JMIR Publications

ABSTRACT

Reduction in typhoid burden in low-income countries requires implementation of feasible water, sanitation, and hygiene (WASH), particularly in densely populated urban slums. Using a large prospective cohort in a densely populated contemporary slum in Mirpur, Bangladesh, we assessed the association between household WASH status and typhoid risk in a training subpopulation (n=98,087). We applied a machine learning algorithm to create dichotomous (“Better”, “Not Better”) composite WASH variable based on three variables – private toilet facility, safe drinking water source and presence of water filter – followed by evaluation of sensitivity and specificity in a distinct validation subpopulation (n=65,286). Using Cox regression models, we evaluated protection associated with living in “Better” WASH households and in clusters with increasing levels of “Better” WASH prevalence during 2 years of follow-up. Residence in “Better” WASH households was associated with 38% reduction in typhoid risk (adjusted hazard ratio [aHR]=0.62; 95% confidence interval [CI]: 0.49–0.78; p<0.001). This reduction was mainly observed in persons <10 years old at first census participation: aHR=0.49; (95%CI: 0.36–0.66; p<0.001). The prevalence of “Better” WASH in clusters appeared inversely related to typhoid incidence; however, this association was not significant in the multivariable model. The adjusted hazard of typhoid decreased by 0.996 (95%CI: 0.986–1.006) for each percent increase in prevalence of “Better” WASH in the cluster (p=0.396). In conclusion, existing variations in household WASH were associated with typhoid risk in a densely populated urban slum. The findings suggest that achievable improvements in WASH facilities may enhance typhoid control in slums where major infrastructural improvements are challenging.