(C) PLOS One
This story was originally published by PLOS One and is unaltered.
. . . . . . . . . .



Examination of water quality at the household level and its association with diarrhea among young children in Ghana: Results from UNICEF-MICS6 survey [1]

['Mohammed Husein', 'Department Of Nutrition', 'Food Science', 'University Of Ghana', 'Accra', 'Carole Debora Nounkeu', 'Limbe Regional Hospital', 'Limbe', 'Seth Armah', 'University Of North Carolina At Greensboro']

Date: 2023-07

Abstract Ghana has made significant progress in expanding water services, but microbial contamination of water is still a major public health problem. The objectives of the study were to: 1) Examine sociodemographic and water access related predictors for the point of use or drinking water quality among rural and urban households, and; 2) Determine the association between the point of use water quality and prevalence of diarrhea among young children in rural and urban households. A secondary data analysis was carried out using the Ghana UNICEF-MICS6 survey taking into account the complex survey design. Logistic regression models were used to carry out the objectives. Among the 2317 households included for water quality testing, majority reported using improved source of drinking water. However, use of unimproved source of water was more common among rural households. In examining water quality at the point of use, it was found that more than 60% of the samples had mid to high levels of E. coli count, with significantly more common among rural compared to urban households (p < .0001). The prevalence of diarrhea among under 5 children was 16.9%. In estimating the risk, E. coli count was not associated with higher diarrhea prevalence. However in urban areas, water storage was associated with increased risk of caregiver-reported diarrhea in children. Also, other factors such as child’s age, maternal education, region and household wealth index predicted diarrhea prevalence. In Ghana, contaminated point of use drinking water is more common in rural household, and in urban areas, water storage is associated with increased risk of caregiver-reported diarrhea in children. In the future, investigation of living condition and environmental hygiene is warranted to further understand different pathways through which risk of diarrhea increases among children.

Citation: Husein M, Nounkeu CD, Armah S, Dharod JM (2023) Examination of water quality at the household level and its association with diarrhea among young children in Ghana: Results from UNICEF-MICS6 survey. PLOS Water 2(6): e0000049. https://doi.org/10.1371/journal.pwat.0000049 Editor: Vikram Kapoor, University of Texas at San Antonio, UNITED STATES Received: June 14, 2022; Accepted: May 22, 2023; Published: June 13, 2023 Copyright: © 2023 Husein 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: We used publicly available Ghana UNICEF-MICS6 survey. UNICEF MICS. (2022). MICS6 TOOLS. Retrieved 6 1, 2022, from UNICEF: https://mics.unicef.org/tools#datacollection. Funding: The authors received no specific funding for this work. Competing interests: The authors have declared that no competing interests exist.

Introduction Safe and clean water is critical to sustainable development and for human survival. It is recognized that access to clean and safe water is a right of every human being [1]. However, according to the progress report on the Sustainable Development Goal (SDG) # 6 on access to safe and clean water for all, about 2 billion people or 26% of the world’s population lack access to safely managed drinking water services [2]. It is recognized that lack of universal access to clean water is impeding the progress in meeting nutrition and public health goals in low- and middle-income countries. Based on the WHO/UNICEF’s Joint Monitoring Program indicators, a drinking water source is considered improved if it is protected from contamination by nature of its construction or through active intervention [2]. Hence, improved water source types include piped water, standpipe, borehole, and protected dug well. However, a review of the water quality studies highlights that getting water from an improved source does not guarantee clean and safe point of use water. Several factors, including water storage facility at the household level, availability of improved sanitation at the community level, and condition of the improved water source affect the safety and quality of water [3]. The 2022 WHO/UNICEF’s JMP report indicated that access to an improved water source is more common in urban compared to rural areas in Sub-Saharan Africa [2]. Especially studies in the rural areas show that not only the type of water source (improved vs. unimproved) but the distance to the water source and related water fetching burden, such as, collecting and storing water, type of water container and proximity to livestock affect the point of use water quality [4–6]. Understanding the sources and pathways through which fecal contamination occurs in the drinking or point of use water is critical in planning concerted efforts and meeting SDG # 6. The achievement of this goal will allow meeting other SDG targets too, such as ending preventable deaths of newborns and children under 5 years of age, since unsafe water is the major risk factor for diarrheal diseases among children causing high levels of morbidity and mortality [7]. In the study involving the UNICEF-MICS data from 27 countries, it was found that low income and rural households were at significant increased risk of using contaminated drinking water than their counterparts, however, the relationship between water quality and diarrhea risk among children was not explored [8]. In Ghana, access to water has improved significantly in the past decade, but microbial contamination of drinking water is still an issue in the country. According to the latest Ghana- MICS report, Escherichia coli (E. coli) contamination in drinking water is more common than in the source water [9]. Hence, to understand water quality issues at the household level and whether disparities exist between rural and urban households, the objectives of the study were to utilize the latest national level data to: 1) Examine sociodemographic and water access related predictors for the point of use or drinking water quality among rural and urban households, and; 2) Determine the association between the point of use water quality and prevalence of diarrhea among young children in rural and urban households.

Methods To meet the study objectives, UNICEF-MICS 2018 dataset for Ghana was used. Ghana is a country in the western part of Africa with a population of about 30 million. The country is currently divided into 16 regions. However, at the time of the collection of the MICS 6 data, there were only 10 regions in Ghana. Thus, our analysis presents the results for the original 10 regions. The Ghana UNICEF-MICS 2018 was designed to be useful in assessing multiple indicators relating to women and children at various levels including national, regional and area (rural/urban) levels. The fieldwork was done from October 15th, 2017 to January 15th, 2018, representing mostly the dry season in Ghana. In the design, urban and rural areas designated by the government within each of the 10 regions were used as the main sampling strata, and households were selected using a two-stage sampling design with a probability approach. Oversampling was done for households with women within the age range of 20–24 years to improve the precision for estimates of indicators of early marriage. Due to the nature of the sampling, sample weights were provided in the data to ensure accurate estimates when reporting results. A total of 3219 rural and urban households were involved in water quality testing. One of the children under 5 from the water quality testing households was selected to assess the diarrhea prevalence and its association with the water quality at the household level. To test the objectives, we selected water quality testing households that had at least one under 5 child which resulted in a sample 2317 households. We used the household data (hh.sav) and the children under 5 data (ch.sav) files. From the household data file, we used the following variables: household wealth index, household head’s education, mother’s or caretaker’s education, urban vs rural household, and region. The computed wealth index and its categories representing a proxy measure of the long-term standard of living of the household was utilized for household income status. The water access related variables and drinking water quality test results were also extracted from household data file. The water access related variables used were: improved vs unimproved drinking water source, location (in the vicinity or not), if not in the vicinity then distance and time to water source, water treatment (yes or no) and water storage (yes or no). In case of water quality measurement, details can be found in the survey manual [10]. But, a range of quality control measures were put in place to ensure accuracy in measurement. For instance, “blank” samples known to be free of contamination were tested to ensure that contamination in the water samples was not due to poor survey testing techniques, such as dirty hands or equipment. Training was also carried out to ensure accuracy and robustness of the water quality data. E. coli count was reported in the range of 0 to 100 cfu. The water testing results of > 100 cfu per 100 mL sample water was grouped as one category of >100 cfu. For this reason, we grouped households into four categories (0 cfu or absent, 1–10 cfu or low, 11–100 cfu or medium and >100 cfu). From the children under 5 data file, child’s sex, age and reported prevalence of diarrhea in the past 2 weeks were retrieved. Statistical analysis Data were analyzed using R software. We used the package survey in R [11]) taking into account the strata, principal sampling unit (PSU), and sampling weight. For continuous variables, such as time to water source, we reported the means and the standard error (SE). For categorical variables, the adjusted number (frequencies) and percentages were reported. For each analysis households with missing data were excluded. We used chi-squared analysis with the adjusted Wald statistics to determine the bivariate association between categorical variables, such as area (rural vs. urban) and sociodemographic characteristics of household. To determine sociodemographic and water access related predictors of water quality overall and specific by urban and. rural households, we used ordinal logistic regression analysis utilizing 4 E. coli count categories as dependent variable. The independent variables were water source, time to water source, water treatment, water storage, caretaker education level, wealth index category of the household, and the region. For prevalence of diarrhea, a binary variable (yes or no) in reference to question on whether the reference child in the household experienced diarrhea in the past two weeks was used. Separate models were constructed for urban and rural areas, and an overall model including both areas was also constructed. In all the regression models, initially we also included the following sanitation variables: 1) Type of toilet facility used (improved vs unimproved vs open defecation) and 2) whether the toilet facility was shared with other households. Using the WHO/UNICEF JMP ladder for sanitation, the combination of these two variables were used to group households into the following 4 sanitation categories: improved (improved facility and not shared with other households) vs. limited (improved facility shared with other households) vs. unimproved facility vs. open defecation. However, this sanitation ladder did not significantly affect water quality or caregiver-reported diarrhea prevalence in children. Therefore, it was excluded to maintain statistical power. Statistical significance was set at p ≤ 0.05.

Discussion The results of this nationally representative study in Ghana indicates that though use of improved drinking water source is common, the point of use water contaminated with E. coli exist. Prevalence of drinking water or the point of use water contaminated with E. coli was 62.3% in urban and 88.7% in rural households. This study also shows that disparities exist in the odds of drinking clean water by region and income level, indicating regional and area level specific investigation and initiatives are needed to improve water quality and bring equity in water security. Since the beginning of the “Water and sanitation for all” initiative launched by the UN agencies and charitable organizations in the early 1980s, water supply and sanitation infrastructures coverage in middle- and low-income countries has improved significantly [12,13]. However, consumption of contaminated drinking water still represents a significant issue in low and middle countries and represents a huge burden on country’s productivity and economic potential [14,15]. Our study showed that fecal contamination of water was significantly influenced by rural location and poverty. Similarly, a systematic review of 345 studies involving coliform measurement of 133,460 water samples in Africa and South-East Asia, indicated the rural areas are most affected by fecal contamination [3]. In understanding the water contamination risk, reviews have shown that post-collection contamination is more common and plays a major role in predicting water quality compared to just knowing whether the water source is improved or unimproved [16,17]. Consequently, the estimation of access to safe and clean water based on use of improved water sources has been criticized since it causes an underestimation of the burden of waterborne diseases. For instance, in a meta-analysis of 45 studies to assess the extent of fecal contamination at the source and in household stored water, it was found that whereas 46% of the point of collection samples were contaminated, the percentage of contaminated household water samples was almost double (75%) [16]. To address poor water quality among rural households where water collection from piped systems is uncommon, the use of water treatment techniques, is recognized as a necessary step in ensuring clean water. In this study, the use of treatment methods to purify water was not common. In the analyses of the UNICEF-MICS 2014-Nepal survey, Kandel et al. found that households who boiled water had about 50% lower odds of fecal contamination than households who did not boil water [18]. Further, the results of this study demonstrated thathousehold wealth status was a significant predictor of poor water quality. Households in richest wealth index category were less likely to consume unclean water, this pattern was seen in both rural and urban households. In comparison by regions in Ghana, in urban areas Volta region had significantly increased risk of contamination compared to Greater Accra (the reference region). Guzmán and Stoler conducted a study in Ghana to compare water quality in rural and semi-urban communities in several regions including the Volta region [19]. Their results also showed similar regional differences in water quality and concluded that intermittent water shortage and non-functional improved sources caused mixed use of improved and open surface water in certain areas increasing the risk of drinking contaminated water. In recent years, use of the bagged sachet water for drinking has become very common in Ghana in both rural and urban areas [19,20]. In our study, we were not able to specifically identify the use of sachets or bottled water for drinking. But, according to the 2014 Ghana Demographic and Health Survey results, sachet water was the primary drinking water source for 29% of households overall, with about 43% of urban and 12% of rural households reporting its use [21]. In our study, no significant association was found between E. coli in drinking water and the caregiver-reported prevalence of diarrhea among young children. Similarly, in a meta-analysis of 28 observational and intervention studies, no significant positive relationship has been found between point-of-use water contamination (E. coli and thermotolerant coliforms) and diarrhea among children [22]. In the study utilizing Bangladesh UNICEF-MICS, Khan et al found a significant dose-response association i.e., as the E. coli count increased in drinking water the risk of diarrheal episodes among under-5 children also increased [23]. In a prospective longitudinal study, it was found that contaminated drinking or point of use water was associated with subsequent incidence of diarrhea among children [24]. As argued by Levy, measurement of water quality just once with concurrent measurement of diarrhea might mask the link between the two as most diarrhea causing pathogens have incubations periods of > 24 hours [25]. Further, results of the meta-analysis highlights that not only contaminated drinking water but contaminated surfaces such as child’s hands play a role in determining risk for diarrhea [26]. While we did not find any association between E. coli count and diarrhea prevalence, water storage was associated with increased risk of reported diarrhea in children, particularly in urban areas. It was also not surprising that maternal education was a significant predictor for caregiver-reported diarrhea among children in our study. Maternal education has consistently remained significant, with children of less-educated mothers shown to be at higher risk of experiencing diarrhea [27]. In a study involving 915 censuses and nationally representative surveys, it was found that between 1970 and 2009, an increase in the average years of education among women contributed 50% reduction in under-5 mortality [28]. Children of more educated mothers are more likely to receive vaccines, visit a doctor for sickness, sleep under insecticide-treated nets, and benefit from other health-protective practices. In our study too, the protective effect of maternal education can also be due to high likelihood of practicing behaviors that help prevent water contamination, such as more likely to treat water, use covered containers to store water, and/or wash hands with soap and water. Notwithstanding the evidence, there are some limitations to consider while interpreting our findings, as well as for future research. First, the results are based on a momentary snapshot of household water quality. However, evidence suggests that water quality varies significantly depending on the season (dry vs. wet), from day to day based on water services functioning, and can even be conditioned by the time availability of the main water fetcher. Further, secondary data analyses carry inherent limitations, including utilizing only the variables that are reported and there is also possibility of unmeasured and residual confounding affecting the ability to measure the strength of the association between behaviors and outcome variables. This study targeted one major source of contamination i.e. drinking water, but other sources, such as food, children’s hands and surroundings, can also cause fecal-oral route transmission. Finally, the use of the ordinal logistic regression model in the analysis of this complex survey data assumes that across all levels of the outcome variable, as we move from one level of to the next, the effect of an independent variable is constant (proportional odds assumption).

Conclusions Our study results indicate that, disparity in water quality exist in Ghana, with poor water quality more common in rural versus urban households. While we did not find any association between E. coli count and diarrhea prevalence, water storage was associated with increased risk of reported diarrhea in children in urban areas. In light of our findings, future investigation on how other factors such as environmental hygiene, living condition, and use of food preparation practices affect diarrhea risk, is warranted. A full understanding of factors and specific pathways that affect water quality at the point of use is also critical in meeting the SDG goals and ensuring reduction in inequity due to lack of access to safe and clean water.

[END]
---
[1] Url: https://journals.plos.org/water/article?id=10.1371/journal.pwat.0000049

Published and (C) by PLOS One
Content appears here under this condition or license: Creative Commons - Attribution BY 4.0.

via Magical.Fish Gopher News Feeds:
gopher://magical.fish/1/feeds/news/plosone/

You are viewing proxied material from magical.fish. The copyright of proxied material belongs to its original authors. Any comments or complaints in relation to proxied material should be directed to the original authors of the content concerned. Please see the disclaimer for more details.