographic data from the Sixth National Census in 2010,elevation, the coverage of vegetation types, the survey data of natural focus, the survey data of the vectors and hosts associated with TBEV.(1)We described the epidemiological characteristics, dynamic change and regional differences of human TBE cases from 2006 to 2013 in mainland China; to analyze the geographical distribution of natural focus of TBE combined with human TBE cases and survey data of natural focus, and to summarize the features of natural focus of TBE in the aspects of vectors, host animals, geographic and geomorphic conditions, vegetation types, and climatic conditions; a boosted regression trees(BRT) model was applied at the county level to assess the risk environmental and meteorological factors associated with the occurrence of TBE, based on which a map was created to predict the potential distribution of natural focus of TBE across the mainland China.(2)In the northeastern endemic regions, we used Kulldorff’s spatiotemporal scan statistic to detect the hotspots of TBE at the township level; a comparative analysis of the epidemic characteristics of TBE was performed within and outside identified clusters; Spearman correlation analyses was applied to examine the association and lag effects between meteorological variables and monthly TBE incidence; Negative binomial regression model was used to explore land cover factors associated with the spatial distribution of TBE. Softwares including Microsoft office 2010, Arc GIS 10.2, STATA10.0, R Language 3.1.1 and Sa TScan 10.0 were used in the analysis.Results:(1)A total of 2117 human TBE cases were reported from 2006 to 2013, and the number of cases were increasing, among which 2093 cases(about 99% of the total cases) were distributed in forest areas of northeastern China, including Heilongjiang Province, Jilin Province and the northeastern of Inner Monglia Autonomous Region; a spatiotemporal expansion of the disease towards west and southwest in recent years was observed in the three provinces; Inner Mongolia Autonomous Region and Heilongjiang Province followed a similar pattern with national TBE incidence. However, thereseemed to be a long-term increasing trend in Jilin Province; Seasonality of TBE was obvious, with the majority of reported infections occurring during May to July each year.More males were observed than females among the cases, and mostly in 30–59 years among all age groups by 10-year age group(about 75% of the total cases). Males in40–49 and 50–59 had the highest and quickest increase in the number of TBE cases during 2009–2011. For occupation, farmers, domestic workers and forest workers accounted for the majority of the patients, and the proportions of patients from farmers and domestic workers were increasing in recent years. The epidemiological features of TBE differed slightly among Heilongjiang Province, Jilin Province, Inner Monglia Autonomous Region and other affected provinces. For example, TBE cases were reported in about three quarters of counties in Heilongjiang Province, but in about a third of counties in Inner Mongolia Autonomous Region; the mean and median age of patients were younger in other provinces than in the three provinces of northeastern China; the most patients were farmers in Heilongjiang, Jilin and other provinces, but were forestry workers and domestic workers in Inner Mongolia Autonomous Region.The distribution of main natural focus of TBE in mainland China were in foresty areas in the Northeast and the Northwest, and Gaoligong Mountains in Yunnan Province in the Southwest. The Northeast natural focus of TBE comprise Daxing’an Mountains in Inner Mongolia Autonomous Region, Xiaoxing’an Mountains in Heilongjiang Province,and Changbai Mountains in Jilin Province, and the Northwest natural focus include Tianshan Mountains and Altai Mountains in Xinjiang Autonomous Region. The vegetation type of typical natural focus was with mixed broadleaf-conifer forests or broad-leaved forests as the dominant vegetation. Ixodes persulcatus ticks are the most important vector in Northeastern China and Xinjiang, however, Ixodes ovatus ticks are the dominant vector in Yunnan. The main reservoir hosts of TBEV are rodents. The result of BRT models revealed that the presence of TBE at the county level in our country was significantly associated with meteorological factors, elevation, thecoverage of broad-leaved forest, the coverage of mixed broadleaf-conifer forest, and the distribution of Ixodes persulcatus ticks. Non-linear relationships existed between the risk of TBE and each of the important predictors aboved. The risk prediction map of TBE in mainland China showed that the high risk areas overlapped well with the observed TBE cases, specially, with all the confirmed natural focus distributing in the predictive high risk areas in Xinjiang.(2)In the northeastern regions of China(including Heilongjiang Province, Jilin Province and Inner Mongolia Autonomous Region), 2093 TBE patients were reported during January 2006 to December 2013, and the results of spatiotemporal cluster analysis for 2033 TBE cases that could be geocoded to the township-level showed that one primary cluster(cluster 1) and three secondary clusters(cluster 2–cluster 4) encompassing 1086 townships were detected. Within the four clusters, 22.69% of the total study population accounted for 83.08% of TBE cases in the northeastern regions. Cluster 1 was located in Daxing’an Mountains, containing 145 townships from June 2006 to July 2013 and the Relative Risk(RR) of 23.02. Cluster 2was located in Changbai Mountains, including 332 townships with a RR of 9.90, and emerged between June 2010 and July 2013. Cluster 3 was located in Xiaoxing’an Mountain. It included 180 townships with a RR of 14.60, occurring between May 2007 and July 2013. Cluster 4 was located at the junction of Inner Mongolia, Heilongjiang and Jilin provinces, which included 429 townships and a 4.72 RR value, and spanned from May 2011 to July 2011. Spatial clusters of human TBE for each year during2006–2013 were performed to show that the primary cluster was located in Changbai Mountain in 2010 and 2013, and in all the other years the primary cluster was located in Daxing’an Mountains. We found differences in the epidemic characteristics(including the mean or median age, the occupational distribution) of TBE by comparative analysis performed in cluster areas and non-cluster areas, for example, the mean or median age of patients was younger in cluster 4 than in cluster 1–3 and non-cluster areas; the most patients were farmers in cluster 2 and non-cluster areas, and were forestry workers incluster 1, and were domestic workers in cluster 3, and in cluster 4, those patients from farmers and from domestic workers accounted for the same proportions. Spearman correlation analyses were conducted showing that monthly TBE incidence was positively correlated with monthly average temperature and monthly cumulative rainfall at the lag of 0 months within the four spatial-temporal clusters(R>0.5, P<0.05); In cluster 1, 3 and 4, monthly TBE incidence was positively correlated with monthly cumulative sunshine hours at the lag of 1 months, but was negatively correlated with monthly average relative humidity at the lag of 2 months; In cluster 2, monthly TBE incidence was positively correlated with monthly cumulative sunshine hours at the lag of 0 months, but was negatively correlated with monthly average relative humidity at the lag of 4 months. Using Negative binomial regression model, we found that TBE incidence was significantly positive correlation with coverage of mixed broad-conifer forest, coverage of broad-leaved forest, coverage of coniferous forest, coverage of grassland, and coverage of meadow, and was negative correlation with coverage of cropland, and was not significantly correlation with coverage of shrub.Conclusion: our study provided a comprehensive epidemiological overview on the reported TBE cases, as well as distribution and features on natural focus of TBE in mainland China, and we performed risk assessment for potential distribution areas of natural focus and human TBEV infections across China; In the main natural focus- the northeastern regions of China, we identified the spatiotemporal clusters of TBE cases,and explored the role of climate and land-cover factors on the spatial-temporal distribution of the disease. Our findings help understand epidemiological features of the disease, determine high risk areas, figure risk factors associated with the spatial-temporal distribution of TBE and facilitate prevention and control for the disease.