Spatial risk profiling of Schistosoma japonicum in Eryuan county, Yunnan province, China

Abstract

Abstract. Bayesian spatial risk profiling holds promise to enhance our understanding of the epidemiology of parasitic diseases, and to target interventions in a cost-effective manner. Here, we present findings from a study using Bayesian variogram models to map and predict the seroprevalence of Schistosoma japonicum in Eryuan county, Yunnan province, China, including risk factor analysis. Questionnaire and serological data were obtained through a cross-sectional survey carried out in 35 randomly selected villages with 3,220 people enrolled. Remotely-sensed environmental data were derived from publicly available databases. Bivariate and non-spatial Bayesian multiple logistic regression models were used to identify associations between the local seroprevalence and demographic (i.e. age and sex), environmental (i.e. location of village, altitude, slope, land surface temperature and normalized difference vegetation index) and socio-economic factors. In the spatially-explicit Bayesian model, S. japonicum seroprevalence was significantly associated with sex, age and the location of the village. Males, those aged below 10 years and inhabitants of villages situated on steep slopes (inclination ≥20°) or on less precipitous slopes of >5°above 2,150 m were at lower risk of seroconversion than their respective counterparts. Our final prediction model revealed an elevated risk for seroconversion in the plains of the eastern parts of Eryuan county. In conclusion, the prediction map can be utilized for spatial targeting of schistosomiasis control interventions in Eryuan county. Moreover, S. japonicum seroprevalence studies might offer a convenient means to assess the infection pressure experienced by local communities, and to improve risk profiling in areas where the prevalence and infection intensities have come down following repeated rounds of praziquantel administration.

Human infections and co-infections with helminths in a rural population in Guichi, Anhui Province, China

“…The higher prevalence among older individuals also supports this hypothesis. Nevertheless, socioeconomic status (quantified by income) was not a significant factor for S. japonicum infection in our study, which is not consistent with the results from previous studies (Agbo et al, 1999;Steinmann et al, 2007;Xu et al, 2011). We observed that family members with higher household income were mainly fisherman, which might at least in part, explain the inconsistency.…”

“…First, when modelling S. japonicum risks, we used the sero-positive cases of schistosomiasis due to the very small number of stool-positive cases. Although sero-positivity does not necessarily indicate the current status of infection, it provides an estimate of the infection pressure irrespective of recent control activities (Steinmann et al, 2007) and the burden of schistosomiasis in the study area. Screening methods have been frequently used in areas of low infection intensity after repeated rounds of chemotherapy (Hillyer and Soler, 1999;Hoshino-Shimizu et al, 1992;Steinmann et al, 2007).…”

Long-Term Impact of the World Bank Loan Project for Schistosomiasis Control: A Comparison of the Spatial Distribution of Schistosomiasis Risk in China

“…When studying the epidemiology of schistosomiasis and evaluating the impact of control strategies, it is impossible to consider all the potential risk factors related with interested diseases because either the information is unavailable or disease mechanisms are unclear. Hence, previous reports have only included the most important factors or those of specific interest [4], [5], [26], [27], [28], [29], [30], [31], which will contribute to bias in effect estimates because of unadjusted effects from risk factors that have been ignored. From the spatial perspective, all the potential risk factors related with studied diseases can be divided into two latent components of random effects, spatially correlated heterogeneity (CH) and uncorrelated heterogeneity (UH).…”

Spatial analysis and risk mapping of soil-transmitted helminth infections in Brazil, using Bayesian geostatistical models

“…Successful applications include risk profiling of schistosomiasis, soil-transmitted helminthiasis and co-infections at the village, district, national and regional level in Africa and Asia (Clements et al, 2006;Raso et al, 2006;Steinmann et al, 2007;Schur et al, 2011a;Chammartin et al, 2013a). In Latin America, the paucity of data has considerably delayed the mapping, and hence control, of soil-transmitted helminthiasis (Sabin Vaccine Institute/Inter-American Development Bank/PAHO, 2011).…”