Spatio-Temporal Heterogeneity of Schistosomiasis in China Based on Multi-stage, Continuous Downscaling of Sentinel Monitoring


 Background: There is a continuous decline in the prevalence of schistosomiasis and the number of Schistosoma japonicum infections in humans and livestock in China. However, there are a large number of factors that have not been resolved and which may contribute to future transmission of schistosomiasis. These include a range of sources for S. japonicum infection, difficulty in management of S. japonicum sources of infection, frequent emergence and re-emergence of Oncomelania hupensis snail habitats, and the problematic elimination of snail habitats. These factors challenge progress towards the elimination of schistosomiasis in China.Methods: Based on multi-stage continuous downscaling of sentinel monitoring, county-based schistosomiasis surveillance data were captured from the national schistosomiasis surveillance sites of China from 2005 to 2019. The data included S. japonicum infections in humans, livestock, and O. hupensis. The spatio-temporal trends for schistosomiasis were detected using a Joinpoint regression model, with a standard deviational ellipse (SDE) tool, which determined the central tendency and dispersion in spatial distribution of schistosomiasis. Further, spatio-temporal clusters of S. japonicum infections in humans, livestock, and O. hupensis were evaluated by Poisson model. Results: The prevalence of S. japonicum human infections was reduced from 2.06% to zero based on the national schistosomiasis surveillance sites of China during the period from 2005 to 2019, with a reduction from 9.42% to zero for the prevalence of S. japonicum infections in livestock, and from 0.26% to zero for the prevalence of S. japonicum infections in O. hupensis. The decline in prevalence of S. japonicum infections in humans, livestock, and O. hupensis was statistically significant from 2005 to 2019 (P < 0.01). There was an exception to the decline in S. japonicum infections in livestock during the period from 2008 to 2012. Using an SDE tool, schistosomiasis-affected regions were reduced yearly from 2005 to 2014 in the endemic provinces of Hunan, Hubei, Jiangxi, and Anhui, as well as in the Poyang and Dongting Lake regions. Poisson model revealed 11 clusters of S. japonicum human infections, six clusters of S. japonicum infections in livestock, and nine clusters of S. japonicum infections in O. hupensis. The clusters of human infection were found to be highly consistent with clusters of S. japonicum infections in livestock and O. hupensis. These clusters were in the five provinces of Hunan, Hubei, Jiangxi, Anhui, and Jiangsu, as well as along the middle and lower reaches of the Yangtze River. Humans, livestock, and O. hupensis infections with S. japonicum were mainly concentrated in the north of the Hunan Province, south of the Hubei Province, north of the Jiangxi Province, and southwestern portion of Anhui Province. In the two mountainous provinces of Sichuan and Yunnan; human, livestock, and O. hupensis infections with S. japonicum were mainly concentrated in the northwestern portion of the Yunnan Province, the Daliangshan area in the south of Sichuan Province, and the hilly regions in the middle of Sichuan Province. Conclusions: This study demonstrate a significant spatio-temporal heterogeneity of schistosomiasis in China. A remarkable decline in endemic schistosomiasis was observed between 2005 and 2019. However, there continues to be a long-term risk of schistosomiasis transmission in local areas, with high-risk areas primarily located in the Poyang Lake and Dongting Lake regions, with frequent acute S. japonicum infections. Using a One Health approach, further reinforcement of an integrated schistosomiasis control strategy, with an emphasis on the sources of S. japonicum infection, is required to facilitate the elimination of schistosomiasis in China by 2030.


Background
Schistosomiasis is prevalent in 78 countries or territories across Asia, Africa, and South America. Currently, more than 200 million people are estimated to have the disease and more than 800 million are at risk for infection. The global disability-adjusted life years (DALYs) due to schistosomiasis is 70 million [1]. This infectious disease was once hyper-endemic in southern China, with 11.6 million schistosomiasis patients and 1.2 million bovine infections with Schistosoma japonicum. There were 14.2 billion m 2 snail habitats at the founding of the People's Republic of China [2]. The national schistosomiasis control strategy in China has shifted three times. The rst was a snail controlbased integrated strategy (from the founding of the People's Republic of China to the early 1980s), followed by a morbidity control strategy based on synchronous chemotherapy for humans and bovines (from the middle 1980s to 2003). The third was an integrated strategy based on control of the source of S. japonicum infections (from 2004 to present) [3]. With these concerted efforts, endemic schistosomiasis has continuously declined in both prevalence and numbers of S. japonicum infections in humans and livestock. Of the 450 Chinese counties endemic for schistosomiasis, 337 counties (74.89%) eliminated schistosomiasis, 97 (21.56%) achieved transmission interruption, and 16 (3.55%) achieved transmission control by the end of 2020 [4]. However, there are still a large number of factors associated with the transmission of schistosomiasis in China including; a wide range of S. japonicum sources of infection, a high degree of di culty in the management of S. japonicum sources of infection, frequent emergence and re-emergence of snail habitats, and the problematic nature of snail habitat elimination [5]. Further, currently available tools do not meet the requirements of the national schistosomiasis elimination program, which impedes the elimination of schistosomiasis from China [6].
understanding of the prevalence of S. japonicum infections in humans and livestock, as well as the endemic foci of Oncomelania hupensis in China [8,9]. The focus of currently available surveillance data is on the description of spatial, temporal, and population distribution of schistosomiasis. However, knowledge of the spatio-temporal heterogeneity of schistosomiasis in China is lacking. Spatial epidemiology effectively quanti es spatial distribution and provides disease mapping, cluster analysis, and risk factor identi cation. Spatial epidemiology provides insights into disease control, prevention, and health resource allocation [10]. Recently, Li and colleagues [11] used both global Moran's I and Anselin's local Moran's I statistics (LISA) to construct a retrospective space-time permutation model for identi cation of the spatial and temporal distributions of emerging snail-infested sites in the Hunan Province from 1949 to 2016. The model was based on annual snail survey data. Pinheiro and colleagues [12] employed spatial epidemiological summaries to determine the spatial-temporal distribution of schistosomiasis-related mortality in Brazil from 2003 to 2018. Spatial analysis has considerable potential for the assessment of schistosomiasis transmission risk and for the development of a schistosomiasis control strategies. Based on multi-stage downscaling and continuous sentinel monitoring data, the aim of this study was to determine the spatial-temporal distribution of S. japonicum infections in humans, livestock, and O. hupensis across the endemic foci of China. In this manner those areas in need of schistosomiasis control will be identi ed. Further, results will provide insight into schistosomiasis management and surveillance during the elimination stage of the program.

Data collection
Schistosomiasis surveillance data were provided by the National Institute of Parasitic Diseases, Chinese Center for Disease Control and Prevention. According to the epidemiological pro le of schistosomiasis, a strati ed method was used to assign national schistosomiasis surveillance sites within endemic province of China [13]. Permanent residents aged 6 to 65 years underwent serological screening at national schistosomiasis surveillance sites for S. japonicum infection, as well as measurements of antibody titers for S. japonicum. Seropositive individuals were tested for parasites by the Kato-Katz technique and by the miracidium hatching test [8,9]. The miracidium hatching test was used to detect S. japonicum infections in livestock [8,9]. Snail surveys were conducted in snail habitats and in suspected snail-infested habits by systematic sampling and environmental sampling at national schistosomiasis surveillance sites. All captured snails were assessed for viability and S. japonicum infection [8,9]. Field schistosomiasis surveillance was conducted by local schistosomiasis control institutions. Annual schistosomiasis surveillance data were reported to the National Institute of Parasitic Diseases, Chinese Center for Disease Control and Prevention.

Data processing
County-based schistosomiasis surveillance data were captured from national schistosomiasis surveillance sites in China from 2005 to 2014 that included the number of; those tested for S. japonicum infection using sero-diagnostic assays, seropositive individuals, individuals tested for S. japonicum infections using parasitology testing, egg-positive individuals, livestock tested for S. japonicum infection, egg-positive livestock, living O. hupensis snails, and O. hupensis snails with S. japonicum infection. The sero-prevalence and prevalence of S. japonicum infection in humans, livestock, and O. hupensis were calculated using the following formulas: Sero-prevalence of S. japonicum human infections = number of sero-positives/number of individuals with serological screening for S. japonicum infection × 100%. Annual percent change (APC) from 2005 to 2014 was estimated. A Joinpoint model, based on an algorithm, was used to assess signi cant change in schistosomiasis trends.

Central tendency and dispersion analysis
A standard deviational ellipse (SDE) tool, also termed a directional distribution tool, was used to identify the spatial distribution of disease [15]. The long and short axes of an ellipse indicate the direction of major and minor trends in spatial distribution of a disease. The size of the long and short axes identi es the degree of deviation from center for the spatial distribution of schistosomiasis in major and minor trend directions [15]. In this study, an SDE tool was used to determine the central tendency and dispersion of spatial distributions for schistosomiasis.

Temporal-spatial cluster analysis
Temporal-spatial clusters of S. japonicum infections were identi ed in humans, livestock, and O. hupensis from 2015 to 2019 using a Poisson model and SaTScan version 9.4.2 [16]. The space-time permutation scan statistic allows a dynamic scan using a cylindrical window in dimensions of time scales and geographical location. A log likelihood ratio (LLR), a statistic that tests the difference between the observed number and the expected number in and outside the window, was estimated and relative risk (RR) calculated. The Monte Carlo method was used for the permutation test, which identi ed high-prevalence clusters [17]. In China, the national schistosomiasis control program was implemented in counties, with the endemic status of schistosomiasis assigned to an administrative village as a national schistosomiasis surveillance site. In this manner the overall prevalence of schistosomiasis in a county was determined. In this study, we used the schistosomiasis clusters in an administrative village to de ne the cluster of schistosomiasis in a county.

Trends in prevalence of schistosomiasis
The prevalence of S. japonicum infection trended toward continuous decline in humans, bovines, and O. hupensis based on the national schistosomiasis surveillance sites of China from 2005 to 2019 ( Figure 1). There was a reduction in human infections from 2.06% in 2005 to zero in 2019. For livestock the reduction was from 9.42% to zero and from 0.26% to zero for O. hupensis (Table 1)  Temporal-spatial clusters of schistosomiasis transmission risk At the county level, space-time scan analysis identi ed six clusters of S. japonicum human infections across ve provinces in lake areas (RR: 1.77 to 12.6) and ve clusters in two mountainous provinces of Sichuan and Yunnan (RR: 6.5 to 9.19) based on the national schistosomiasis surveillance sites from 2005 to 2019. The six clusters covered 32 counties in ve province lake areas. There were 13 counties in Hunan, 13 in Hubei, four in Jiangxi, and three in Anhui. The clusters were primarily in: Yueyang City (5), Yiyang City (3), and Changde City (4) of the Hunan Province; Jingzhou City (7) of the Hubei Province; and Shangrao City (3) of the Jiangxi Province accounting for 68.75% (22/32) of all clusters ( Figure 5, Table 2). In addition, the ve clusters in the two mountainous provinces covered two counties in the Yunnan Province and seven in the Sichuan Province. The clusters were primarily in the Liangshan Prefecture (2) and Meishan City (3) of the Sichuan Province, accounting for 55.56% (5/9) of all clusters ( Figure 5, Table 2).
At the county level, space-time scan analysis identi ed four clusters of S. japonicum livestock infections across ve provinces in lake areas (RR: 3.01 to 9.04) and two clusters in two provinces of Sichuan and Yunnan (RR: 3.4 to 7.32) based on the national schistosomiasis surveillance sites from 2005 to 2019. The four clusters covered 32 counties across ve provinces in lake areas, including nine counties in Hunan, nine in Hubei, nine in Jiangxi, and seven in Anhui. The clusters were primarily in: Jingzhou City (7) of the Hubei Province; Changde City (4) and Yueyang City (4) of the Hunan Province; Jiujiang City (4) and Nanchang City (3) of the Jiangxi Province; and Anqing City (4) and Chizhou City (2) of the Anhui Province accounting for 82.35% (28/34) of all clusters ( Figure 6, Table 3). In addition, two clusters in the mountainous regions covered two counties in the Yunnan Province, including Weishan County and Eryuan County ( Figure 6, Table 3 Table 4). In addition, two clusters in the two mountainous provinces covered two counties in Yunnan Province and one in Sichuan Province including Eryuan County and Dali County of the Yunnan Province, and Dechang City of the Sichuan Province (Figure 7, Table 4).
These results identify clusters of S. japonicum infections in humans, livestock, and O. hupensis to be concentrated in the north of Hunan Province, south of Hubei Province, north of Jiangxi Province, and southwest of Anhui Province. Infections were across ve provinces in lake areas, with clusters predominantly located around Poyang Lake, Dongting Lake, and along the middle and lower reaches of the Yangtze River.  [19]. During the period from 2005 through 2019, an integrated strategy, with an emphasis on the management of the source of S. japonicum infections, was implemented for schistosomiasis control in China. The strategy included; expanded examination and therapy for schistosomiasis, raising livestock in pens, replacement of bovine with machines, improved sanitation, night soil management, and cementing of ditches [20]. Based on the principle of "prevention rst, scienti c control, highlighting key points and classi ed guidance", this integrated strategy reinforced schistosomiasis examination and therapy for humans and livestock, as well as snail survey and control. By 2019, there were two individuals and seven bovines with S. japonicum infections and 3.624 billion m 2 of snail habitat in China [21].
Understanding factors that affect schistosomiasis transmission and awareness of endemic status alterations are prerequisites for preventive disease control. During the period from 2005 through 2014, a total of 80 national schistosomiasis surveillance sites were established in China. These included; fork beach, islet without embankment, islet with embankment, inner embankment, plateau, mountain, hill, and waterway networks. Data from the national schistosomiasis surveillance sites identi ed the main types of schistosomiasis-endemic foci as well as trends in the prevalence of S. japonicum infections in China [8,13]. To further understand the potential transmission risk for schistosomiasis, a total of 454 national schistosomiasis surveillance sites were established during the period from 2015 through 2019. The sites covered all schistosomiasis-endemic counties as well as four counties in the Three Gorges Reservoir area. Surveillance included case monitoring, transmission factors, and transmission risk [9]. From 2015 to 2019, the prevalence of S. japonicum infections reduced from 2.0682% to zero in humans, from 9.42% to zero in livestock, and from 0.26% to zero in O. hupensis based on the national schistosomiasis surveillance sites. Joinpoint regression analysis showed; the APC of S. japonicum infections in livestock to be 41.4% from 2005-2008 and 67.2% from 2012-2019; an APC of 10.65% for S. japonicum human infections from 2005-2013 and 20.83% from 2013-2019; and an APC of 14.07% for S. japonicum infections in O. hupensis from 2005-2019. These data demonstrate the remarkable effectiveness of the integrated strategy for schistosomiasis control, which included improved sanitation, protection of human health, and reductions in poverty due to endemic schistosomiasis foci. However, a number of factor lead to occult infections such as contact with low-intensity S. japonicum cercariae in low-endemic areas and the di culty of low-intensity infection diagnosis, results in an underestimation of the prevalence of S. japonicum infection [22]. Currently, schistosomiasis is weakly endemic in China, with the prevalence of S. japonicum infections in humans, livestock, and O. hupensis approaching zero in all national schistosomiasis surveillance sites. However, it is likely that this is an underestimate of schistosomiasis because of the insensitivity of diagnostic assays.
In this study we found a yearly reduction in the schistosomiasis-affected regions of China from 2005 to 2014. By weighted SDE, the affected areas were predominantly located in endemic foci of four provinces; Hunan, Hubei, Jiangxi, and Anhui. These foci were concentrated in Poyang and Dongting Lake regions. These results are similar to the national report on schistosomiasis in China [23,24] and to high-risk areas for schistosomiasis based on modeling [25,26]. There are widespread marshlands around the Dongting and Poyang Lake areas, with many O. hupensis infested sites and a large number of livestock that can serve as reservoirs for S. japonicum [27]. There is an increased possibility of exposure to snail-infested sites and S. japonicum-infested waters for local residents who sh, pasture, and farm. Further, the residents are typically in boats and release their feces directly into lakes, resulting in long-term and extensive infections [28]. In marshlands and lake areas, local residents breed and pasture their animals, which makes elimination of schistosomiasis di cult [29]. Currently, livestock and shermen frequent snail-infested areas, with risk of S. japonicum infection a persistent threat [30].
Because of the life cycle and transmission pattern of the parasite, no changes in high-risk environments are possible and as a consequence there is a high risk for schistosomiasis transmission in local Chinese regions [31]. Poisson analysis identi ed eleven clusters of S. japonicum human infections, six clusters of livestock infections, and nine clusters of O. hupensis infections. The clusters of human infection were highly consistent with those of livestock and O. hupensis infection. The clusters were mainly located; around the Poyang and Dongting Lake areas, Jianghan Plain areas, the middle and lower reaches of the Yangtze River, the northwestern part of Yunnan Province, the Dalianshan Mountain area in the south of Sichuan Province and the hilly regions in the middle of Sichuan Province. In addition, clusters of schistosomiasis transmission risk identi ed by hotspot analysis were essentially consistent with schistosomiasis transmission-controlled counties and the neighboring transmission interrupted-areas [21]. Our ndings indicate a high risk of schistosomiasis transmission in these clustering areas, with the possibility of a rebound in schistosomiasis requiring close attention. The clusters of S. japonicum infection in humans, livestock, and O. hupensis are widespread and concentrated in marshland and lake areas, associated with widespread snail habitats, with large numbers of oating boatmen and shermen, as well as livestock management di culties [32,33]. While the clusters of S. japonicum infection were relatively small and dispersive in the two mountainous provinces, they were associated with the block-or dot-like distribution of snails [34,35]. These ndings provide an extensive coverage of high-risk areas for schistosomiasis in marshland and lake areas, where rebound of schistosomiasis requires careful attention during the elimination stage. Although there is limited schistosomiasis transmission risk in mountainous areas, socio-economic under-development and complex natural environments will likely lead to a rebound in schistosomiasis. Surveillance for S. japonicum infection and O. hupensis needs to be intensi ed in marshlands, lake areas, and mountainous regions. Further, reinforcement of an integrated schistosomiasis control strategy that emphasizes consideration of; the sources of S. japonicum infection, human S. japonicum infections, livestock infections, wild animal infections, and O. hupensis infections are recommended for elimination of schistosomiasis in China [36]. A surveillance-response system using a One Health approach is appropriate. This study has limitations. First, the spatio-temporal analysis of S. japonicum infections in humans, livestock, and O. hupensis was based at the county level. Future studies at a ner scale (at a village or individual scale) are needed. Second, natural and socio-economic factors were not considered with regard to the spatio-temporal heterogeneity of S. japonicum infections in humans, livestock, or O. hupensis. Further analysis of climate, geography, and social developments at a ner scale would provide important insights into precision control of schistosomiasis.

Conclusions
In summary, the results of this study demonstrate signi cant spatio-temporal heterogeneity for schistosomiasis in China. Based on the multistage downscaling of continuous sentinel monitoring data, a remarkable decline was seen in endemic schistosomiasis in China during the period from 2005 through 2019, with markedly reduced disease. However, there remains a long-term risk of transmission in local areas, with the highest-risk areas primarily in Poyang Lake and Dongting Lake regions, where frequent acute S. japonicum infections occur. Using a One Health approach, reinforcement of an integrated schistosomiasis control strategy with emphasis on the sources of S. japonicum infection, and on the inclusion of human, livestock, wild animal, and O. hupensis S. japonicum infections will provide an effective surveillance-response system that will insure elimination of schistosomiasis in China by 2030.