Eco-social determinants of Schistosoma japonicum infection supported by multi-level modelling in Eryuan county, People's Republic of China

- We assessed eco-social determinants of Schistosoma japonicum infection in Yunnan, China.
- We used a Bayesian multilevel modelling approach.
- We identified and discuss key risk factors for schistosomiasis.
- A multi-level approach can deepen the understanding of eco-social determinants at a small scale.

Schistosomiasis remains of considerable public health concern in many tropical and subtropical regions of the world, including the People's Republic of China (P.R. China). The effectiveness of schistosomiasis control interventions are, among other factors, governed by the social-ecological context. However, eco-social determinants of schistosomiasis are poorly understood, particularly at the household or village levels. In the current study, residents in 26 villages of Eryuan county, Yunnan province, P.R. China, were screened for Schistosoma japonicum infection with a serological assay that was followed by stool examination for sero-positive individuals. Bayesian multilevel models with spatial random effects were employed to profile the S. japonicum infection risk based on known transmission sites of S. japonicum that are scattered across individual land parcels in this part of the country. The key risk factors identified with this approach were the absence of a sanitary stall house for livestock and presence of living and infected intermediate host snails in close proximity. We conclude that a spatially explicit Bayesian multilevel approach can deepen our understanding of eco-social determinants that govern schistosomiasis transmission at a small geographical scale.

Schistosomiasis remains of considerable public health concern in China. The effectiveness of control interventions is, among other factors, governed by social-ecological contexts. We selected 26 villages of Eryuan county, Yunnan province in south-western China and assessed the eco-social determinants of Schistosoma japonicum infection in humans, using a Bayesian multilevel modelling approach.