The joint effects of efficacy and compliance: a study of household water treatment effectiveness against childhood diarrhea.


The effectiveness of household water treatment (HWT) at reducing diarrheal disease is related to the efficacy of the HWT method at removing pathogens, how people comply with HWT, and the relative contributions of other pathogen exposure routes. We define compliance with HWT as the proportion of drinking water treated by a community. Although many HWT methods are efficacious at removing or inactivating pathogens, their effectiveness within actual communities is decreased by imperfect compliance. However, the quantitative relationship between compliance and effectiveness is poorly understood. To assess the effectiveness of HWT on childhood diarrhea incidence via drinking water for three pathogen types (bacterial, viral, and protozoan), we developed a quantitative microbial risk assessment (QMRA) model. We examined the relationship between log(10) removal values (LRVs) and compliance with HWT for scenarios varying by: baseline incidence of diarrhea; etiologic fraction of diarrhea by pathogen type; pattern of compliance within a community; and size of contamination spikes in source water. Benefits from increasing LRVs strongly depend on compliance. For perfect compliance, diarrheal incidence decreases as LRVs increase. However, if compliance is incomplete, there are diminishing returns from increasing LRVs in most of the scenarios we considered. Higher LRVs are more beneficial if: contamination spikes are large; contamination levels are generally high; or some people comply perfectly. The effectiveness of HWT interventions at the community level may be limited by imperfect compliance, such that the benefits of high LRVs are not realized. Compliance with HWT should be carefully measured during HWT field studies and HWT dissemination programs. Studies of pathogen concentrations in a variety of developing-country source waters are also needed. Guidelines are needed for measuring and promoting compliance with HWT.

MIDAS Network Members