This matched observational study compared health outcomes in households served by continuous water supply (CWS) and those served by intermittent water supply (IWS) in Hubli-Dharwad, India. The investigators enrolled 2,220 CWS households and 2,218 IWS households, matching households on socioeconomic status and sanitation. The analysis focused on two primary outcomes: prevalence of diarrhea in children under 5 years and incidence of typhoid fever. The authors used generalized linear models with robust standard errors, adjusting for sociodemographic characteristics and sanitation, to estimate prevalence ratios (PRs) and cumulative incidence ratios (CIRs).
Among households on IWS, the study reports a median supply frequency of once every 8 days (interquartile range [IQR] = 7–8 days). The median duration of supply per cycle in IWS households was 4 hours (IQR = 3–5 hours). The authors stratified IWS households by supply frequency and duration, including quintile-based comparisons and a comparison to the UN availability criterion (having water available when needed defined as ≥12 hours/day or ≥4 days/week).
Overall, IWS households had higher prevalence of diarrhea among children under 5 years compared with CWS households. The reported prevalence ratio for child diarrhea in IWS versus CWS was 1.36 (95% confidence interval 1.00–1.84), indicating a 36% higher prevalence in IWS households after adjustment. When IWS households met the UN availability criterion or fell into the top quintiles of supply frequency and duration (receiving water once every 1–6 days or for 7–24 hours per supply cycle), child diarrhea prevalence was similar to that observed in CWS households. In contrast, IWS households below the UN criterion and those in the bottom quintiles of supply frequency and duration (receiving water once every 9–15 days or for ≤3 hours per supply cycle) showed substantially higher child diarrhea prevalence—reported increases ranged from 45% to 72% compared with CWS households (statistical significance reported as p<0.05 for these comparisons).
Typhoid fever incidence was also higher in IWS households compared with CWS households. The cumulative incidence ratio comparing IWS to CWS households was 1.78 (95% CI 1.05–3.02), corresponding to a 78% higher number of typhoid cases in IWS households overall. IWS households that met the UN availability criterion had typhoid incidence similar to CWS households. Households below the UN criterion experienced more typhoid fever cases than CWS households (CIR=1.93, 1.15–3.22) and more typhoid than IWS households meeting the UN criterion (CIR=3.66, 1.37–9.79). Households in the bottom quintiles of supply frequency and duration had roughly twice as many typhoid cases as CWS households (p<0.05).
The study evaluated IWS performance against the UN Sustainable Development Goals' availability threshold of having water when needed—operationalized as ≥12 hours/day or ≥4 days/week. IWS households that met this UN criterion or that were in the highest quintiles for frequency and duration showed health outcomes comparable to CWS households for both child diarrhea and typhoid incidence. IWS households failing to meet the UN criterion had significantly worse outcomes for both indicators, and the lowest-frequency/duration quintiles had the worst outcomes.
The authors interpret these findings as supporting global efforts to implement continuous water supply as a measure to reduce waterborne infections. They further highlight a pragmatic interim message: increasing supply frequency and duration in existing IWS systems—moving toward the UN criterion of availability when needed—can deliver measurable reductions in child diarrhea prevalence and typhoid cases. The study therefore provides empirical evidence linking specific features of supply intermittency (frequency and duration) to health outcomes in a large matched cohort.
The article states that ethical approval was obtained from the Institutional Review Board of the University of Massachusetts, Amherst (protocol 2018-5063) with a reliance agreement by North Carolina State University, and that participant consent procedures were followed. The authors declare no competing interests. All data produced will be made available on the Open Science Framework upon publication.
The manuscript is posted as a preprint and has not been peer reviewed; the report itself notes this status. Specific methodological details beyond those summarized in the abstract—such as additional covariates included in adjusted models, the exact matching algorithm and balance diagnostics, or details of case ascertainment for typhoid—are not fully reported in the abstract and would need to be consulted in the full preprint and supplementary materials for comprehensive appraisal.
Conclusion
In this Hubli-Dharwad study, households served by intermittent water supply experienced higher child diarrhea prevalence and greater typhoid incidence compared with households on continuous water supply. IWS systems that provided water with greater frequency and longer duration, including those meeting the UN availability criterion, had health outcomes similar to CWS. The findings support implementing CWS where feasible and improving frequency and duration in IWS systems as an interim public health measure.