Unprotected water sources and low latrine coverage are contributing factors to persistent hotspots for schistosomiasis in western Kenya

Background Evidence indicates that whereas repeated rounds of mass drug administration (MDA) programs have reduced schistosomiasis prevalence to appreciable levels in some communities referred to here as responding villages (R). However, prevalence has remained high or less than anticipated in other areas referred to here as persistent hotspot villages (PHS). Using a cross-sectional quantitative approach, this study investigated the factors associated with sustained high Schistosoma mansoni prevalence in some villages despite repeated high annual treatment coverage in western Kenya. Method Water contact sites selected based on observation of points where people consistently go to collect water, wash clothes, bathe, swim or play (young children), wash cars and harvest sand were mapped using hand-held smart phones on the Commcare platform. Quantitative cross-sectional surveys on behavioral characteristics were conducted using interviewer-based semi-structured questionnaires administered to assess water usage/contact patterns and open defecation. Questionnaires were administered to 15 households per village, 50 pupils per school and 1 head teacher per school. One stool and urine sample was collected from 50 school children aged 9-12 year old and 50 adults from both responding (R) and persistent hotspot (PHS) villages. Stool was analyzed by the Kato-Katz method for eggs of S. mansoni and soil-transmitted helminths. Urine samples were tested using the point-of-care circulating cathodic antigen (POC-CCA) test for detection of S. mansoni antigen. Results There was higher latrine coverage in R (n = 6) relative to PHS villages (n = 6) with only 33% of schools in the PHS villages meeting the WHO threshold for boy: latrine coverage ratio versus 83.3% in R, while no villages met the girl: latrine ratio requirement. A higher proportion of individuals accessed unprotected water sources for both bathing and drinking (68.5% for children and 89% for adults) in PHS relative to R villages. In addition, frequency of accessing water sources was higher in PHS villages, with swimming being the most frequent activity. As expected based upon selection criteria, both prevalence and intensity of S. mansoni were higher in the PHS relative to R villages (prevalence: 43.7% vs 20.2%; P < 0.001; intensity: 73.8 200.6 vs 22.2 96.0, P < 0.0001), respectively. Conclusion Unprotected water sources and low latrine coverage are contributing factors to PHS for schistosomiasis in we