key: cord-0064320-e2vilql5
authors: Nguyen, Thi Yen Chi; Fagbayigbo, Bamidele Oladapo; Cissé, Guéladio; Redi, Nesre; Fuhrimann, Samuel; Okedi, John; Schindler, Christian; Röösli, Martin; Armitage, Neil Philip; Carden, Kirsty; Dalvie, Mohamed Aqiel
title: Diarrhoea among Children Aged under Five Years and Risk Factors in Informal Settlements: A Cross-Sectional Study in Cape Town, South Africa
date: 2021-06-04
journal: Int J Environ Res Public Health
DOI: 10.3390/ijerph18116043
sha: d9e637b0ebbb91dad515f3e60f191e73c53177dc
doc_id: 64320
cord_uid: e2vilql5

Background: There is limited data on the association between diarrhoea among children aged under five years (U5D) and water use, sanitation, hygiene, and socio-economics factors in low-income communities. The study investigated U5D and the associated risk factors in the Zeekoe catchment in Cape Town, South Africa. Methods: A cross-sectional study was conducted in 707 households in six informal settlements (IS) two formal settlements (FS) (March–June 2017). Results: Most IS households used public taps (74.4%) and shared toilets (93.0%), while FS households used piped water on premises (89.6%) and private toilets (98.3%). IS respondents had higher average hand-washing scores than those of FS (0.04 vs. −0.14, p = 0.02). The overall U5D prevalence was 15.3% (range: 8.6%–24.2%) and was higher in FS than in IS (21.2% vs. 13.4%, respectively, p = 0.01). Water storage >12 h was associated with increasing U5D (OR = 1.88, 95% CI 1.00–3.55, p = 0.05). Water treatment (OR = 0.57, 95%CI 0.34–0.97, p = 0.04), good hand-washing practices (OR = 0.59, 95%CI 0.42–0.82, p = 0.002) and Hepatitis A vaccination (OR = 0.51, 95%CI 0.28–0.9, p = 0.02) had significant preventing effects on U5D. Conclusions: The study highlights that good hygiene practice is a key intervention against U5D in informal settlements. The promotion of hand-washing, proper water storage, and hygienic breastfeeding is highly recommended.

In 2017, diarrhoea was among the leading cause of death among all ages globally (more than 1.7 million deaths) and the fifth leading cause of death among children younger than five years with more than 0.44 million deaths) [1] . According to the World Health Organization (WHO) 2017 report, more than a quarter (26.9%) of diarrhoea deaths occurred among children younger than five years, and about 90% of diarrhoea deaths occurred in South Asia and sub-Saharan Africa [2] . Statistics South Africa (Stats SA) estimates that 20% of under-five deaths may be attributed to diarrhoea [3] , with 31,436 diarrhoea cases recorded in 2016 [4] , while other sources estimate the mortality rate between 8-24% [5, 6] .

There is an unequal geographical distribution of diarrhoea burden among children younger than five years of age across Africa [6] . Children in low-and middle-income

The study was conducted in six informal and two formal settlements of Cape Town, located in the Zeekoe Catchment in the Cape Flats (a low-lying area of Cape Town). The map of the catchment study area is represented in Figure 1a -c. The catchment area overlays the Cape Flats Aquifer (CFA), a potential groundwater resource, and is marked by various land uses, precarious socio-economic conditions, black and grey water flows from settlements and industries, and a wastewater treatment plant (WWTP) that discharges treated effluent to the sea (Figure 1 ). 

The Zeekoe Catchment area consists of a wide range of communities, with many formal settlements blended into a poor informal settlement and sharing the same polluted premises [19] . Six informal settlements in the catchment selected in a previous phase of the study were also selected for this study, namely: Barcelona, Lotus Park, Phola Park, Sweet Home, Weltevrede, and Pelican Park (Figure 1c ). In addition, we selected nearby formal houses from Gugulethu and Manenberg. The selection of these sites was based on 

The Zeekoe Catchment area consists of a wide range of communities, with many formal settlements blended into a poor informal settlement and sharing the same polluted premises [19] . Six informal settlements in the catchment selected in a previous phase of the study were also selected for this study, namely: Barcelona, Lotus Park, Phola Park, Sweet Home, Weltevrede, and Pelican Park (Figure 1c ). In addition, we selected nearby formal houses from Gugulethu and Manenberg. The selection of these sites was based on the following criteria established with the civil society actors working in the areas: (i) Accessibility of the area; (ii) Interviewer's availability for the survey in the settlement; and (iii) Security issues. Of note, all the settlements from the catchment considered for inclusion in the study were high-risk crime areas, and thus, it was important to minimize the threat to the safety of the field staff. Areas were visited if site facilitators were available to accompany field staff. Houses in the selected areas were selected opportunistically.

The sample size was chosen for use in a Monte Carlo simulation. A sample size of approximately 90 households per each settlement was selected, assuming that the assumed prevalence of childhood diarrhoea in study sites was 40% with a standard deviation of 5%, and the ratio between exposed and non-exposed groups was 1:1. 'Diarrhoea' was defined as "having three or more loose or liquid bowel movements over a 24-h period, as reported by the mother or caregiver of the child, at any point during the seven days preceding the interview" [7] .

Data were collected from March to June 2017. The questionnaire was developed in English and translated into Xhosa and Afrikaans for appropriate communication within the local community, given that many residents are not fluent in the English language. A pilot study was carried out three days before the beginning of the household survey to identify possible shortcomings of the questionnaire and to verify the anticipated duration of the interview. The questionnaire covered the following sections: (i) basic demographic information, (ii) socio-economic status, (iii) water source, (iv) sanitation and hygiene, and (v) child health status. The survey questions were addressed to the head of the household or its legal representative at the time of the interview. The questionnaires were administered by six groups of trained interviewers consisting of two persons per group. The anonymity of the respondents was preserved throughout the course of the research.

Statistical analyses were carried out using Stata14 software (StataCorp LP, Lakeway, TX, USA). Using factor analysis, a wealth index was created based on possessions of household items, and a hand-washing index was created based on hand-washing. All variables were screened for irregularities and outliers. Basic descriptive statistical analyses, such as median, mean, interquartile ranges, standard deviation (SD), and confidence interval (CI), were obtained for the relevant continuous variables, and a frequency table was generated for other variables.

Due to the hierarchical structure of collected data where child's data being nested in the household and household data in communities, multi-level logistic modelling was performed [20] . Mixed-effects logistic regression modelling was used to examine the effects of child-related variables (level 1) and household-related variables (level 2) on childhood diarrhoea.

A total of 707 households participated in the survey conducted in six informal settlements and two nearby formal settlements. There was a total of 778 children aged 0-5 years old in the recruited households. The demographic and health-related information of those children stratified by living location is shown in Table 1 . The mean age of the children was 28.8 months (SD = 17), and 51.9% were female. While the gender of children was distributed equally in the IS, more females than males participated in FS (58.2% vs. 41.8%, respectively). The vast majority of caretaker respondents (>90%) were female. The median number of people per household in FS was 5 (SD: 2.2, range 1-18). It was also observed that residents of FS stay longer in their current address relative to those from the IS with an average duration of (23 years vs. 8 years, p = 0.005). Although FS families had a statistically significantly higher mean wealth index than those living in IS (p = 0.05), there was not much difference in the proportion of children going to daycare (47.1% vs. 47.3%, p = 0.99). 

The majority of children in participating households had access to potable water (>97%) for domestic use. In IS, 74.4% of households used public taps as the main source of drinking water (Figure 2a ). All residences of IS had access to water predominantly from a public tap, with the exception of Pelican Park that had piped tap water inside the yard as the main water resource (Figure 2b ). In contrast, piped tap-water on the premises (89.6%) was the most common water resource in formal areas (Figure 2c ). Most IS residents reported storing water either in plastic bottles or containers (96.3%), with the water storage duration exceeding 12 h in 13.1% of households. FS residents were more likely to store water <12 h than those in IS (91.3% vs 76.2%, Chi-square χ 2 =26.9, p < 0.001). Only a small proportion (0.6%) of the caretakers, all from IS, reported that they need over 30 min to get to the water source. A general overview of access to water resources of participating households during the study is represented in Table 2 . Most IS residents reported storing water either in plastic bottles or containers (96.3%), with the water storage duration exceeding 12 h in 13.1% of households. FS residents were more likely to store water <12 h than those in IS (91.3% vs. 76.2%, Chi-square χ 2 =26.9, p < 0.001). Only a small proportion (0.6%) of the caretakers, all from IS, reported that they need over 30 min to get to the water source. A general overview of access to water resources of participating households during the study is represented in Table 2 . 

Most households (98.3%) reported access to latrines, but only 150 households (21.2%) had access to a flush toilet system. The proportion of IS households having access to an improved sanitation facility (modern toilet, potable water, waste disposal) was considerably lower than those of FS households (5.0% vs. 75.0%, Chi-square test, p < 0.001). Shared facilities were the most common in all IS (93%), while a private toilet was the most common among FS households (75%) (Figure 3a) . The mean number of households that shared a sanitary facility in most informal households was 4 (range 1-11). Most of the sanitary facilities were located less than 50 m from the houses (94.8%). While flush toilets were the most frequently reported facility in formal areas (>96%) (Figure 3c ), facility types varied across informal sites, with the flush toilet the most common in Pelican Park (96.6%) and Phola Park (97.7%); the portable toilet system is most common in Barcelona (71%) and Shengu (chemical toilets) and flush toilets in Lotus Park, Weltevrede, and Sweet Home (Figure 3b ). Most households (98.3%) reported access to latrines, but only 150 households (21.2%) had access to a flush toilet system. The proportion of IS households having access to an improved sanitation facility (modern toilet, potable water, waste disposal) was considerably lower than those of FS households (5.0% vs. 75.0%, Chi-square test, p < 0.001). Shared facilities were the most common in all IS (93%), while a private toilet was the most common among FS households (75%) (Figure 3a) . The mean number of households that shared a sanitary facility in most informal households was 4 (range 1-11). Most of the sanitary facilities were located less than 50 m from the houses (94.8%). While flush toilets were the most frequently reported facility in formal areas (>96%) (Figure 3c ), facility types varied across informal sites, with the flush toilet the most common in Pelican Park (96.6%) and Phola Park (97.7%); the portable toilet system is most common in Barcelona (71%) and Shengu (chemical toilets) and flush toilets in Lotus Park, Weltevrede, and Sweet Home (Figure 3b ). 

With regard to hygiene, the participants were asked, "When do you usually wash your hands?" with options: before eating, after eating, after defecation, after latrine use, before feeding child, after handing rubbish, after handling child's diaper/feces, before food preparation, after touching animals. The questionnaire also included the hygiene question, "Have you seen any rats/cockroaches around your house in the last 7 days." Additionally, interviewers recorded the availability of soap in the house using an observation checklist.

The 

With regard to hygiene, the participants were asked, "When do you usually wash your hands?" with options: before eating, after eating, after defecation, after latrine use, before feeding child, after handing rubbish, after handling child's diaper/feces, before food preparation, after touching animals. The questionnaire also included the hygiene question, "Have you seen any rats/cockroaches around your house in the last 7 days." Additionally, interviewers recorded the availability of soap in the house using an observation checklist.

The 34 .2% FS), before food preparation (58.9% IS vs. 64.6% FS), and after touching animals (12.5% IS vs. 12.2% FS). The results of an analysis using a hand-washing index derived from the above indicators using principal component analysis showed that respondents from IS had better hand-washing practices than FS (t-test, p = 0.02). The availability of soap at hand-washing places reported in both formal and informal settlements are similar (28.5% vs. 26.2%) and not statistically significantly different. With regard to wastewater disposal, most IS respondents reported that wastewater was disposed of onto open ground (63.8%), while most FS respondents reported that they disposed wastewater into the toilet (51.2%) and onto open ground far away from their houses (18.3%). Pest (rats, cockroaches) invasion 2 weeks prior to the survey occurred more frequently in IS compared to FS (87.0% vs. 46.0%, p = 0.00); however, using pesticides inside the house during 2 weeks prior to the survey was less frequently reported in IS than FS households (44.0% vs. 57.0%, p = 0.004). Hygiene practices of respondents from informal and formal households are presented in Table 3 .

The health-related indicators for participating children include vaccines monitored by the South Africa Expanded Programme on Immunisation (EPI-SA). Children from FS were more frequently reported to be vaccinated than IS children: 74.1% vs. 50.9% for Hep B vaccination and 76.2% vs. 64.2% for Rotavirus, respectively (Chi-square test, p < 0.001). Interestingly, for HepA vaccination, which is not sponsored under EPI-SA, children living in informal households were reported to get injected more frequently than those living in formal households (45.0% vs. 33.9%, Chi-square test, p < 0.001). Overall, 12.3% of all children were breastfed exclusively for 6 months after birth. More than twothirds of studied children (76.4%) reportedly had no breastfeeding at all. There was no statistically significant difference in breastfeeding rates between FS and IS children (21.1% vs. 24.5%, Chi-square test, p = 0.33). 

The prevalence of diarrhoea among children under five years old was calculated based on the total number of cases that reported having diarrhoea during the 7 days prior to the interview in all the participating households. The distributions of diarrhoea cases by study site, age and gender distribution are illustrated in Figure 4 .

to the interview in all the participating households. The distributions of diarrhoea cases by study site, age and gender distribution are illustrated in Figure 4 .

Among a total of 778 recruited children, diarrhoea occurrence was recorded in 119 children (15.3%). The period prevalence was highest in Gugulethu (24.2%), followed by Pelican Park (19.4%), Manenberg (17.8%), Sweet Home (15%), Barcelona (14.4%), Lotus Park (11.1%), Phola Park (11.1%), and Weltevrede (8.6%) (Figure 4b) . The prevalence of diarrhoea cases in this study was higher in formal settlements compared to informal settings (21.2% vs. 13.4% respectively, Chi-square test, χ 2 = 6.6, p = 0.01) (Figure 4a ). As seen in Figure 4c , the age group of 6-11 months had the highest prevalence of diarrhoea (26%), following by the group of 24-35 months (18%); conversely, the diarrhoea prevalence in other age groups was lower, ranging from 12%-13%. Diarrhoea prevalence was higher among girls than boys in all age groups, except in children aged 12-23 months. However, the gender difference was not statistically significant (Chi-square test, χ 2 = 1.5, p = 0.22) (Figure 4d ). Among a total of 778 recruited children, diarrhoea occurrence was recorded in 119 children (15.3%). The period prevalence was highest in Gugulethu (24.2%), followed by Pelican Park (19.4%), Manenberg (17.8%), Sweet Home (15%), Barcelona (14.4%), Lotus Park (11.1%), Phola Park (11.1%), and Weltevrede (8.6%) (Figure 4b ). The prevalence of diarrhoea cases in this study was higher in formal settlements compared to informal settings (21.2% vs. 13.4% respectively, Chi-square test, χ 2 = 6.6, p = 0.01) (Figure 4a ). As seen in Figure 4c , the age group of 6-11 months had the highest prevalence of diarrhoea (26%), following by the group of 24-35 months (18%); conversely, the diarrhoea prevalence in other age groups was lower, ranging from 12%-13%. Diarrhoea prevalence was higher among girls than boys in all age groups, except in children aged 12-23 months. However, the gender difference was not statistically significant (Chi-square test, χ 2 = 1.5, p = 0.22) (Figure 4d ).

The results from the multivariate analysis show that the caregiver's education level was positively associated with the diarrhoea occurrence in children under five years old (OR =1.59, 95% CI 1.06-2.40, p = 0.03). There was no significant association between eco-nomic status, based on the wealth index and under-five-year child diarrhoea (OR = 0.67, 95% CI 0.38-1.17, p = 0.16). Usage of water stored >12 h showed a statistically significant positive association with diarrhoea (OR = 1.90, 95%CI 1.02-3.79, p = 0.05). Children from households who treat water before drinking were less likely to experience diarrhoea compared to children from households who do not treat water (OR = 0.57, 95%CI 0.34-0.97, p = 0.04). None of the drinking water sources were statistically significantly associated with diarrhoea. Interruption of water supply was not found to be statistically significantly associated (OR = 1.21, 95%CI 0.69-2.09, p = 0.5) with diarrhoea. Additionally, sanitation characteristics and ownership status, when summarized as a continuous variable, showed a significant association with under-five diarrhoea (OR = 0.47. 95% CI 0.26-0.84, p = 0.12). Interestingly, a higher number of households sharing the same facility was associated with a lower risk of diarrhoea among study participants. There was a significantly lower risk of diarrhoea among households sharing the facility with more than three other households compared to those using non-shared facilities (OR = 0.35, 95%CI 0.15-0.85, p =0.02). Dysfunctional toilet facilities showed a significant positive association with childhood diarrhoea (OR = 1.88, 95%CI 1.00-3.55, p = 0.05). An increased hand-washing index was found to be strongly associated with a lower risk of diarrhoea (OR = 0.61, 95%CI 0.47-0.79, p < 0.001). There was no significant association between under-five diarrhoea, gender, and age. The duration of breastfeeding showed a weak and statistically non-significant positive association with the risk of diarrhoea (OR = 1.11, 95%CI 1.00-1.23, p = 0.07). The results in Table 4 shows Hepatitis A vaccination to be a statistically significant protective factor (OR = 0.51, 95%CI 0.28-0.90, p = 0.02) while rotavirus vaccination was not statistically significantly associated with under-five diarrhoea (OR = 1.62, 95%CI 0.86-3.08, p = 0.14). Results of multi-level multivariate analyses of risk factors associated with diarrhoea are presented, including the following groups of variables: socio-demographic (caretaker education level, having other under-5 children in the household and wealth index), water-related variables (drinking water sources, water storage >12 h, water treatment options and water interruption during within the period of seven 7 days prior the survey), sanitation and hygiene-related variables (facility access and sharing, reported a problem with toilet facility, disposal of child's feces, hand-washing index), and children-specific variables (child's age, gender, breastfeeding duration and immunization history of Hepatitis A and Rotavirus vaccine).

This study revealed that socio-economic factors were important predictors of underfive diarrhoea. Children living in the formal settlements showed greater evidence of access to socio-economic facilities (accommodation, education and health) relative to those living in the informal settlements. It is important to note, however, that living in a formal settlement does not necessarily suggest better hygiene practice. People living in formal settlements have a surprisingly lower index of hand-washing compared to people living in informal settlements. The combination of poor living conditions, poor hygiene practices and contaminated environment in the neighbourhood contributed to diarrhoea cases in children under five in both formal and informal households. The literature suggests that the caregiver's increasing level of education has been considered as a protective factor against diarrhoea [21] [22] [23] . However, the results in this study show that the higher education levels of caregivers are associated with a consistently increased risk of diarrhoea in children under five. The paradox has been found in some other studies reporting that there were no differences in knowledge of adequate dietary practices between different maternal education levels [24, 25] . In addition, higher education could lead to earlier weaning which can increase episodes of diarrhoea in children [26] .

Water access and resources is a crucial factor that serves as an indicator of community health. The majority of the investigated households in the formal settlements have access to the municipal water supply. Various factors relating to water access do not allow for any concrete conclusions about the safety of the system delivering water in informal households in this study. Inhabitants in informal settlements are not allowed to have private water connections to their homes; however, some make illegal connections without any supervision or instruction from local authorities, thus creating a path for potential pollution. There is seldom any system for draining the grey water generated by the in-house supply from the informal settlement [27] .

Our findings show that sharing sanitation facilities with others was associated with an increased risk of diarrhoea relative to non-sharing facilities. Contamination of drinking water resources was higher in households with poor sanitation compared to those with adequate sanitation, and this may explain the increased risk of diarrhoea in the former. However, there are different types of shared sanitation services in informal settlements, varying from a self-maintained facility that is shared between a few nearby households to publicly run facilities (some might be maintained better than others). Although our data suggested that shared sanitation does not pose higher risks than private access, other factors related to sanitation could be more important in diarrhoea transmission. For instance, IS are constrained both with respect to space and water supply, and the majority of communal sanitation facilities are not linked to a sewage system. Stagnant water is often observed in such dense informal communities, which poses a big risk of spreading diarrhoea pathogens into the environment. Previous studies demonstrated children would face a high risk of diarrhoea from playing in soil and surface water outside the household [28, 29] . Reinforcing the findings from other studies, we also found that dysfunctional toilet facilities to be significantly associated with childhood diarrhoea in this study. This study found that hygiene practices were associated with a reduced risk of childhood diarrhoea, which is consistent with the literature [30, 31] .

It is well-known that good hygiene could enhance healthy living, and treating pointuse water reduces the risk of getting diarrhoea disease [30, 32] . Our findings show that children from households with good hygiene are less likely to experience diarrhoea than those who did not. This study is consistent with similar studies elsewhere [33] . Hygiene practices, such as water purification methods, including water-boiling and filtration, could reduce the pathogen load in water if the process is carried out consistently and correctly according to standard techniques [30, 34] . Another study also suggested that not all water treatment methods are effective if carried out in an unhygienic manner; improper infiltration could cause by unsafe storage water leading to recontamination after treating [35] . Our data shows that drinking water stored longer than 12 h could contribute to the potential risk of childhood diarrhoea. Similar results were found in other studies conducted in developing countries [36] [37] [38] . Our findings suggested that sharing a facility with more than four households significantly lower the risk of diarrhoea compared to those using a non-shared facility. Similar studies conducted in Tanzania showed that shared latrines contained lower concentrations of Escherichia coli compared to private latrines [39] . Both results are unexpected. Other studies and meta-analyses found that shared sanitation is generally dirtier and associated with an increased risk of diarrhoea compared to private sanitation [40] [41] [42] [43] , suggesting the need for further investigation.

In this study, the highest diarrhoea prevalence occurred in children aged 06-11 months. This finding is consistent with that of previous research that suggests diarrhoea is more likely to happen to younger children under 12 months, and the risk reduces with increasing age [44] [45] [46] . Ordinarily, breastfeeding should increase the children's immune system and thus protect them against diarrhoea, but this was not found to be the case for those aged under five in this study. This could be due to exposure to contaminated water lying around the informal settlements, increased exposure of the toddler to contaminated water and/or lack of personal hygiene. Additionally, practicing breastfeeding may contribute to childhood diarrhoea if the nursing mothers engage in unhygienic practices such as dirty breasts. Findings also revealed that children who took the HepA vaccine are at lower risk than those who were not HepA vaccinated. The prevalence of Hep A infection in the study is likely due to exposure to poor water and sanitation conditions. Serological studies detected HepA in fecal samples of children with acute diarrhoea in developing countries [47] [48] [49] . Furthermore, a number of studies have reported the detection or isolation of Hep A virus from water resources, with the detection rate ranging from 76% in surface water to 37% in wastewater plants and irrigation water [50, 51] . However, there was no available data on the Hep A prevalence in children under five years in South Africa, and the role of Hep A virus as an element in diarrhoea has not been documented. The results from this study stress the need for further investigation with routine tracking of Hep A infection as a contributory cause of childhood diarrhoea in sub-urban areas in Cape Town.

The strengths of this study include the opportunity to estimate the effects of multiple environmental factors and diarrhoea burden using individual-level data from a communitybased survey and using multi-level analyses to consider the effects of individual-level and household-level factors in one analysis. Among the limitations of this study, we did not explore the comparison of the ways sanitation facilities are shared in informal settlements with private access facilities. Most households in informal settlements had access to only shared sanitation facilities. There are limited data on how shared sanitation facilities in informal settlements were managed and maintained. In addition, documentation of how stagnant water and leakages were contained and managed was not explored by our study. Those factors may be important sanitation-related confounders on the effects of sanitation on childhood diarrhoea. We were unable to check our data on diarrhoea prevalence against that collected by the local health authorities owing to the difficulty in obtaining the necessary approvals within the limited time frame of the study. A standard limitation is that diarrhoea was reported by respondents with no clinical confirmation. This method is widely used, although clinical confirmations make the disease classification more reliable. The study also did not track the casual, social, and permanent movement of people from informal settlements to nearby formal houses with exposure to different types of water sources and sanitation facilities, which might contribute to the risk of getting diarrhoea. Moreover, the cross-sectional study design only allows us to describe associations but does not allow us to infer causality. Finally, the study did not involve any biomedical tests which could provide stronger information on diarrhoea aetiology.

The study found several factors that were significantly associated with the increased risk of diarrhoea among children under five years old living in informal and formal settlements of the Lotus River catchment area in Cape Town. Namely, these factors are caregiver's education level, storing water longer than 12 h, not using proper water treatment, sanitation facility shared with more than four households, dysfunctional toilet facility, using pesticides inside the houses, poor hand-washing practices, longer duration of breastfeeding, and not having Hepatitis A vaccination. It was clear that good hygiene practices are key in the prevention of diarrhoea. Based on those findings, the promotion of hygienic practices is highly recommended. Caregivers should have more knowledge about and exercise better the practices of hand-washing, hygienic breastfeeding, and food preparation. Proper and healthy water storage is necessary, covering water containers, and basic water treatment techniques, such as boiling, filtration, and disinfection, should be observed. Local authorities should support improvement on the adequate sanitation and drainage service provision in both formal and informal settlements. A periodic water quality monitoring campaign at the communal level will be necessary to identify early enough possible water contamination problems, which can lead to diarrhoea. Finally, further studies on diarrhoea aetiology and transmission pathways in the informal urban context are required to get a better understanding of diarrhoea causality in these communities.

Estimates of the global, regional, and national morbidity, mortality, and aetiologies of diarrhoea in 195 countries: A systematic analysis for the Global Burden of Disease Study

Statistics South Africa. Levels and Trends of Morbidity and Mortality among Children Aged Under-Five Years in South Africa

City of Cape Town. Diarrhoea Statistics Prove Interventions Are Working

Diarrhoea morbidity and mortality in older children, adolescents, and adults

Prevalence of diarrhoea and risk factors among children under five years old in Mbour, Senegal: A cross-sectional study

Preventing Disease through Healthy Environments: A Global Assessment of the Burden of Disease from Environmental Risks

Inheriting a Sustainable World: Atlas on Children's Health and the Environment

Urbanization and Development: Emerging Futures

The challenge of COVID-19 in African cities: An urgent call for informal settlement upgrading

Novel Coronavirus in Cape Town Informal Settlements: Feasibility of Using Informal Dwelling Outlines to Identify High Risk Areas for COVID-19 Transmission from A Social Distancing Perspective

Africa's Urbanisation Dynamics 2020: Africapolis, Mapping a New Urban Geography

Globalization and Spatial Mobilities: Commodities and People, Capital, Information and Technology

City of Cape Town. Diarrhoea Trends Suggest that More Hard Work is Required

Water Quality and Diarrhoeal Diseases in Children in the Lotus River Catchment Area in Cape Town

Reducing diarrhoea deaths in South Africa: Costs and effects of scaling up essential interventions to prevent and treat diarrhoea in under-five children

Urban catchment management in a developing country: The Lotus River project

Keep Calm and Learn Multilevel Logistic Modeling: A Simplified Three-Step Procedure Using Stata, R, Mplus, and SPSS

Association between climate variability and hospital visits for non-cholera diarrhoea in Bangladesh: Effects and vulnerable groups

Effect of climatic variability on childhood diarrhea and its high risk periods in northwestern parts of Ethiopia

Association of climate variability and childhood diarrhoeal disease in rural Bangladesh

Temperature Variability and Occurrence of Diarrhoea in Children under Five-Years-Old in Cape Town Metropolitan Sub-Districts

Childhood diarrhea in high and low hotspot districts of Amhara Region, northwest Ethiopia: A multilevel modeling

Climate Change is Likely to Worsen the Public Health Threat of Diarrheal Disease in Botswana

Heatwaves and health: Reflections on the El Niño phenomenon in Piura, Peru. Cadernos Saúde Pública

Impact of Climate Change on Children's Health in Limpopo Province, South Africa

Diarrhoeal Disease in Relation to Possible Household Risk Factors in South African Villages

Preventing Diarrhoea through Better Water, Sanitaiton and Hygiene Exposures and Impacts in Lowand Middle-Income Countries

Factors Associated with the Duration of Moderate-to-Severe Diarrhea among Children in Rural Western Kenya Enrolled in the Global Enteric Multicenter Study

The Relationship between Distance to Water Source and Moderate-to-Severe Diarrhea in the Global Enterics Multi-Center Study in Kenya

Quality of piped and stored water in households with children under five years of age enrolled in the Mali site of the Global Enteric Multi-Center Study (GEMS)

Manure-borne pathogens as an important source of water contamination: An update on the dynamics of pathogen survival/transport as well as practical risk mitigation strategies

Estimates of global, regional, and national morbidity, mortality, and aetiologies of diarrhoeal diseases: A systematic analysis for the Global Burden of Disease Study

Exploring Household Economic Impacts of Childhood Diarrheal Illnesses in 3 African Settings

Contribution of Water Pollution from Inadequate Sanitation and Housing Quality to Diarrheal Disease in Low-Cost Housing Settlements of Cape Town, South Africa

World Health Organization

Update and SDG Baselines; World Health Organization

Sanitation and Hygiene-Specific Risk Factors for Moderate-to-Severe Diarrhea in Young Children in the Global Enteric Multicenter Study

Descending the sanitation ladder in urban Uganda: Evidence from Kampala Slums

User Perceptions of Shared Sanitation among Rural Households in Indonesia and Bangladesh

Determinants of quality of shared sanitation facilities in informal settlements: Case study of Kisumu

Global assessment of exposure to faecal contamination through drinking water based on a systematic review

Fecal Contamination of Drinking-Water in Low-and Middle-Income Countries: A Systematic Review and Meta-Analysis

Environmental Protection Agency (US EPA). Health Effects Support Document for Perfluorooctanoic Acid

Heavy Metals as a Risk Factor

Surface sediment properties and heavy metal pollution assessment in the Pearl River Estuary

Heavy metals (Pb, Cd, As and MeHg) as risk factors for cognitive dysfunction: A general review of metal mixture mechanism in brain

Arsenic and non-malignant lung disease

Pregnancy outcomes, infant mortality, and arsenic in drinking water in West Bengal, India

Acknowledgments: The authors would like to acknowledge and express thanks to all individuals and families who participated in the study. Special mention is needed for local teams of Edith Stephens Nature Reserve and A.L. Abbott & Associates, especially Dale Issac, who greatly supported the study and helped with logistical arrangements, translation, as well as community engagement during data collection.

The authors declare no conflict of interest. Int. J. Environ. Res. Public Health 2021, 18, 6043 

The study was conducted according to the guidelines of the Declaration of Helsinki, and approved by was approved by the "Ethikkommission Nordwest-und Zen-tralschweiz (EKNZ)" in Switzerland (EKNZ BASEC 2016-00304) and by the "Human Research Ethics Committee (HREC)" of the University of Cape Town in South Africa (HREC REF 248/2016).

Informed consent was obtained from all subjects involved in the study.

The data presented in this study are available from the corresponding author on reasonable request. The data are not publicly available due to the presence of confidential participants' information.