https://orcid.org/0000-0001-5069-5670
Figures
Abstract
Low-income urban areas experience challenges in accessing safe water due to poor infrastructure and illegal connections, leading to microbial contamination and worsening health risks. This study investigates the association of household water handling practices with microbial drinking water quality in Kibera – the largest urban informal settlement in Nairobi, Kenya. We conducted a cross-sectional study in December 2022 in Kibera with 262 randomly selected households to assess water handling practices, and physicochemical and microbial water quality. Most households (84%) had Free Chlorine Residual (FCR) below the recommended levels (≥ 0.2 mg/L) indicating insufficient protection against microbial contamination. Water samples from 56% and 15% of households tested positive for total coliforms and E. coli respectively. Only 42% of households treated their drinking water at home. Multivariate logistic regression revealed significantly higher odds (AOR = 3.16, p < 0.001) of contaminated drinking water (E. coli or total coliforms) amongst households not treating water. Those maintaining recommended FCR levels had substantially lower odds (AOR = 0.04, p < 0.001) of microbial contamination. In a bivariate sub-analysis, chlorine-based water treatment methods were protective against microbial contamination (OR=0.26, p = 0.025). In contrast, filtration-based methods were linked to higher odds of contamination (OR=3.25, p = 0.008) and boiling had no effect (p = 0.671). This study underscores the need for public health actors to promote effective and sustained water handling practices, especially for settings at high risk of contamination such as low-income urban areas.
Citation: Musyoka D, Agira J, Busienei P, Okoth B, Amondi C, Odundo J, et al. (2026) Household water handling practices and microbial water quality in informal settlements of Kibera, Kenya. PLOS Water 5(7): e0000577. https://doi.org/10.1371/journal.pwat.0000577
Editor: Michael B. Fisher, UNC-Chapel Hill: The University of North Carolina at Chapel Hill, UNITED STATES OF AMERICA
Received: April 5, 2025; Accepted: June 10, 2026; Published: July 10, 2026
Copyright: © 2026 Musyoka et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: The data supporting this manuscript has been uploaded as Supporting Information (S1 Data).
Funding: This research was funded by the Manitou Foundation - now KHR Family Fund - through the London School of Hygiene & Tropical Medicine (LSHTM) [Grant number: 58-2277856 to IR]. The funders had no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
In 2022, 1.7 billion people globally used a fecally-contaminated water source [1]. With a projected population of over 50 million people, the 2022 Kenya Demographic and Health Survey estimated that about 32% of Kenyans rely on unimproved water sources which are inadequately protected from fecal contamination [2]. Rapid urbanization and population growth has put pressure on existing water infrastructure, disproportionately affecting access to safe water in informal settlements where majority of urban poor populations reside [3]. These challenges are further compounded by climate change patterns, such as prolonged droughts and erratic rainfall patterns [4].
Low-income areas such as Kibera informal settlements in Nairobi continue to grapple with significant challenges in access to safe water due to lack of proper infrastructure for water supply [5]. There are, however, some efforts being made by the government, non-governmental organizations, and international agencies to improve water and sanitation infrastructure in low-income areas and informal settlements in Kenya [6,7]. In such areas, however, illegal connections, cracks, and leaks in the water distribution system are common, allowing contaminants to enter the water supply [8]. However, contamination can arise from improper hygiene practices during collection, use of unclean storage containers, unsanitary transportation methods, and inadequate treatment or filtration at the point of use [9–12]. These pathways significantly increase the risk of exposure to enteric pathogens that cause diarrhea, cholera and typhoid fever [13–17].
Household water handling practices such as improper storage, use of contaminated containers, and inadequate hygiene during water collection and use, contribute to poor drinking water quality in different settings [18,19]. Chlorine disinfection and use of clean storage containers with narrow opening for filling and dispensing water have been shown to significantly improve microbial quality of drinking water [20–22]. Intra-day variation in water quality are common in slum households with positive Escherichia coli (E. coli) tests in water from sources perceived as safe or classified as improved [23,24]. These perspectives on how water handling practices influence water quality lack insights on safety variations potentially experienced in unhygienic environments such as slums. Therefore, efforts to initially treat water by public or private utilities that supply water in informal settlements may be lost in numerous recontamination events during collection, storage, or usage if temporal dynamics of water quality are not considered.
Monitoring drinking water quality in urban slums remains uncommon and our understanding of how household water handling practices in these settings influence microbial water quality overlooks temporal dynamics in water handling. Water quality deterioration may occur with continued usage, and handling even within hours, yet such shorter time scales are often overlooked. Empirical evidence of superior practices that are protective over time in dynamic contexts such as urban slums is limited. This study investigates the effect of household water handling practices on drinking water quality in informal settlements of Kibera, Kenya. It attempts to show how household characteristics and water handling practices (status of the storage container, water treatment) influence household water quality.
Materials and methods
Ethics statement
This study was approved by the AMREF Ethics and Scientific Research Committee (protocol number AMREF-ESRC P1020-2021). The National Commission for Science, Technology, and Innovation granted a research permit (License No.: NACOSTI/P/22/19012) for the study. Further permits from the sub-national level were sought from the County Commissioner, and the Directorate of Health Services Nairobi City County. Additionally, we obtained written informed consent from adult participants who were eligible and had agreed to participate in the study, before commencing data collection. A copy of the participant information sheet and consent form, including contact information to reach the study team, was given to each participant for their records.
Study setting
This study was conducted in Kibera, an informal settlement located in the Southwest of Nairobi, the capital city of Kenya (Fig 1). Kibera has an approximate population of over 250,000 people [25] and is considered the most extensive and possibly the oldest current urban informal settlement in sub-Saharan Africa [26]. The UN Habitat estimates an average density of 87,500 inhabitants per km2 with Soweto West and Laini Saba at 129,000 and 48,000 inhabitants per km2 respectively [27]. Kibera is characterized by poor water, sanitation and hygiene (WASH) infrastructure with residents having limited access to essential WASH services [28]. Kibera’s piped water network is limited to a few households with the majority fetching water for drinking and other purposes from water kiosks usually connected to the Nairobi City Water and Sewerage Company (NCWSC) distribution grid where free chlorine levels are set to 0.5 mg/l at water collection points. About 50% of water points/kiosks in Kibera experience rationed water supply in at least 3 days a week and 60% of population has access to water kiosks within 50 meters, although Soweto West has poorer access to water kiosks compared to Makina and Laini Saba [27]. Other sources include boreholes operated by private individuals, non-governmental and/or faith-based organizations.
Map data from OpenStreetMap https://www.openstreetmap.org/. Shape file obtained from Humanitarian Data Exchange website https://data.humdata.org/dataset/kenya-elections.
Study design and sampling
We conducted a cross-sectional survey from 1st to 17th December 2022 in Makina, Laini Saba, and Soweto West villages in Kibera. The NCWSC distributes water into Kibera informal settlement through three service lines as shown in Table 1. Water from service lines is then distributed to users through cluster chambers which contain individual metered connections. A conceptual model of water distribution in Kibera is presented in Supporting Information, S1 Text. Our sampling approach was balanced between service line/cluster chamber and village representativeness. To ensure representativeness and through engagements with NCWSC, one village was selected from each of the three main service lines. Villages were purposively selected if they had a high density of functional connections to ensure representativeness in selection of users. To ensure representativeness, 8–9 water points were randomly selected per village. A maximum of 10 households was selected randomly within a 200m radius of the identified water points. In total, 262 households were selected in all villages comprising users and non-users of utility-supplied water.
Quantitative household survey
The study respondents were household heads or their spouses who were aged at least 18 years. A structured survey tool was administered to these respondents to understand WASH services in Kibera, household WASH practices, and estimate the occurrence of diarrheal disease in the households within the last 7 days (S2 Text). The survey tool integrated the observation of household water handling practices - including the type and status of water storage containers - and availability of handwashing supplies to complement information on reported behavior.
Onsite water testing
As part of the survey, participants were asked to provide a cup or jug of drinking water if available, fetching it as they would normally do for drinking. Approximately 500ml of the water was used to test for turbidity, free, and total chlorine while in the household. Water turbidity was tested using a calibrated turbidity tube with a range 5–500 NTU (DelAgua, UK) following manufacturer instructions. A Palintest pool-tester kit (Palintest, UK) was used to test for free and total chlorine following manufacturer instructions. Temperature, electrical conductivity, and pH of the water were also measured onsite using Hanna Combo pH/EC/TDS/C/PPM Tester (HI98129 Hanna Instruments, USA) by dipping the probe of the instrument into the water provided in a Whirlpak bag. The probe was held for a couple of minutes to achieve stability after which a reliable reading of temperature, electrical conductivity, and pH was taken.
Microbial drinking water quality
Up to 120 ml of household drinking water was aseptically collected in a Whirl Pak bag (Thermofisher, UK) with sodium thiosulfate, placed in a sample cooler bag, and transported to the laboratory for detection of fecal indicator bacteria within 6 hours. At the laboratory, the membrane filtration method (EPA Method 1604) was used to detect and enumerate E. coli and total coliforms with few alterations [29]. Briefly, 100 ml drinking water samples were filtered through a 0.45µm membrane filter which was then cultured on HiCromeTM Chromogenic Coliform agar plates (Himedia, India) and incubated at 37°C for 24 hours. All samples were filtered and tested undiluted with a detection limit of 1 (one) E. coli and/or 1 (one) total coliform per sample volume. After incubation, the results were read and interpreted based on the morphological characteristics and reported in CFU/100ml. Where present, E. coli colonies were further confirmed using the Indole test. Here, 5 distinct colonies of E. coli were picked from a plate and inoculated into 10 ml of tryptic soy broth and incubated at 37°C for 24 hours. Observation of color change (from clear to orange/pink) following the addition of three drops of Kovac’s Indole Reagent (Himedia, India) was confirmatory of E. coli colonies. Detection of either E. coli or total coliforms was indicative of microbial contamination of household drinking water.
Occurrence of fecal indicator bacteria (FIB) in drinking water can vary by seasonality [30]. Water sampling and microbial analysis was conducted in December, during short rains. A seasonal rainfall analysis from 1st October to 31st December 2022 reported by the Kenya Meterological Department shows that depressed rainfall was received over several stations including in Nairobi County. See S3 Text for daily precipitation estimates from the Climate Hazards Group InfraRed Precipitation with Station (CHIRPS) v2.0 data for the study area [31,32].
Quality assurance
Field staff received one-week intensive training on basic data collection procedures, navigation on the SurveyCTO platform, and key ethical considerations including voluntary participation, data confidentiality, and benefits. They were also taken through the proposed study design and objectives to gain a better understanding of the data collection tools. Handheld tablets preloaded with the survey and water quality questionnaire were used for data collection, enabling direct transmission of survey and onsite water quality test results to a central server whose access was restricted to the study management team. Submitted data were reviewed, cleaned, and checked for inconsistencies, duplicates, and out-of-range values. Any data found to be inconsistent was validated to verify the anomaly. Appropriate quality assurance and quality control on water samples were ensured by preparing blanks for every 20 Petri dishes to confirm that adequate sterile technique was employed. Blanks were only prepared in the lab using commercially available distilled water. Additionally, duplicate samples were prepared for every 10th household.
Data management and analysis
Microbial water quality was set as the outcome variable and was defined as the presence of either E. coli > 0 or total coliforms > 0 in water tested at the household level. This was classified as a binary outcome with 1 indicating > 0 CFU/100ml, and 0 indicating 0 CFU/100ml of either E. coli or total coliforms in drinking water. The binary classification of microbial quality was based on detection rather than enumeration of enteric pathogens as a proxy for fecal and environmental contamination and aligns with WHO guidelines of <1 CFU/100ml. Household characteristics and WASH practices were used as predictor variables; namely village, education level, source of drinking water, round trip travel time taken to fetch the water, status of the storage container, whether household treats water, water storage duration, chlorine levels in drinking water, sanitation facility used, handwashing facility used, and self-reported diarrheal episode in the last 7 days. Descriptive statistics were used to summarize variables, i.e., frequencies and percentages for categorical variables, and means and standard deviation for continuous variables. Free Chlorine Residual (FCR), the amount of chlorine remaining in treated water that is available to protect against microbial contamination, was classified as “below recommended levels” for values below 0.2 mg/L and “at recommended levels” for values equal or above 0.2mg/L. Experimental laboratory data was summarized descriptively (mean, standard deviation) and compared against WHO guidelines on drinking water (turbidity: < 5 NTU (nephelometric turbidity unit), FCR: ≥ 0.2 mg/L, pH: 6.5–8.5). The Kruskal Wallis test was used to determine differences in pH, temperature, and electrical conductivity of drinking water across villages. Significant differences were confirmed by Dunn’s test with p-value adjustment. The Chi-square test was used to assess the relationship between presence of E. coli or total coliforms and study villages. Concentrations of E. coli and total coliforms in the water samples were separately categorized from safe to high risk according to WHO guidelines.
Bivariate logistic regression was used to explore the effect of independent variables (household characteristics, and household WASH practices) and the outcome variable (drinking water quality). Independent variables that showed association with a p-value of < 0.2 were included in the multivariate analysis, with Adjusted Odds Ratio (AOR) reported. Only variables with a p-value less than 0.05 were declared as statistically significant. All data was managed and analyzed using STATA 17 (Stata Corp LP, College Station, TX, USA).
Results
Household demographics
A total of 262 respondents were interviewed, with similar proportions between the study villages as shown in Table 2. Most (91%) respondents were either heads of their households or were spouses of the household head. About 65% of participants were female, and the mean age was 39 years. Only 2% of respondents had not attended school while the rest had completed basic or advanced levels of education. Public tap/kiosks formed the main source of water (94%), and the majority (87%) of households reported taking ≤5 minutes per trip of fetching water and coming back home. We asked respondents to estimate how long household drinking water had been stored, to understand the duration of storage. Most households (72%) stored their water for less than 24 hours and only few (6%) stored water beyond 72 hours. The majority of households (84%) were observed to cover their water storage containers. Amongst the 42% of households who treated water at home, the reported methods were boiling (52%), filtration (29%), and chlorination (20%). The specific filtration method referred to here was the use of Sawyer PointONE filters based on a 0.1 micron fiber membrane.
The mean pH (7.7), temperature (22.3°C), and conductivity (128.23 μS/cm) were within desirable ranges (pH range 6.5-8.5; temperature ≤25°C; conductivity <1500 μS/cm) while about 98.5% of samples tested for turbidity had < 5 NTU indicating acceptable water clarity across villages. However, free and total chlorine levels in 84% of the samples were below the WHO-recommended value of ≥0.2 mg/L. E. coli and total coliforms were detected in 15% and 56% of household water samples respectively. While most samples were safe for E. coli, concentrations for positive samples spread from low to high risk, with concentrations for total coliforms falling under medium and high-risk categories. All laboratory blanks were negative for E. coli and total coliforms while duplicate samples demonstrated consistent results. There were no significant differences by village among the majority of microbial, physical or chemical properties except for electrical conductivity and total coliforms (Table 3).
Effect of household characteristics and water handling practices on microbial water quality
The results of the bivariate and multivariate logistic regression are shown in Table 4. Households residing in Makina village were twice as likely to have contaminated water (COR = 2.15, CI = 1.18-3.91) than those living in Laini Saba. An advanced education level of the household head or their spouse was associated with 53% reduced odds (COR = 0.47, CI = 0.23-0.98) of having contaminated drinking water compared to household heads possessing basic education. Likewise, households not treating drinking water at the point of use were at a higher risk of having contaminated water (presence of E. coli or total coliforms) than households who treated drinking water (COR = 2.24, CI = 1.36-3.70). Households whose drinking water had the recommended FCR levels (≥ 0.2 mg/L) were 94% less likely to test positive for E. coli or coliform (COR = 0.06, CI = 0.02-0.16) than households whose drinking water did not have the recommended FCR.
Longer water storage periods were associated with higher odds of contamination. Households who stored water for 49–72 hours were nearly three times (COR = 2.84, CI 1.21 - 6.68) as likely to test positive for E. coli or total coliforms than households whose water was stored for <24 hours. Those who stored water for >72 hours were nearly four times (COR = 3.96, CI = 1.08 - 14.48) likely to test positive for E. coli or total coliforms. An association between floor material and E. coli contamination in the domestic environment has been reported [33]. Accordingly, housing status and floor material variables were included as covariates in the multivariate analysis. Households that reported keeping livestock/pets (chicken or cat) did not have higher odds of contamination.
Only indicated covariates and significantly associated variables (p ≤ 0.2) in bivariate analysis were included in the multivariate logistic regression – i.e., location of the household, education of household head or their spouse, household water treatment, drinking water storage time, and FCR levels. The location of households denoted by village (AOR = 3.13, CI = 1.47-6.64), water treatment at the household level (AOR = 3.20, CI = 1.72-5.96), and FCR levels (AOR = 0.04, CI = 0.01-0.14) showed a significant association with microbial water quality at the household level (Table 4). Ordinal logistic regression was applied to both E. coli and total coliforms data and yielded similar results (S1 Table).
Discussion
Access to clean drinking water is important to promote health, especially in poor urban households. Household drinking water whether from piped or non-piped water systems, must not have any detectable microbial contamination. The presence of total coliforms in drinking water is indicative of environmental contamination, with E. coli being specific for fecal pollution which increases the risk of diarrhea, especially in children under 5 years of age. Approximately 56% and 15% households sampled had their drinking water contaminated with total coliforms and E. coli respectively, levels beyond the WHO guidelines of 0 c.f.u/100ml. The majority of samples met the guideline for E. coli (0 CFU/100 mL), suggesting generally low levels of fecal contamination. However, among the positive samples, contamination ranged from low to high-risk categories. In contrast, total coliforms were detected in more than half of the samples, with a substantial proportion falling within medium and high-risk categories. This reflects the broader environmental origin of total coliforms, beyond fecal contamination but still important in identifying contamination points in water distribution systems.
Our results show differential exposure to contamination based on location/water service line. As reported, households in Makina had higher odds of testing positive for E. coli or total coliforms than in Laini Saba. This is surprising since over 96% of households in these 3 villages use water from the same NCWSC source [27]. In addition, households do not always exclusively fetch/use water from one village, as choices may vary based on water availability, accessibility, and affordability in different water kiosks. In essence, these results may not sufficiently link microbial contamination at the household level to specific service lines or locations within the study site.
Although microbial contamination at the point of water collection (water kiosk) was not tested, the design of the distribution network reflects a scenario of homogeneity in water quality at collection points irrespective of service line or village. Bivariate analysis indicated that the reported source of drinking water, container diameter, or observed status of the storage container (open, partially closed or fully closed) did not influence household drinking water quality. Comparable results on water quality at the source were reported in Malawi, where a limited effect of water source on microbial quality was attributed to similar water quality between different water sources [34]. Just 6 months prior to the start of this study, a study implemented across 7 villages in Kibera [35] found that most water sources contained bacterial contamination (total coliforms). While these studies emphasize the validity of our findings, we cannot be certain about where in the service chain the contamination occurred, and therefore these results must be interpreted with caution. Important social, cultural, and economic factors beyond the scope of this study may explain these differences by location with implications for contextualized and targeted interventions. Furthermore, water quality deterioration from the point of collection to the point of use has been documented [36–38], and there is evidence that these changes occur at finer time scales [24].
At the household, drinking water can be exposed to fecal contamination through poor water handling practices and the use of uncovered storage containers. In our case, observed status of storage containers did not significantly influence microbial water quality. Households in informal settlements usually rely on water storage containers to ensure sustained water availability at the household even when water is unavailable at collection points [8]. Extended water storage time, which may accelerate chlorine decay, appeared to influence water quality beyond 48 hours. Longer water storage means multiple handling events with risk of repeat exposure during usage or refill, and probable recontamination [39–41]. However, storage time was not statistically significant when other factors were controlled for. This suggests that the effect of chlorination on microbial contamination is not likely to be lost with increased storage time, especially where households have constrained water storage capacity and experience increased water demands, such as in conditions commonly presented in urban slums.
Water treatment and FCR levels at the household level may have an overriding effect on handling practices, providing a quenching effect to contamination/recontamination events during storage as has been suggested in other studies [42]. Descriptive analysis showed that the majority of households stored water had FCR below the WHO-recommended levels of 0.2 mg/L which was too low to protect against further contamination. Since 94% of households used public tap or kiosk, this result is suggestive of sub-optimal dosing in pre-distribution treatment processes, or chlorine dilution/deterioration during storage and usage. These findings are consistent with previous studies that reported similar proportions of households with low FCR in Kisumu, Nairobi and Hawassa [43–45]. The high proportion of households positive with total coliforms or E. coli was partly due to low FCR levels making drinking water susceptible to contamination. These proportions are comparable to microbial water quality reported in similar settings where household water samples were positive for total coliforms and E. coli in Nairobi and in Hawassa [44,45]. While these variations in positivity rates exist between studies, households in slum settings like Kibera consistently experience high exposure to microbial contamination.
Improving water quality and minimizing health risks associated with poor water quality can be achieved by water treatment either through boiling, filtration, chemical disinfection (chlorination), or exposure to ultraviolet radiation before consumption [46]. In this study, household microbial water quality was significantly associated with water treatment practices at the household, FCR levels, and the location of the household. The effect of FCR suggests that due to continuous water use and handling at the household level, reliance on chlorine-based treatment methods would protect against further contamination events with increased storage time and usage [47–49]. Studies in other settings further support that water treatment methods such as boiling and filtration do not necessarily eliminate risks posed by inappropriate water storage and handling practices [50–52]. The Sawyer PointOne filter for example, is a common filtration method among households in Kibera, and its effectiveness has been reported in Kibera and elsewhere [18,53,54]. However, while these filters are effective in the short term, functionality concerns have been reported in long term use an aspect which may alter their effectiveness [55] thereby suggesting efficacy of other approaches such as chlorination [56–58]. While studies have often assessed water treatment methods as universally effective in different settings, the use of chlorine-based methods in slum settings may prove superior to filtration or boiling. Overall, while water treatment at the household marked a crucial practice, sustained protection against recontamination post-collection or treatment played an equally important role [47,49]. Households that rely on filtration or boiling to treat water may benefit from continuous education (on filter use and maintenance) and behavior change strategies on water handling post treatment. This is particularly important since households may not prefer to use chlorination due to acceptability thresholds on taste and odor [59,60].
Finally, the link between water quality and diarrhea is well established [45,61–63] and most WASH interventions target health outcomes especially among under five children. In this study, microbial water quality was not significantly associated with self-reported diarrhea. A plausible explanation is that diarrhea prevalence was reported for all household members and was not segregated by age, potentially leading to dilution of expected associations, especially among children under five. Reporting diarrhea for one’s self or other family members (especially other adults not in one’s care) is also a sub-optimal measure of disease [64].
Limitations
Our study was not without limitations. The sample size may have limited statistical power and therefore reported associations should be interpreted with caution. Furthermore, we recognize that rainfall events during the data collection period may have contributed to the observed contamination levels as shown in other studies [30,65]. This study did not incorporate rainfall data limiting the ability to assess the influence of rainfall on observed associations. Although field duplicates were analyzed alongside lab-prepared blanks, no field blanks were collected and processed alongside field samples. The use of E. coli and total coliforms as proxy for microbial quality excludes other pathogens like enteric viruses which may present varying associations with water treatment variables. Finally, predictor variables reported here do not fully incorporate the wider exposure risks among older children and adults who have less restricted mobility beyond the study villages. Moreover, there are many pathways to diarrhea (e.g., fingers, flies, fields, food and fluids) as has been reported elsewhere and contaminated water may only be one pathway [66].
Conclusion
Slum households continue to experience heightened exposure to microbial contamination of drinking water. This study shows that free chlorine residual at the household level is likely to play a dominant role in determining microbial water quality especially in slum settings where households exist in fecally-contaminated spaces. There is need for water utilities to identify strategies that increase likelihood of recommended microbial water quality at the point of use among slum households. This presents an opportunity for future studies to reassess the effectiveness of non-chlorine-based water treatment methods, and whether their use in real world conditions can offer effective and sustained protection against microbial contamination. To make progress in realizing SDG 6 targets for safely managed drinking water, the focus must shift from current household water treatment and storage (HWTS) practices towards provision of piped, adequately treated water on-premises to reduce recontamination risks.
Supporting information
S1 Text. Conceptual model of the water distribution.
https://doi.org/10.1371/journal.pwat.0000577.s001
(DOCX)
S2 Text. Survey, and water sampling and testing tools.
https://doi.org/10.1371/journal.pwat.0000577.s002
(DOCX)
S3 Text. Daily precipitation estimates during study period.
https://doi.org/10.1371/journal.pwat.0000577.s003
(DOCX)
S4 Text. Stata code used for data management, cleaning, and analysis.
https://doi.org/10.1371/journal.pwat.0000577.s004
(DO)
S1 Table. Ordinal logistic regression results.
https://doi.org/10.1371/journal.pwat.0000577.s006
(DOCX)
Acknowledgments
The project team would like to acknowledge and thank survey respondents and research assistants for their time, input, and involvement throughout the implementation of this study.
References
- 1. WHO. Drinking-water [Internet]. 2023 [cited 2024 Dec 4]. Available from: https://www.who.int/news-room/fact-sheets/detail/drinking-water
- 2.
KNBS and ICF. Kenya Demographic and Health Survey 2022. Key Indicators Report. Nairobi, Kenya, and Rockville, Maryland, USA: KNBS and ICF; 2023.
- 3. Dos Santos S, Adams EA, Neville G, Wada Y, de Sherbinin A, Mullin Bernhardt E, et al. Urban growth and water access in sub-Saharan Africa: Progress, challenges, and emerging research directions. Sci Total Environ. 2017;607–608:497–508. pmid:28704674
- 4. Kamau N, Njiru H. Water, Sanitation and Hygiene Situation in Kenya’s Urban Slums. J Health Care Poor Underserved. 2018;29(1):321–36. pmid:29503303
- 5. Blomkvist P, Nilsson D, Juma B, Sitoki L. Bridging the critical interface: Ambidextrous innovation for water provision in Nairobi’s informal settlements. Technology in Society. 2020;60:101221.
- 6. Mulwa F, Li Z, Fangninou F. Water Scarcity in Kenya: Current Status, Challenges and Future Solutions. OALib. 2021;8:1–15.
- 7. SHOFCO. SHOFCO. SHOFCO [Internet]. 2024 [cited 2024 Aug 11]. Available from: https://www.shofco.org/approach/wash/
- 8. Kumpel E, Nelson KL. Intermittent Water Supply: Prevalence, Practice, and Microbial Water Quality. 2016.
- 9. Gizachew M, Admasie A, Wegi C, Assefa E. Bacteriological Contamination of Drinking Water Supply from Protected Water Sources to Point of Use and Water Handling Practices among Beneficiary Households of Boloso Sore Woreda, Wolaita Zone, Ethiopia. Int J Microbiol. 2020;2020:5340202. pmid:32351573
- 10. Mahmud ZH, Islam MS, Imran KM, Hakim SAI, Worth M, Ahmed A, et al. Occurrence of Escherichia coli and faecal coliforms in drinking water at source and household point-of-use in Rohingya camps, Bangladesh. Gut Pathog. 2019;11:52. pmid:31695751
- 11. Schweitzer L, Noblet J. Water contamination and pollution. Green chemistry: an inclusive approach. 2018. p. 261–90.
- 12. Shields KF, Bain RES, Cronk R, Wright JA, Bartram J. Association of Supply Type with Fecal Contamination of Source Water and Household Stored Drinking Water in Developing Countries: A Bivariate Meta-analysis. Environ Health Perspect. 2015;123(12):1222–31. pmid:25956006
- 13. Worrell CM, Wiegand RE, Davis SM, Odero KO, Blackstock A, Cuéllar VM. A cross-sectional study of water, sanitation, and hygiene-related risk factors for soil-transmitted helminth infection in urban school- and preschool-aged children in Kibera, Nairobi. PLoS One. 2016;11(3):e0150744. pmid:26950552
- 14. Ahmed J, Wong LP, Chua YP, Channa N, Mahar RB, Yasmin A, et al. Quantitative Microbial Risk Assessment of Drinking Water Quality to Predict the Risk of Waterborne Diseases in Primary-School Children. Int J Environ Res Public Health. 2020;17(8):2774. pmid:32316585
- 15. Shayo GM, Elimbinzi E, Shao GN, Fabian C. Severity of waterborne diseases in developing countries and the effectiveness of ceramic filters for improving water quality. Bull Natl Res Cent. 2023;47(1).
- 16. Sila ON. Physico-chemical and bacteriological quality of water sources in rural settings, a case study of Kenya, Africa. Scientific African. 2019;2:e00018.
- 17. Kim DD, Swarthout JM, Worby CJ, Chieng B, Mboya J, Earl AM, et al. Contaminated drinking water facilitates Escherichia coli strain-sharing within households in urban informal settlements. Nat Microbiol. 2025;10(5):1198–209. pmid:40312516
- 18. Berihun G, Abebe M, Hassen S, Gizeyatu A, Berhanu L, Teshome D, et al. Drinking water contamination potential and associated factors among households with under-five children in rural areas of Dessie Zuria District, Northeast Ethiopia. Front Public Health. 2023;11:1199314. pmid:37361152
- 19. Ondieki JK, Akunga DN, Warutere PN, Kenyanya O. Bacteriological and physico-chemical quality of household drinking water in Kisii Town, Kisii County, Kenya. Heliyon. 2021;7(5). pmid:34007931
- 20. Sobsey MD, Handzel T, Venczel L. Chlorination and safe storage of household drinking water in developing countries to reduce waterborne disease. Water Sci Technol. 2003;47(3):221–8. pmid:12639033
- 21. Mellor JE, Smith JA, Samie A, Dillingham RA. Coliform Sources and Mechanisms for Regrowth in Household Drinking Water in Limpopo, South Africa. J Environ Eng (New York). 2013;139(9):1152–61. pmid:25190902
- 22. Meierhofer R, Wietlisbach B, Matiko C. Influence of container cleanliness, container disinfection with chlorine, and container handling on recontamination of water collected from a water kiosk in a Kenyan slum. J Water Health. 2019;17(2):308–17. pmid:30942780
- 23. Bain R, Cronk R, Wright J, Yang H, Slaymaker T, Bartram J. Fecal contamination of drinking-water in low- and middle-income countries: a systematic review and meta-analysis. PLOS Medicine. 2014;11(5):e1001644.
- 24. Price HD, Adams EA, Nkwanda PD, Mkandawire TW, Quilliam RS. Daily changes in household water access and quality in urban slums undermine global safe water monitoring programmes. Int J Hyg Environ Health. 2021;231:113632. pmid:33202361
- 25. Desgroppes A, Taupin S. Kibera: The biggest slum in Africa?. Les Cahiers d’Afrique de l’Est / The East African Review. 2011;44:44.
- 26. Shimamura Y, Yang J, Baffoe G, Dodoo J, Mwaketsi I, Matsuda H. Rethinking the Kibera ‘Slum’: The Role of Social Networks as a Mechanism for Coping with the Increased Demand for Job Opportunities. J Econ Sustain Dev. 2017;8(4):157–69.
- 27.
UN Habitat. Informal settlements ’ vulnerability mapping in Kenya The Case of Kibera. 2020. p. 1–22.
- 28. Kim J, Hagen E, Muindi Z, Mbonglou G, Laituri M. An examination of water, sanitation, and hygiene (WASH) accessibility and opportunity in urban informal settlements during the COVID-19 pandemic: Evidence from Nairobi, Kenya. Sci Total Environ. 2022;823:153398. pmid:35092785
- 29. EPA. Method 1604: total coliforms and Escherichia coli in water by membrane filtration using a simultaneous detection technique (MI Medium) [Internet]. United States, Environmental Protection Agency, Office of Water, [2002]. 2002. Available from: http://www.epa.gov/safewater/methods/met1604.pdf
- 30. Kostyla C, Bain R, Cronk R, Bartram J. Seasonal variation of fecal contamination in drinking water sources in developing countries: a systematic review. Sci Total Environ. 2015;514:333–43. pmid:25676921
- 31. Climate Hazards Center Infrared Precipitation with Stations Version 3 (CHIRPS3) Data Repository. 2025.
- 32. Funk C, Peterson P, Harrison L, Saldivar R, Landsfeld M, Pedreros D, et al. The Climate Hazards Center Infrared Precipitation with Stations, Version 3. Sci Data. 2026;13(1):718. pmid:41965329
- 33. Tazin S, Islam M, Pickering AJ, Kwong LH, Mertens A, Niven C, et al. Associations between floor material and E. coli contamination in rural Bangladeshi households. Int J Hyg Environ Health. 2025;269:114641. pmid:40738050
- 34. Price HD, Adams EA, Nkwanda PD, Mkandawire TW, Quilliam RS. Daily changes in household water access and quality in urban slums undermine global safe water monitoring programmes. Int J Hyg Environ Health. 2021;231:113632. pmid:33202361
- 35. Tintle NL, Westra J, Griend KVD, Beard V, Turner BN, Huisman NLH, et al. Improvements in diarrheal disease prevalence with point-of-use water filter implementation in the informal settlement of Kibera, Kenya. medRxiv. 2023. 2023.03.03.23286740.
- 36. Daniel D, Diener A, van de Vossenberg J, Bhatta M, Marks SJ. Assessing Drinking Water Quality at the Point of Collection and within Household Storage Containers in the Hilly Rural Areas of Mid and Far-Western Nepal. Int J Environ Res Public Health. 2020;17(7):2172. pmid:32218157
- 37. Cassivi A, Tilley E, Waygood EOD, Dorea C. Household practices in accessing drinking water and post collection contamination: A seasonal cohort study in Malawi. Water Res. 2021;189:116607. pmid:33197683
- 38. Chemuliti JK, Gathura PB, Kyule MM, Njeruh FM. Bacteriological qualities of indoor and out-door drinking water in Kibera sub-location of Nairobi, Kenya. East Afr Med J. 2002;79(5):271–3. pmid:12638814
- 39. Bigohe GN, Tshimanga RM, Bigohe J-BN, Makaly EB, Bola GB, Managa M, et al. Post-supply microbial deterioration of household-stored drinking water in Kinshasa, Democratic Republic of the Congo. Discov Water. 2025;6(1).
- 40. Judah LA, Andriambololonirina C, Rakotoarisoa L, Barrett LJP, Khaliq M, Mihelcic JR. Occurrence and Mitigation of Bacterial Regrowth in Stored Household Water in Eastern Coastal Madagascar. Water. 2024;16(11):1592.
- 41. Levy K, Nelson KL, Hubbard A, Eisenberg JNS. Following the water: a controlled study of drinking water storage in northern coastal Ecuador. Environ Health Perspect. 2008;116(11):1533–40. pmid:19057707
- 42. Daniel D, Diener A, van de Vossenberg J, Bhatta M, Marks SJ. Assessing Drinking Water Quality at the Point of Collection and within Household Storage Containers in the Hilly Rural Areas of Mid and Far-Western Nepal. Int J Environ Res Public Health. 2020;17(7):2172. pmid:32218157
- 43.
Alam M, Shoaib D, Hasan M, Tabassum KF, Wilbur J, Tidwell B, et al. Evaluation of Hygiene and Behaviour Change Coalition for COVID-19 Prevention: Inclusion of Disability and Ageing. UK: London School of Hygiene and Tropical Medicine; 2023.
- 44. Blanton E, Wilhelm N, O’Reilly C, Muhonja E, Karoki S, Ope M, et al. A rapid assessment of drinking water quality in informal settlements after a cholera outbreak in Nairobi, Kenya. J Water Health. 2015;13(3):714–25. pmid:26322757
- 45. Bekele RS, Teka MA. Physicochemical and microbial quality of drinking water in slum households of Hawassa City, Ethiopia. Appl Water Sci. 2022;13(1).
- 46. Agensi A, Tibyangye J, Tamale A, Agwu E, Amongi C. Contamination Potentials of Household Water Handling and Storage Practices in Kirundo Subcounty, Kisoro District, Uganda. J Environ Public Health. 2019;2019:7932193. pmid:30944573
- 47. Ali SI, Ali SS, Fesselet J-F. Evidence-based chlorination targets for household water safety in humanitarian settings: Recommendations from a multi-site study in refugee camps in South Sudan, Jordan, and Rwanda. Water Res. 2021;189:116642. pmid:33246215
- 48. Nielsen AM, Garcia LAT, Silva KJS, Sabogal-Paz LP, Hincapié MM, Montoya LJ, et al. Chlorination for low-cost household water disinfection - A critical review and status in three Latin American countries. Int J Hyg Environ Health. 2022;244:114004. pmid:35816847
- 49. Chlorination and safe storage of household drinking water in developing countries to reduce waterborne disease. Water Sci Technol. 2003.
- 50. Ghaudenson R, Priadi CR, Foster T. Effectiveness of Groundwater Boiling as Household Water Treatment in Metro and Bekasi Cities, Indonesia. E3S Web Conf. 2021;277:04002.
- 51. Gupta SK, Suantio A, Gray A, Widyastuti E, Jain N, Rolos R, et al. Factors associated with E. coli contamination of household drinking water among tsunami and earthquake survivors, Indonesia. Am J Trop Med Hyg. 2007;76(6):1158–62. pmid:17556629
- 52. Lin W, Ye C, Guo L, Hu D, Yu X. Analysis of microbial contamination of household water purifiers. Appl Microbiol Biotechnol. 2020;104(10):4533–45. pmid:32193577
- 53. Sawyer TBM Kibera Final Report December 2022 [Internet]. [cited 2025 Oct 4]. Available from: https://www.sawyer.com/resources/sawyer-tbm-kibera-final-report-december-2022
- 54. Alford KR, Walls LK, DeRuiter SL, Pruim R, VanHorn JE, Bone M, et al. Reduction in diarrhoea rates after household water filter distribution in small and remote communities in Liberia. Cogent Public Health. 2023;10(1).
- 55. Alford KR, Rosendale KD, Koroma AH, DeRuiter SL, Pruim R, VanHorn JE, et al. Evaluating filter functionality and user competence after a hollow fiber membrane filter intervention in Liberia. J Water Sanit Hyg Dev. 2022;12(12):851–61.
- 56. Effectiveness of Groundwater Boiling as Household Water Treatment in Metro and Bekasi Cities, Indonesia. ResearchGate.
- 57. Factors associated with E. coli contamination of household drinking water among tsunami and earthquake survivors, Indonesia. Am J Trop Med Hyg.
- 58. Lin W, Ye C, Guo L, Hu D, Yu X. Analysis of microbial contamination of household water purifiers. Appl Microbiol Biotechnol. 2020;104(10):4533–45. pmid:32193577
- 59. Crider Y, Sultana S, Unicomb L, Davis J, Luby SP, Pickering AJ. Can you taste it? Taste detection and acceptability thresholds for chlorine residual in drinking water in Dhaka, Bangladesh. Sci Total Environ. 2018;613–614:840–6. pmid:28942317
- 60. Heylen C, String G, Naliyongo D, Ali SI, Brown J, Ogira V, et al. Evaluating chlorine taste and odor acceptability to inform drinking water chlorination in humanitarian settings: Lessons from trials in Uganda. PLOS Water. 2025;4(2):e0000267.
- 61. Eugène BL, Jean PLK, Joris LL. Effectiveness of home water treatment for the control of diarrhea in children under five years: an intervention study in a rural health zone of Tshopo in the Democratic Republic of Congo. J Water Sanit Hyg Dev. 2024;15(1):61–74.
- 62. Merid MW, Alem AZ, Chilot D, Belay DG, Kibret AA, Asratie MH, et al. Impact of access to improved water and sanitation on diarrhea reduction among rural under-five children in low and middle-income countries: a propensity score matched analysis. Trop Med Health. 2023;51(1):36. pmid:37322559
- 63. Kirby MA, Nagel CL, Rosa G, Umupfasoni MM, Iyakaremye L, Thomas EA, et al. Use, microbiological effectiveness and health impact of a household water filter intervention in rural Rwanda-A matched cohort study. Int J Hyg Environ Health. 2017;220(6):1020–9. pmid:28596086
- 64. Watson SI, Rego RTT, Hofer T, Lilford RJ. Evaluations of water, sanitation and hygiene interventions should not use diarrhoea as (primary) outcome. BMJ Glob Health. 2022;7(5). pmid:10.1136/bmjgh-2022-008521
- 65. Powers JE, Mureithi M, Mboya J, Campolo J, Swarthout JM, Pajka J, et al. Effects of High Temperature and Heavy Precipitation on Drinking Water Quality and Child Hand Contamination Levels in Rural Kenya. Environ Sci Technol. 2023;57(17):6975–88. pmid:37071701
- 66. Luby SP, Halder AK, Huda TM, Unicomb L, Islam MS, Arnold BF. Microbiological contamination of drinking water associated with subsequent child diarrhea. 2015.