Predictors of bovine Schistosoma japonicum infection in rural Sichuan, China
Description
In China, bovines are believed to be the most common animal source of human Schistosoma japonicum infections, though little is known about what factors promote bovine infections. Because schistosomiasis is a sanitation-related, water-borne disease transmitted by many animals, we hypothesized that several environmental factors – such as the lack of improved sanitation systems, or participation in agricultural production that is water-intensive – could promote schistosomiasis infection in bovines. This data was collected as a part of a repeat cross-sectional study conducted in rural villages in Sichuan, China from 2007 to 2016. During this time, all humans and bovines residing in participating households in the study villages were invited to participate in schistosomiasis infection surveys in 2007, 2010 and 2016. Additionally, the head of each household was asked to complete a household survey each year that contained closed-ended questions related to socioeconomic status, domestic and farm animal ownership, sanitation and water access and agricultural practices. Bovine age, type and sex were also collected at the time of the bovine infection surveys in 2007 and 2010. Each row in the data represents a single bovine participant from a given study year. The outcome in this analysis was bovine Schistosoma japonicum infection, while predictors included village and household-level values of physical and environmental conditions hypothesized to be potentially predictive of bovine S. japonicum infection. Candidate predictors included: 1) physical/biological characteristics of bovines, 2) human sources of environmental schistosomes, 3) socio-economic indicators, 4) animal reservoirs, and 5) agricultural practices. Village-level predictors were generated from the household survey data, representing all households that participated in the household survey from a given village, even if they didn’t own bovines. Notably, the village-level variables excluded all observations from the bovine’s own household, and instead used only the data from the other households in the village. This allowed for an assessment of how the surrounding village environment impacts individual bovine infection risk, independent of the home environment, whereas the household-level variables aim to unpack the influence of the unique household environment on bovine infection status. All personal and/or identifying information – including human infection data – was removed from our datasets prior to publishing here to maintain participant privacy. Nevertheless, our entire analysis code is included in the R scripts published here, in the interest of full transparency. The density of bovines in a village and agricultural practices were among the top predictors of bovine S. japonicum infection in all collection years. Additionally, human infection prevalence in the village, pig ownership and bovine age were found to be strong predictors of bovine infection in at least one year.
Files
Steps to reproduce
This study was conducted in two rural counties in Sichuan, China where schistosomiasis has reemerged. Surveys of environmental and social risk factors, as well as human and bovine infection were conducted in 2007, 2010 and 2016. A village census was conducted in each collection year and all bovines and residents over the age of five were invited to participate in surveys for S. japonicum infection. In the summers of 2007, 2010 and 2016, the head of each household was asked to complete a household survey that contained closed-ended questions related to socioeconomic status, domestic and farm animal ownership, sanitation and water access and agricultural practices. Bovine age, type and sex were collected at the time of the bovine infection surveys in 2007 and 2010 (these data were not collected in 2016). S. japonicum infection surveys were conducted by attempting to test three stool samples on three consecutive days from eligible humans and all bovines, using both Kato-Katz and miracidium hatching tests for human participants, and only miracidium hatching tests for testing the bovine population. Infection surveys were conducted in November and December of 2007 and 2010, and July 2016. Bovines were isolated in a pen or tied up until a stool was produced on three separate days (consecutive, when possible). All stool samples were transported to the central laboratory soon after collection to be examined using the miracidium hatching test, following standard protocols. To account for the short survival and rapid hatching of bovine miracidia, the bovine samples were examined for miracidia at one, three and five hours after preparation for at least two minutes each time, whereas human samples were assessed at two, five and eight hours after preparation. One sample from each human was also examined using the Kato-Katz thick smear procedure in 2007 and 2010, using three slides per stool, and 41.7 mg Kato-Katz template. A bovine was considered positive for S. japonicum if any miracidium hatching test was positive. A human was considered positive for S. japonicum if any miracidium hatching test or the Kato-Katz test was positive. Household questionnaire and infection survey data were all stored in RedCap prior to analysis. The raw data was exported to Stata for cleaning and variable generation. Prior to publishing our datasets, all personal and/or identifying information was removed (e.g. names and unique identifiers, geographic coordinates, participant birth date, etc.). The data was then exported to csv for this analysis, conducted in RStudio 4.0. Because human infection data was considered personal information in this study, these variables were removed from the CSV files prior to publishing on this site. As such, annotations were added to the R-Script files that are shared here, in order to indicate which sections of the code can be run to fully reproduce our results using the data shared here.