Thank you for visiting Nature. The browser version you are using has limited support for CSS. For the best experience, we recommend that you use a newer version of the browser (or turn off the compatibility mode in Internet Explorer). At the same time, to ensure continued support, we will display sites without styles and JavaScript.

npj Clean Water Volume 4, Article Number: 47 (2021) Cite this article

Reliable access to water, sanitation, and personal hygiene (WASH) services is an important part of children's health and development. However, since there are few piped water systems close to household taps in rural areas in sub-Saharan Africa, there is limited evidence of its impact. We conducted a quasi-experimental study in four rural areas in southern Zambia from April 2018 to May 2019, in which we measured the impact of installing a local piped water supply system on storage water and fecal contamination of the hands of caregivers. Access to tap water was associated with a 0.5 log10 reduction in E. coli concentration in drinking water (p <0.05), but there was no change in hand contamination. The pipeline water supply system in this study reduced the median distance to a safe drinking water source by more than 90%, but the microbiological results we measured improved only slightly, while the self-reported household water storage duration remained unchanged. These findings emphasize the need for future impact assessments on pipeline water supply systems to measure a comprehensive set of indicators directly related to human well-being, such as saving time.

Reliable access to water, sanitation and personal hygiene (WASH) services is an important part of the health and development of young children1. In the past 20 years, significant progress has been made in obtaining improved water sources; however, progress in sub-Saharan Africa is much more limited than in other regions2. Due to insufficient WASH services, children living in low- and middle-income countries bear a heavy burden of disease1. In 2016, diarrhoeal disease alone caused the deaths of nearly 300,000 children under 5 years of age, of which about 50% occurred in the sub-Saharan region. South Africa 1.

There are many ways for children to come into contact with fecal contamination, including their own hands, the hands of their caregivers and stored drinking water3, 4, 5, 6, 7, 8, 9. Recent studies have measured the health effects of increased use of chlorine-based water for disinfection, improved pit toilets, child manure management tools, and portable hand washing stations10,11,12. However, these family-level interventions did not significantly reduce the prevalence of linear growth in children under 5 years of age10 or diarrhea11,12. The limited impact observed in these studies has led some to suggest that future investments must focus on transformative solutions that are “more effective in reducing fecal pollution”13,14.

A piped water supply system with a yard or household tap is designed to distribute a relatively large amount of water to a convenient nearby location. It is assumed that the transition to these systems is to improve access to safe drinking water and increase the frequency and thoroughness of hand washing15,16. However, the per capita investment in household water supply in pipelined water supply systems is also much higher. Such systems are rare in rural areas of low-income countries, especially in sub-Saharan Africa. Currently, only 18% of rural households in sub-Saharan Africa have access to tap water, and since 2002, coverage has only expanded by 5 percentage points.

As pipelines, local water supply systems are not common in low- and middle-income countries, and they are basically missing in the literature that measures their impact on hand hygiene, water quality, or health. For example, studies that measure hand hygiene indicators do not include households that rely on piped water sources, which may obscure the relationship between water sources, water supply, and hand hygiene3,17. A recent systematic review and meta-analysis by Wolf et al.18 included 70 studies on shared community resources, and 9 studies on pipelines and local services. This is in stark contrast to high-income countries, which have thoroughly investigated the relationship between treated tap water and health19,20.

However, similar to other interventions designed to interrupt household fecal exposure 11,12, studies on the impact of transitioning to tap water do not show that the prevalence of diarrhoeal diseases has indeed decreased. For example, a review by Overbo et al. 21 found that in five of ten rigorous studies, there was a statistically significant association between on-site piped water supply and the reduction of diarrhoeal disease in children under 5 years old, while the remaining studies reported The result of is invalid. It is necessary to further clarify the ways in which piped water can significantly reduce exposure to fecal pathogens and/or improve children’s health.

In this study, we first hypothesized the mechanism by which the transition from unimproved off-site water supply to local piped water supply affects fecal exposure and children’s health. We then used empirical data from rural Zambia to understand the impact of this transition on the combination of microbes and child health outcomes reported by caregivers.

Our conceptual model was developed through a literature review and pilot study conducted in Zambia in January 2018 (Figure 1). The model elaborated on the hypothetical impact of interventions (access to shared or private plumbing taps) on fecal contamination on the hands and in the water. At our research site, the pipeline water supply system is connected to a deep groundwater source without fecal pollution. We assume that by bringing the tap water closer to the home, the home will use more water and use more water for sanitary behaviors such as hand washing15,16. Washing hands more frequently is expected to reduce fecal contamination on the hands of caregivers, in which case the caregivers are usually women. By cleaning their hands more frequently and/or thoroughly, caregivers can reduce the chance of contaminating the food and water they prepare for their children, thereby reducing their children’s exposure to fecal pathogens 22,23. Cleaning young children's hands more frequently can also reduce the chance of ingesting feces when they open their hands8,24.

Hypothetical relationship between pipeline water supply and main research results.

We also hypothesized that bringing the water supply closer to the household will cause the household to fetch water more frequently, but with less water per trip. If the family only fetches water when needed, we assume that this will reduce dependence on the family’s storage of water. Previous systematic reviews and meta-analysis found that storing water without a lid or on the ground is related to water pollution observed after collection from the source25,26. Reducing the time of storing water in the home is expected to result in lower pollution of stored water and lower consumption of water with a lower concentration of fecal indicator bacteria, which is related to a lower prevalence of diarrhea27.

The study was conducted in four villages in southern Zambia in 2018-19. The two treatment villages are equipped with a mechanized pipeline water supply network, which pumps untreated deep groundwater to 20-25 taps through underground pipelines and submersible pumps. After installation, each faucet is used by 1 to 4 households [average = 2.9, SD = 1.2]. The map of the study area is included in Supplementary Information (SI) (Supplementary Figure 2).

At the baseline, there were no significant differences in the socioeconomic or demographic characteristics measured between the two treatment villages and the two comparison villages (Table 1). A higher percentage of households in comparison villages have improved primary drinking water sources (36% vs. 7%, p <0.001), which is consistent with the focus of implementing partner World Vision, which prioritizes projects in the poorest communities. This difference is caused by the fact that the treatment households at the baseline almost completely rely on unimproved wells as the main water source, while the comparison households can use hand pumps to pump water. Unimproved wells are classified according to the definition of the joint monitoring plan, which refers to wells dug without linings, aprons and/or coverings. In addition, before collecting baseline data, World Vision instructed the processing family to build a handwashing station and a shared or private toilet as a prerequisite for the completion of the pipeline water supply system. This resulted in the reported use of shared or private toilets to defecate between the treatment group and the control group (82% vs. 48%, p <0.001) and the observed presence of handwashing stations (57% vs. 9%, p <0.001).

Concentrations of E. coli in all source water samples from shared or private taps (N = 16) were below the detection limit of 1 most probable number (MPN)/100 mL (-0.3 log10 MPN/100 mL) (figure 2). In samples collected in three time periods, we measured the median of 2.0 log10 MPN E. coli/100 mL in unmodified wells (interquartile range [IQR] = 1.4 to 2.4 log10 MPN/100 mL, N = 78 ), with a median of -0.3 log10 MPN/100 mL (IQR = -0.3 to -0.3 log10 MPN/100 mL, N = 17) in a borehole using a hand pump. Overall, 99% of unimproved water well samples, 18% of hand pump drilling samples, and 0% of shared or private faucet samples contained detectable E. coli. Other descriptive statistics can be found in SI (Supplementary Table 4).

Box plot of E. coli concentration (log10 MPN/100 mL) in source water samples by source type, combined in all phases of the study.

At baseline, households use hand pumps to draw water from unimproved wells and boreholes. The average distance between a family residence and its main water source is 286 m (IQR = 109–471 m). After installing shared and private faucets, the families in the treatment group lived at a median distance of 13 m from the main drinking water source (IQR = 8-27 m). There was no measurement change in the water source access of the comparative villages, therefore, their distance from the main water source did not change during the study period. All distances are measured using global positioning system equipment.

We found that there are only marginal differences in the frequency of self-reported hand washing, measured as the self-reported hand-washing event the day before the interview, treatment and the finish line were compared between families [estimated value = 0.6 hand-washing event of the day, 95% confidence interval (CI) = (- 0.0 to 1.2), p = 0.07] (full model reported in SI, Supplementary Table 2). Similarly, when using the microbiological analysis of hand sanitizer samples, we measured no significant changes in MPN E. coli per hand (estimated value = -0.3 log10 MPN per two hands, 95% CI (-0.9 to 0.3), p = 0.4) (Supplementary Table 2).

Receiving tap water is associated with a 71% reduction in E. coli concentration in storage water (307 observations in 148 households) (estimated value = -0.5 log10 MPN, 95% CI: (-1.0 to 0.0), p = 0.03, Table 2, in SI The complete model of the report, Supplementary Table 3). However, in 63% (N = 33) of the treatment households, E. coli contamination continued to exceed the detection limit. Descriptive statistics of E. coli concentration by treatment group and stage can be found in SI Supplementary Table 5.

In the household survey, treatment and control households reported that on the day of the interview, the terminal line fetched water more frequently than at baseline, but the difference between treatment groups was not statistically significant (χ2 correlation test, two-sided, χ2 = 0.2, p = 0.6) (Figure 3). In addition, we found no significant correlation between the distance to the main water source and the storage time. In other words, households living less than 13 m from the main water source (quintile 1) reported the same storage practices as households living more than 500 m (quintile) from the main water source (other details in the SI). Information, Supplementary Figure 1). 3).

Self-reported water storage collection days at home by group and research stage. The area of ​​the circle represents the relative proportion in each column.

In treating families, there was no significant difference between the baseline and end points of the 7-day prevalence of diarrhoeal disease reported by nursing staff in children under 5 years of age (18% to 9%, p = 0.24, Figure 4). However, in comparative households, this indicator has a significant increase (5% to 28%, p <0.001). In summary, the provision of tap water is associated with a decrease in the chance of nurses reporting diarrhea in children under 5 years of age (odds ratio = 0.1, 95% CI: (0.0-0.7), p <0.05). The analysis was performed on subsamples of our population that reported having children under 5 years of age during the study period (N = 79 at baseline and N = 74 at endpoint). In any cohort, whether it was toothache or cuts and abrasions, the health results of the negative control did not change significantly over time.

The 7-day prevalence of clinical diarrhea in children under 5 years of age reported by caregivers, divided by study stage and group.

Shared and private pipe faucets in close proximity to rural households in Zambia are related to the reduction of water pollution at the point of consumption, but they are statistically significant. After receiving the intervention, there was no significant change in self-reported hand washing frequency or microbiological indicators of hand contamination.

Receiving tap water is associated with a 71% (0.5 log10) reduction in the concentration of fecal indicator bacteria in stored drinking water (MPN/100 mL). The scale of these results is similar to another recent study of Zimbabwe's pipeline water system7. However, of the 52 (63%) stored water samples of the processing households, 33 continued to have measurable E. coli contamination. These findings indicate that pollution occurs between collection and consumption at the source, even in households that draw water from nearby piped sources.

Post-supply pollution is related to the expansion of household storage before consumption25,30. We assume that households that have access to tap water will store it for a shorter period of time before using it (Figure 1). However, we did not find evidence of a difference in storage time between treatment households and comparison households, nor did we find any correlation between storage time and distance to the main source (Figure 3 and Supplementary Figure 3). We speculate that these findings are due to community members' lack of trust in the continued functioning and sustainability of the water system. The intermittent supply may be the key factor explaining the continued storage of water by households. If the water supply may be exhausted or accidentally shut down, it is risky for the household to give up water storage altogether. Therefore, even if the faucet is close to home, the previously adopted water storage behavior may not change in the short term. In addition, although the pipeline water supply system was in operation throughout the investigation period, there is sufficient evidence that the rural water supply infrastructure often fails 31, 32, 33, 34. This may cause families to continue their previous storage behaviors until they have confidence in the long-term function of the system. Given the short time between intervention and evaluation (<12 months), respondents are unlikely to build this level of trust or change entrenched water storage behaviors. However, a long-term assessment of water storage behavior may reveal greater changes.

The existing literature on hand washing shows that improving access to water is the main obstacle to increasing the frequency of hand washing 15,16,35,36,37. In addition, the supply of water may be related to the thoroughness of hand washing or the increase in duration. However, we found no change in the number of self-reported hand washing incidents or microbiological indicators of hand contamination (E. coli log10 MPN per two hands). Previous research has shown that self-reported hand-washing frequency is an indicator that is susceptible to deviations in social expectations. Compared with observed or directly measured frequency, it may represent an exaggerated measure of hand-washing frequency38,39. Therefore, the use of this indicator has no significant impact and strengthens our assessment, that is, the handwashing rate has not increased significantly. This finding is consistent with evidence from high-income countries that isolated infrastructure interventions are not sufficient to promote improved hand hygiene behavior40. Therefore, more community or family-level information may be needed to further encourage hand washing.

We measured the significant increase in the prevalence of diarrhea in the comparison village and the non-significant decrease in the treatment village. An analysis of the differences in the differences revealed that there was a statistically significant reduction in diarrhea reported by caregivers of children under 5 years of age. Comparing the long-term increase in the prevalence of diarrhoea in villages is an unexpected finding and has made an important contribution to the difference in the difference in results. Although the protective effect of the tap water system may have prevented treatment villages from experiencing a similar increase in the prevalence of diarrhea during the study period, we believe that other factors may contribute to this finding. For example, the interviewee may answer our question about the prevalence of diarrhea, intentionally or unintentionally, to show need, gratitude, or for other reasons41. This may be especially common in comparative villages, where no benefits were obtained during the study period. However, we did not see evidence of this behavior in the measurement of other health indicators, which suggests that if this deviation occurs, it is isolated from the report of diarrhea episodes.

In addition, we acknowledge that the allocation of treatment status in this study is non-randomized. Although the comparison villages were matched according to a set of pre-specified covariates, there was a significant difference in the proportion of households using shared or private toilets and hand washing stations at baseline. Although our model statistically controls for these differences, non-random allocation may also cause other unmeasured confounding factors to be unbalanced. In addition, the number of diarrhea episodes in our study population is very small. Therefore, although we have found that tap water infrastructure has a significant protective effect on diarrhoeal diseases, we believe that these findings should be interpreted with caution before being confirmed by follow-up studies.

Finally, our research only focuses on the potential advantages of the piped water supply system in the sample. Researchers have no access to cost data for installation, maintenance, and management, which represents the limitations of this work. Although some literature provides life-cycle cost estimates for rural pipeline water systems in low-income countries, the analysis was conducted on a system several orders of magnitude larger than the system in this study42. Therefore, there are still gaps in the literature on measuring the life cycle costs of small rural water supply systems.

A recent large-scale randomized controlled experiment found that improvements in household WASH have limited impact on children’s health and linear growth10,11,12. As a result, there are calls for additional investment in civil infrastructure, local water sources and more “transformation” solutions2,13,14. The piped water supply system in this study embodies a transformational approach, providing a local piped water supply system that can provide fecal-free water and shorten the distance to the main drinking water source by more than 90%. Nevertheless, we only measured a small improvement in the quality of stored water, and there were no changes in the concentration of E. coli on the hands, the frequency of hand washing, or the amount of water stored in the household.

However, the provision of locally piped water is still a worthy goal, related to the substantial improvement of results directly related to human well-being. For example, similar interventions are related to time savings 43, household income 44, 45, water production uses 46, food security 47, musculoskeletal injuries 48, 49 and improvements in psychosocial health 50, 51. Therefore, we recommend that future research aims to understand the potential impact of transformative WASH. In addition to further examination of microbiological indicators, a wide range of well-being measures should be prioritized, and health indicators should be verified where possible.

The study was conducted in two neighboring regions in southern Zambia in 2018-19. In 2018, the international non-governmental organization World Vision built 14 water supply systems in rural areas. The local World Vision resident office selected villages that had access to a piped water supply network based on a relatively high level of perceived demand. The treatment and comparison villages are matched based on their access to health care facilities and their proximity to the elementary school. In two treatment villages (T1 and T2) and one comparison village (C1), all female heads of household were required to participate in the research. In the second comparative village (C2), due to its large scale, a systematic sampling procedure was used to select two from every three households. One family refused to participate in the baseline. The research design implemented is a non-random, quasi-experimental design. The SI contains a flowchart of the participants and a map of the study site (Supplementary Figures 1 and 2). We report our methods and results in accordance with the non-random design transparent report evaluation guidelines in SI (Supplementary Table 1)52.

The treatment village is equipped with a mechanized pipeline water supply network, including underground PVC pipelines and submersible pumps. In these two villages, 20-25 faucets were installed, mixing private courtyard faucets and faucets shared by two to five households. Water supply is about 8-16 hours per day. The drilling and piping systems in the research village use deep underground water sources.

Data collection occurred in May 2018 (baseline), September 2018 (midline) and June 2019 (final line). Baseline and final line data collection occurs at the beginning of the dry season, while midline data collection occurs at the end of the dry season. The interview was conducted in the participant’s yard and covered topics including water collection and treatment practices, sanitation and hygiene behaviors. A total of 434 household surveys were conducted during the three study periods. The interview was conducted by Zambian research assistants in Tongan, the local language of the research site. They have completed 2 weeks of intensive training and pre-testing, and are fluent in Tongan and English. All surveys are written in English, translated into Tonga by a third-party translator, and back-translated into English by the research assistant. All data is collected using a tablet or mobile phone using SurveyCTO software (Dobility, Cambridge, MA).

During each data collection period, the survey team visited the same households for interviews. At the baseline, 154 households were interviewed. At the midline and final line, 90% (N = 138) and 81% (N = 125) of households completed the repeat survey respectively. Three-quarters (N = 114) of households were interviewed in all three time periods. A maximum of 3 interview attempts were made with each family to maximize the retention of the entire research phase.

The main result of the study was the level of E. coli in the respondent's hands and stored drinking water at home. The secondary results were the level of E. coli in the water collected directly from the source, the interviewee's self-reported frequency of washing hands with soap, self-reported storage time at home, and the 7-day prevalence of diarrhea reported by nursing staff. child.

To measure hand contamination, we collected hand sanitizer samples from a subset of households that also provided stored water samples during each data collection period. Due to the limitation of on-site laboratory capacity, it is impossible to collect hand sanitizer samples from the same household at every time period. Therefore, during the three data collection periods, 175 total samples were drawn from 109 unique households. Following the procedure described in 37, samples were collected by asking the interviewees to immerse their hands in a sterile Whirl-Pak bag containing 350 ml of distilled water. We also asked respondents how many times they washed their hands in the previous day and the number of times they washed their hands with soap.

To measure E. coli contamination in storage water, we collected stored water samples from a subset of respondents (N = 267 samples from 148 unique households) during all three data collection periods. Respondents were asked to pour about 300 ml of water from the container they currently use (or will use) in the way they usually serve themselves, and put it into a sterile Whirl-Pak bag. The samples are kept on ice and shipped to the on-site laboratory for analysis within 6 hours of collection. Due to traffic restrictions from the rural community to the on-site laboratory, if the interview starts before 1100 hours, no water and hand samples will be collected, as the time between sampling and processing is expected to exceed 6 hours.

To measure the duration of household water storage, the interviewee was asked: "If you drink a glass of water now, where would you get it from?". If they mentioned a water storage container, they would be asked, "When was this water taken?" and the source of the water. The response record is the afternoon of the current day, the morning of the current day, yesterday, the day before yesterday, or before.

The IDEXX Colilert and Quanti-tray 2000 process (Westbrook, ME) was used to analyze E. coli in all source water, storage water, and hand rinse samples. In order to measure the E. coli pollution in the village's water sources, we identified all the water sources that the respondents drank during the household survey (N = 111 during the three data collection periods). During each data collection period, the enumerator collected approximately 300 milliliters of water from each water source. The samples were collected in sterile Whirl-Pak bags (Nasco, Fort Atkinson, WI), stored on ice, and then transported to the on-site laboratory for analysis within 6 hours after the sample was collected.

For hand-washed samples, after stirring, transfer 100 mL of water from a 350 mL sample bag for analysis. The detection limit is 3.5 MPN/2 lots and 8468 MPN/2 lots. Water samples with a concentration below the detection limit are designated as 0.5 MPN, and water samples with a concentration above the upper limit are designated as 2420 MPN/100 mL. Hand samples with concentrations below the detection limit were designated as 2 MPN/2 hands, and samples with concentrations above the upper limit were designated as 8469 MPN/2 hands53.

A field blank, a laboratory blank, and a randomly selected laboratory repeat (two aliquots from the same sample bag) are performed daily for quality control. The on-site blank is a sample bag filled with distilled water, carried in the refrigerated cooler to the research village, opened and closed outside the project vehicle, and returned to the refrigerated cooler. The laboratory blank is 100 ml aliquots of laboratory distilled water. All blanks were tested for E. coli. According to the processed sample volume, the detection limits are 1 MPN/100 ml and 2419 MPN/100 mL. During the study period, E. coli was not detected in any blank samples on site or in the laboratory.

We measured the prevalence of diarrhea in children under 5 years of age through a caregiver survey. Caregivers of weaned children under 5 years of age were asked whether their children had experienced specific health symptoms in the past 7 days. We use the World Health Organization's case definition of diarrhea (3 or more loose or watery stools in the past 24 hours) as our indicator of diarrhea disease. Nursing staff were also asked about other health symptoms of the child, including fever, vomiting, and cough, and two negative controls: toothache, cuts, and abrasions54.

The data was analyzed using R version 3.6.1 (R Foundation, Vienna, Austria). We used a linear mixed-effects model with a random intercept to calculate the difference in log10 transformed E. coli concentrations between hand sanitizer and storage water to explain the clustering at the household level using lme4 packages55,56. Both models control the choice of treatment group, the type of toilet reported for adults to defecate (if any), the observed presence of dedicated handwashing stations for household use, and whether soap interviews are observed during the handwashing station. These covariates were pre-specified based on previous studies that reported the possibility of hand-contaminated water during the extraction event17, 30, 57. A generalized linear mixed-effects model with the same control variables, random effects, and analysis package was used to measure the difference in the prevalence of diarrhea under 5 years old and the duration of water storage reported by nursing staff. The model that estimates the difference in stored water quality uses an additional control variable to determine whether households report "somewhat frequently" or "frequently" treating their drinking water. We use the DHARMa package (version 0.4.1) in R to verify that all the assumptions of the linear and generalized linear mixed-effects model are satisfied58. All models used data from all three data collection periods.

All models include the comprehensive wealth score as the final control variable. The SI contains information on how to calculate the wealth score.

We define intervention exposure as households reporting that they use shared or private taps connected to the village's water network as their main source of water. Fourteen families lived in a treatment village but did not receive the services of the new system, so they were excluded from the sample. Therefore, the analysis of all reports is "treatment of the person being treated", of which only those who use tap water as the main water source are considered as treatment group59.

Ethical approval was obtained from the Institutional Review Board of Stanford University (Protocol 44976) and the Biomedical Research Ethics Committee of the University of Zambia (Reference 001-03-18). Before the start of the research activity, the researchers met with community leaders in four villages to describe the research plan. Before participating, all interviewees voluntarily provided informed verbal consent to enumerators proficient in the local language. All study participants are at least 18 years old.

Because the depth of the collected household-level survey data is very sensitive, the data supporting the results of this study are not publicly available. However, all anonymous data sets generated and/or analyzed during the current research period, as well as the accompanying code, may be obtained from the corresponding author upon reasonable request.

Prüss-Ustün, A. etc. Disease burden due to selected adverse health outcomes due to insufficient water, sanitation, and personal hygiene: the latest analysis for low- and middle-income countries. internationality. J. Hagrid. environment. Health 222, 765–777 (2019).

WHO/UNICEF joint monitoring plan. The progress in household drinking water, sanitation and sanitation from 2000 to 2017. Pay special attention to inequality. WHO/UNICEF joint monitoring plan, 2019).

Ercumen, A. et al. The impact of single and combined water, sanitation, and handwashing interventions on fecal pollution in the household environment: a cluster randomized controlled trial in rural Bangladesh. environment. science. technology. 52, 12078–12088 (2018).

Ercumen, A. et al. Can improved sanitation facilities reduce the contamination of water, hands, food, soil and flies by feces? Evidence from cluster randomized controlled trials in rural Bangladesh. environment. science. technology. 52, 12089–12097 (2018).

Pickering, AJ, etc. Fecal indicator bacteria and subsequent childhood diarrhea along multiple environmental transmission routes (water, hands, food, soil, flies) in rural Bangladesh. environment. science. technology. 52, 7928–7936 (2018).

Mattioli, M., Davis, J., Mrisho, M. & Boehm, AB Bagamo, Tanzania, quantitative human norovirus GII in the hands of mothers of children under 5 years of age. Yes. J. Tropical. medicine. Hagrid. 93, 478–484 (2015).

Navab-Daneshmand, T. etc. E. coli contamination in multiple environmental compartments (soil, hands, drinking water and handwashing water) in Harare: relevance and risk factors. Yes. J. Tropical. medicine. Hagrid. 98, 803–813 (2018).

Kwong, LH, etc. Participating in a cluster randomized trial of water, sanitation and sanitation intervention (WASH welfare), rural children in Bangladesh ingest fecal bacteria along multiple routes, https://doi.org/10.1021/acs.est.0c02606 (2020) ).

Holcomb, DA etc. In Maputo, Mozambique, the water, soil and surfaces of households sharing inferior sanitation facilities are contaminated with human feces. internationality. J. Hagrid. environment. Health 226, 113496 (2020).

Rupee, SP, etc. The impact of water quality, sanitation, hand washing and nutrition interventions on diarrhea and child growth in rural Bangladesh: a cluster randomized controlled trial. Lancet ball. heal. 6. E302–e315 (2018).

Null, C. etc. The impact of water quality, sanitation, hand washing and nutrition interventions on diarrhea and child growth in rural Kenya: a cluster randomized controlled trial. Lancet ball. heal. 6. E316–e329 (2018).

Humphrey, JH, etc. The independent and combined effects of improved water, sanitation, and personal hygiene, as well as improved complementary foods, on stunting and anemia in rural Zimbabwean children: a cluster randomized trial. Lancet ball. heal. 7. e132-e147 (2019).

Pickering, AJ, etc. WASH benefits and SHINE trial: Explain the effect of WASH intervention on linear growth and diarrhea. Lancet ball. heal. 7. e1139–e1146 (2019).

Cumming, O. et al. The impact of three major new trials on the impact of water, sanitation, and hygiene on children with diarrhea and stunting: a consensus statement. BMC medicine. 17, 1-9 (2019).

Oswald, we wait. Provide private piped water supply and sewer connections in communities in the suburbs of Lima, Peru, and directly observe the mother’s hand washing. tropical. medicine. internationality. heal. 19, 388–397 (2014).

Stelmach, RD & Clasen, T. Household water consumption and health: a systematic review. internationality. J. Environment. Reservoir Public Health 12, 5954–5974 (2015).

Pickering, AJ, etc. Hands, water and health: fecal pollution in Tanzanian communities has improved non-connected water supply. environment. science. technology. 44, 3267–3272 (2010).

Wolf, J. et al. The impact of drinking water, sanitation and hand washing with soap on childhood diarrhoeal disease: an updated meta-analysis and meta-regression. tropical. medicine. internationality. heal. 23, 508–525, https://doi.org/10.1111/tmi.13051 (2018).

Alsan, M. and Goldin, C. Watershed in Child Mortality: The Role of Effective Water Supply and Wastewater Treatment Infrastructure, 1880-1920 Marcella Alsan Claudia Goldin. J. Politics. economy. 127, 586–638 (2019).

Cutler, D. & Miller, G. The role of public health improvement in health progress: America in the 20th century. Demographics 42, 1-22 (2005).

Overbo, A., Williams, AR, Evans, B., Hunter, PR & Bartram, J. On-site drinking water supply and health: a systematic review. internationality. J. Hagrid. environment. Health 219, 317–330 (2016).

Mattioli, M., Davis, J. & Boehm, AB In Tanzania, the fecal intake caused by hand-to-mouth contact is greater than the water intake in the diet: a quantitative fecal exposure assessment model. environment. science. technology. 49, 1912-1920 (2015).

Pickering, AJ, Julian, TR, Mamuya, S., Boehm, AB & Davis, J. Bacterial contamination of the hands of Tanzanian mothers changes with time and family activities. tropical. medicine. internationality. heal. 16, 233–239 (2011).

Kwong, LH, etc. Mouths and object mouths of children aged 3 to 18 months in rural Bangladesh. internationality. J. Environment. Reservoir Public Health 13, 563 (2016).

Wright, J., Gundry, S. & Conroy, R. Domestic drinking water in developing countries: a systematic review of microbial contamination between source and point of use. tropical. medicine. internationality. heal. 9, 106–117 (2004).

Shields, KF, Bain, RES, Cronk, R., Wright, JA & Bartram, J. Association of Supply Types of Fecal Pollution in Source Water and Household Storage Drinking Water in Developing Countries: A Bivariate Meta-Analysis. environment. Health perspective. 123, 1222–1231 (2015).

Rupee, SP, etc. Microbial contamination of drinking water associated with subsequent childhood diarrhea. Yes. J. Tropical. medicine. Hagrid. 93, 904–911 (2015).

WHO/UNICEF. The core issues of drinking water and sanitation in household surveys (WHO/UNICEF, 2018).

World Health Organization. Guidelines for drinking water quality (World Health Organization, 2017).

Harris, AR, Boehm, AB & Davis, J. Pollution mechanism after drinking water supply in Bagamoyo, Tanzania. J. Water Health 11, 543–554 (2013).

Foster, T. Predicting the sustainability of hand pump management in communities in sub-Saharan Africa: Evidence from Liberia, Sierra Leone, and Uganda. environment. science. technology. 47, 12037–12046 (2013).

Marks, SJ & Davis, J. Will user participation lead to a sense of ownership in the rural water supply system? Evidence from Kenya. World Development 40, 1569–1576 (2012).

Whittington, D. et al. How does the demand-driven community management model of the rural water supply system perform? Evidence from Bolivia, Peru and Ghana. Water Policy 11, 696–718 (2009).

Dhoba, L. Expanding the scale of rural water supply: Rethinking the experience of the pipeline water supply program managed by rural communities in Zimbabwe. J. Water sanitation. Hagrid. Development. 527-538, https://doi.org/10.2166/washdev.2020.081 (2020).

Coombes, Y. & Devine, J. Introducing foam: Analyze handwashing behavior to design a framework for handwashing procedures. Water and Sanitation Planning Working Paper (World Bank, 2010).

Dreibelbis, R. et al. Comprehensive behavioral models for water, sanitation, and personal hygiene: a systematic review of behavioral models and a framework for designing and evaluating behavior change interventions in infrastructure-constrained environments. BMC Public Health 13, 1015 (2013).

Pickering, AJ, Boehm, AB, Mwanjali, M. & Davis, J. The efficacy of waterless hand hygiene compared with hand washing with soap: a field study conducted in Dar es Salaam, Tanzania. Yes. J. Tropical. medicine. Hagrid. 82, 270–278, https://doi.org/10.1111/tmi.13051 (2010).

Biran, A. et al. Compare the performance of the handwashing habit indicators of rural households in India. tropical. medicine. internationality. heal. 13, 278–285 (2008).

Ram, PK, etc. Is structured observation an effective technique for measuring hand washing behavior? Use an acceleration sensor embedded in the soap to evaluate the responsiveness to structured observations. Yes. J. Tropical. medicine. Hagrid. 83, 1070–1076 (2010).

Gould, DJ, Moralejo, D., Drey, N., Chudleigh, JH & Taljaard, M. Interventions to improve hand hygiene compliance in patient care. Cochrane database system. Revised version https://doi.org/10.1002/14651858.CD005186.pub4 (2017).

Zwane, AP, etc. The investigation can change the behavior and related parameter estimates in the future. Process National Academy of Sciences. science. United States 108, 1821–1826 (2011).

Nyarko, KB, Dwumfour-Asare, B., Appiah-Effah, E. & Moriarty, P. Cost of water service in rural areas and small towns in Ghana. In: IRC Symposium 2010: Pumps, Piping and Commitments 172-191 (IRC WASH, 2010).

Winter, JC, Darmstadt, GL & Davis, J. The role of piped water supply in promoting the health, economic development and gender equality of rural communities. society. science. Medicine 270, 113599 (2021).

Renwick, M. et al. Multipurpose water service for the poor: assessing the state of knowledge (Winrock International, 2007).

Van Houweling, E., Hall, RP, Diop, AS, Davis, J. & Seiss, M. The role of water for production in women’s livelihoods: Evidence from rural Senegal. Alternate water. 5, 658–677 (2012).

Hall, RP, Vance, EA and van Houweling, E. The productive use of tap water in rural Senegal. Alternate water. 7, 480–498 (2014).

Brewis, A. et al. Household water insecurity is closely related to food insecurity: evidence from 27 locations in low- and middle-income countries. Yes. J. Humph. biology. 32, 1-13 (2020).

Geere, J. et al. In low-income countries, carrying water may be the main cause of disability from musculoskeletal diseases: cross-sectional surveys in South Africa, Ghana, and Vietnam. J. Globe. Health 8, 1-14 (2018).

Geere, JL, Hunter, PR & Jagals, P. Domestic water and its effects on health: a review and mixed method pilot study in Limpopo Province. South Africa. environment. heal. 9, 52 (2010).

Bisung, E. & Elliott, SJ The psychosocial impact of lack of water and sanitation in low- and middle-income countries: a scoping review. J. Water Health 15, 17-30 (2017).

Devoto, F., Duflo, E., Dupas, P., Parienté, W. & Pons, V. Happiness in tap water: Moroccan cities use tap water. Yes. economy. J. Economy. Policy 4, 68-99 (2012).

Des Jarlais, DC, Lyles, C. & Crepaz, N., TREND Group. Improving the quality of reporting of non-random evaluations of behavioral and public health interventions: a trend statement. Yes. J. Public Health 94, 361–366 (2004).

Eaton, AD, Clesceri, LS & Greenberg, AE Standard Methods for Water and Wastewater Testing (APHA, 2005).

Arnold, BF, Ercumen, A., Benjamin-Chung, J. & Colford, JM Brief report: Negative controls for detection of selection bias and measurement bias in epidemiological studies. Epidemiology 27, 637–641 (2016).

Winter, B. The linear model and linear mixed-effects model and language application in Winter, B. R. Preprint, https://arXiv.org/abs/1308.5499 (2013).

McCulloch, CE, Searle, SR & Neuhaus, JM Generalized linear mixed model. cycle. Biostatistics. 1-27, https://doi.org/10.1002/0470011815.b2a10021 (2005).

Trevett, AF, Carter, RC & Tyrrel, SF The mechanism that caused the deterioration of water quality in rural communities in Honduras after water supply. internationality. J. Hagrid. environment. Health 208, 153–161 (2005).

Hartig, F. DHARMa: Residual diagnosis of hierarchical (multi-level/mixed) regression models. R package version 0.4.4. http://florianhartig.github.io/DHARMa/ (2021).

Gupta, SK The concept of intention-to-treat: a review. perspective. clinical. Reservoir 2, 109–112 (2011).

JCW's graduate funding is provided by the Department of Civil and Environmental Engineering of Stanford University, the Stanford Graduate Scholarship, and the Fulbright Research Program in the United States. Additional funding for this research was provided by the Stanford Gold Center for Global Development and the Stanford Institute for Innovation in Developing Economies. Formative research is supported by the Stanford Office of International Affairs and the Stanford Global Engineering Project. We are grateful to the families who agreed to participate in this study. We also recognize the countless contributions of our team of investigators to the project: Angel Mutale, Chookwa Lubombo, Cholwe Mweene, Felistas Chuunga, Harriet Ng'ombe, Orety Handiindo, Puumba Muchindu, and Savior Sikalumbi. We thank the staff of World Vision, especially Maybin Ng'ambi, Emmanuel Nyundu, John Hasse and Mark Kelly, for their enthusiastic cooperation and assistance in the logistics work of Zambia.

Department of Civil and Environmental Engineering, Stanford University, Stanford, California, USA

James C. Winter, Alexandria B. Boehm & Jennifer Davis

Pediatrics, Stanford University School of Medicine, Stanford, California, USA

Woods Institute for the Environment, Stanford University, Stanford, California, USA

You can also search for this author in PubMed Google Scholar

You can also search for this author in PubMed Google Scholar

You can also search for this author in PubMed Google Scholar

You can also search for this author in PubMed Google Scholar

All authors contributed to the conceptualization. JCW, GLD, and JD contributed to the design of the survey tool. JCW, ABB and JD contributed to the development of laboratory methods. JCW drafts the manuscript text, conducts field surveys and conducts data analysis. JD, ABB and GLD provide supervision. JD and GLD provided funding. All authors participated in the review and editing of the manuscript.

Correspondence with James C. Winter.

World Vision funded the implementation of the pipeline water supply plan described in this study. Before and after the project, Stanford University received a separate grant from World Vision for unrelated projects, in which three authors (JCW, GLD, and JD) participated. The author declares no other competing interests.

The publisher states that Springer Nature remains neutral on the jurisdiction claims of published maps and agency affiliates.

Open Access This article has been licensed under the Creative Commons Attribution 4.0 International License Agreement, which permits use, sharing, adaptation, distribution and reproduction in any media or format, as long as you appropriately indicate the original author and source, and provide a link to the Creative Commons license , And indicate whether any changes have been made. The images or other third-party materials in this article are included in the article’s Creative Commons license, unless otherwise stated in the material’s credit line. If the article’s Creative Commons license does not include the material, and your intended use is not permitted by laws and regulations or exceeds the permitted use, you need to obtain permission directly from the copyright owner. To view a copy of this license, please visit http://creativecommons.org/licenses/by/4.0/.

Winter, JC, Darmstadt, GL, Boehm, AB, etc. The impact of on-site pipeline water supply on fecal pollution in rural Zambia. npj Clean Water 4, 47 (2021). https://doi.org/10.1038/s41545-021-00138-x

DOI: https://doi.org/10.1038/s41545-021-00138-x

Anyone you share the following link with can read this content:

Sorry, there is currently no shareable link in this article.

Provided by Springer Nature SharedIt content sharing program

npj Clean Water (npj Clean Water) ISSN 2059-7037 (online)