Syed Anjerul Islam , Argaw Ambelu , Zakariah Seidu , Ryan D. Cronk , Michael B. Fisher 

Microbial contamination of drinking water contributes to a substantive and preventable burden of enteric disease that disproportionately impacts infants and children. The World Health Organization (WHO) has published guidance on water safety and water quality monitoring, including sanitary inspection (SI) of water systems to detect and manage hazards such as fecal contamination. Sanitary inspection is a low-cost, on-site risk assessment tool for water supply systems based on observable risk factors (RFs) associated with potential hazards or defects. While water quality sampling and analysis methods are well characterized, SI as a tool for risk management in drinking water systems is under-studied. We used SI and water quality data from 966 rural boreholes in Ethiopia, Ghana, and Burkina Faso and merged these with remote-sensing rainfall estimates based on household location. Logistic regressions (binary and ordinal) were used to characterize associations of total SI risk score, as well as individual risk factors (RFs), and classes of RFs (i.e., “Source,” “Transport,” and “Barrier” risks) with fecal indicator bacteria (FIB) occurrence as the outcome, controlling for estimated cumulative rainfall (over the past 1-15 days before sampling). We found associations (P<0.05, OR: 3.5, 95% CI 1.05-11.66) between SI scores and E. coli risk categories controlling for fifteen-day cumulative rainfall. Furthermore, risk factors in the “barrier” category, such as the presence and adequacy of fencing around boreholes, concrete pads, and walls extending below the ground, were associated with the E. coli risk category. When examining individual RFs in the regression models, the presence of human excreta 10 m from the source (OR: 2.53), absence of cement floor (OR: 0.16), handpumps that were loosely attached at the base (OR:1.57), adequate fencing to keep animals out (OR: 0.57) and the presence of stagnant water (OR:1.39) were significantly associated with microbial contamination. Incorporating precipitation into models improved model fit characteristics (improved Pseudo R squared and AIC value); specifically, accounting for cumulative rainfall during the fifteen days before sampling improved model fit (increased pseudo-R2 from 0.035 to 0.05) for E. coli contamination. These findings can inform the design, construction, maintenance, and monitoring of boreholes and prompt timely remediation of critical hazards in such systems, potentially enhancing water safety.