(C) PLOS One
This story was originally published by PLOS One and is unaltered.
. . . . . . . . . .
Impact on childhood mortality of interventions to improve drinking water, sanitation, and hygiene (WASH) to households: Systematic review and meta-analysis [1]
['Hugh Sharma Waddington', 'Environmental Health Group', 'Department Of Disease Control', 'London School Of Hygiene', 'Tropical Medicine', 'Lshtm', 'London International Development Centre', 'Lidc', 'London', 'United Kingdom']
Date: 2023-05
The findings are congruent with theories of infectious disease transmission. Washing with water presents a barrier to respiratory illness and diarrhoea, which are the two biggest contributors to all-cause mortality in childhood in L&MICs. Community-wide sanitation halts the spread of diarrhoea. We observed that evidence synthesis can provide new findings, going beyond the underlying data from trials to generate crucial insights for policy. Transparent reporting in trials creates opportunities for research synthesis to answer questions about mortality, which individual studies of interventions cannot be reliably designed to address.
Further analysis by WASH technology indicated interventions providing improved water in quantity to households were most consistently associated with reductions in all-cause mortality. Community-wide sanitation was most consistently associated with reductions in diarrhoea mortality. Around one-half of the included studies were assessed as being at “moderate risk of bias” in attributing mortality in childhood to the WASH intervention, and no studies were found to be at “low risk of bias.” The review should be updated to incorporate additional published and unpublished participant flow data.
We included evidence from 24 randomized and 11 nonrandomized studies of WASH interventions from all global regions, incorporating 2,600 deaths. Effects of 48 WASH treatment arms were included in analysis. We critically appraised and synthesised evidence using meta-analysis to improve statistical power. We found WASH interventions are associated with a significant reduction of 17% in the odds of all-cause mortality in childhood (OR = 0.83, 95% CI = 0.74, 0.92, evidence from 38 interventions), and a significant reduction in diarrhoea mortality of 45% (OR = 0.55, 95% CI = 0.35, 0.84; 10 interventions).
We conducted a systematic review and meta-analysis, using a published protocol. Systematic searches of 11 academic databases and trial registries, plus organisational repositories, were undertaken to locate studies of WASH interventions, which were published in peer review journals or other sources (e.g., organisational reports and working papers). Intervention studies of WASH improvements implemented under endemic disease circumstances in L&MICs were eligible, which reported findings at any time until March 2020. We used the participant flow data supplied in response to journal editors’ calls for greater transparency. Data were collected by two authors working independently.
In low- and middle-income countries (L&MICs), the biggest contributing factors to the global burden of disease in childhood are deaths due to respiratory illness and diarrhoea, both of which are closely related to use of water, sanitation, and hygiene (WASH) services by households. However, current estimates of the health impacts of WASH interventions use self-reported morbidity, which may fail to capture longer-term or more severe impacts. Reported mortality is thought to be less prone to bias than other reported measures. This study aimed to answer the question: What are the impacts of WASH interventions on reported childhood mortality in L&MICs?
Copyright: © 2023 Sharma Waddington 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.
In this paper, we present a different approach to estimate the health effects of WASH interventions. There is a large number of trials of WASH interventions, sufficient numbers on which to estimate global effects on mortality, even when the individual studies themselves did not aim to do so. We conducted a systematic review of the effects of WASH interventions on child mortality in L&MIC contexts, drawing on a number of sources including losses to follow-up due to mortality as reported in participant flows. It is an established finding that study participants do not misreport death, even in open studies [ 15 , 16 ]. This might be because death of a child is a rare and salient event. The crucial advantage of this approach, therefore, is that reported mortality is less prone to bias.
The key advantage of randomised controlled trials (RCTs) over other methods is the clarity with which randomisation balances unobservable differences across groups in expectation, not in any single trial, but over multiple draws from the population [ 21 ]. Thus, the “gold standard” for evidence on health impacts from these studies uses meta-analysis of findings from multiple studies [ 22 ]. However, meta-analysis can also magnify biases, because it is harder to identify errors where they pervade the whole data set. Some approach is clearly needed to address reporting bias. Of great potential concern is publication bias, the phenomenon whereby studies are more likely to be published if they find significant effects, a factor that is made more likely when they are funded by private manufacturers, as has been common in trials of water treatment (chlorine, water filters) and hygiene (soap) [ 23 ].
A common finding in existing reviews is that bundling WASH together does not produce additive effects in comparison with single water, sanitation, or hygiene technologies [ 12 ]. One possible reason for this finding is bias in reporting. For example, the most common method of collecting health outcomes data in impact evaluations of WASH interventions is through participant report [ 10 ]. However, data on reported illness have been shown to be biased in open (also called “unblinded”) trials [ 13 – 16 ]. Perhaps carers might misrepresent illness to minimise the time spent with enumerators when data are collected repeatedly over time [ 17 , 18 ]. Social desirability bias may also arise where participants are inadvertently induced to report favourably. Briscoe and colleagues [ 19 ] highlighted how diarrhoeal illness becomes normalised among highly exposed groups over time, which leads to underreporting, a problem we might expect to become worse when reporting is done by someone other than the patient, in this case the child’s carer. Or illness may be acknowledged differently by sex [ 20 ], if girls who complain about pain are less likely than boys to be pacified by their carers and, therefore, report less illness. In other words, we may not see additive effects of multiple WASH technologies provided together if bias in the reporting of disease outcomes, rather than diarrhoea epidemiology, is driving the findings.
GBD estimates of WASH-related mortality are presently calculated using estimated coefficients on diarrhoea morbidity impacts from systematic reviews and meta-analyses. Estimates vary widely (Table B in S1 Annex ), suggesting great imprecision affecting our measurement of the gravity of the diarrhoea problem, globally or in any specific context. Of the 44 systematic reviews included in a recent WASH sector-wide interventions evidence map [ 10 ], half of which concerned effects of WASH provision on diarrhoea, none had synthesised the evidence on mortality in childhood. The most recent systematic evidence on WASH interventions and diarrhoeal illness was reported in The Lancet in July 2022 [ 11 ].
Unfortunately, studies of the effects of WASH interventions on diarrhoea and other causes of mortality are beset by such ethical and logistical difficulties that, with few exceptions (e.g., [ 5 ]), practically none were carried out until recently (e.g., [ 6 – 8 ]). For example, it could not be ethical to design a prospective study to measure mortality as a primary outcome when lifesaving oral rehydration solution (ORS) is widely available and affordable. As a result, and in accordance with the recommendation of the Grading of Recommendations, Assessment, Development, and Evaluations (GRADE) procedure [ 9 ], the focus shifted from mortality to morbidity—mainly from diarrhoea—as a more accessible outcome.
Diarrhoeal diseases and respiratory infections are thought to kill 4.1 million people each year [ 1 , 2 ]. Half of these deaths are of infants and young children aged less than 5 years old [ 3 ], around 1.2 million of whom live in circumstances without adequate drinking water, sanitation, and hygiene (WASH) in low- and middle-income countries (L&MICs) [ 4 ]. The global burden of disease (GBD) for communicable causes is weighted heavily by mortality in childhood, the two biggest single causes of which are diarrhoea and respiratory infection. Approximately 90% of the total diarrhoea GBD and 99% of the total respiratory infection GBD are due to years of life lost (YLL) (Table A in S1 Annex ).
We evaluated the likelihood that potential biases could cast doubt on the results (review question 4). The effects of WASH improvements on mortality are largely expected to occur by blocking transmission of infectious diseases, primarily faeco-oral and respiratory infections, in childhood. People who survive beyond the age of 5 are thought to have developed sufficiently robust immunity to these diseases; hence, the effects of WASH improvements on mortality among older groups is expected to be far weaker. Therefore, as a negative control [ 41 , 42 ], meta-analysis was estimated for those studies that reported all-cause mortality among a “placebo population” of participants aged over 5 years. We also assessed the sensitivity of the pooled effects to exclusion of each single effect, examined whether there was a correlation between risk-of-bias rating and the estimated effect, and tested for small-study effects (publication bias) at the review level using graphical inspection of funnel plots and regression tests.
Overall pooled effects were estimated for all-cause mortality (review question 1) and diarrhoea mortality (review question 2) using Stata. We assessed the consistency of the pooled effects using I-squared and tau-squared statistics to measure the relative and absolute heterogeneity between studies. We tested for effect moderators in meta-analysis and meta-regression analysis, including the WASH intervention technology provided to study participants, water supply and sanitation conditions in the counterfactual group, participant characteristics (age and if from immunocompromised group), and study characteristics (season of data collection and length of follow-up). We report forest plots showing country and WASH technologies for each analysis (we also report the same forest plots by study author in Figs A-G in S1 Annex ). To aid interpretation of the meta-regression coefficients, we calculated OR prediction values at the means, minima and maxima of the dichotomous variables, and the mean and interquartile range of the continuous variable. Moderator variables were prespecified based on theory and previous reviews, with the exceptions of the moderator analysis by baseline mortality rate and the negative control. We used meta-regression plots to assess the predicted effects of the interventions by baseline mortality rate (review question 3).
Mortality rates were computed over a standard period, as mortality measurements increase over longer exposure periods. Age-specific (e.g., under 2) mortality rates were defined where these data were available [ 6 – 8 , 34 ], or, if they were not, crude mortality rates were taken over the data collection period. Intervention effects were measured as the odds ratio (OR) of the mortality rates, and their 95% confidence intervals. Where studies reported multiple intervention arms against a single control arm, we split the control sample assuming an equal mortality rate for each comparison. We applied a continuity correction in study arms where there were no deaths, by adding 0.5 to all frequencies, which can cause bias in meta-analysis of rare events [ 35 ]. These studies were assessed as being at “high risk of bias” in the outcome measurement domain [ 36 – 40 ].
The primary outcomes for the review were all-cause mortality and mortality due to diarrhoeal illness. Outcomes data were collected independently by two researchers from two sources. The first source was the few studies that reported mortality alongside statistical information [ 6 – 8 , 32 , 33 ]. Mortality data were also recoverable from studies that reported losses to follow-up (attrition) in sample populations. Participant flow diagrams were reviewed in all studies of WASH technologies in L&MICs to obtain crude mortality rates for field trials by intervention group. These studies, therefore, formed the major source of evidence on all-cause mortality. Some studies also reported cause-specific mortality rates, including diarrhoea and other infections, defined by carers in verbal autopsy and/or clinicians, or collected from vital registries.
A risk-of-bias tool was developed for WASH impact evaluations that drew on Cochrane’s tools for RCTs [ 27 ], cluster RCTs [ 28 ], and nonrandomised studies of interventions [ 29 ], and a tool for appraising quasi-experiments [ 30 ]. Six bias domains were assessed: confounding, selection bias, departures from intended interventions, missing data, outcome measurement bias, and reporting bias. The studies were assessed on the likelihood of bias in estimating effects of WASH access on mortality in children aged 5 years or under. This may or may not have been a primary research question in the papers themselves; hence, our ratings do not provide risk-of-bias assessments for the study overall. The risk-of-bias assessments were done by two researchers working independently, at the outcome level for each included study arm, as recommended by Cochrane [ 22 ] and the Campbell Collaboration [ 31 ]. Template data collection forms are available in the study protocol [ 24 ]. Data extracted from included studies are provided in Table C in S1 Annex . The dataset used in analysis is provided in S1 Dataset .
Counterfactual conditions were categorised as “improved” or “unimproved” according to the WHO/UNICEF Joint Monitoring Programme (JMP) classification. Improved water supplies were defined where the majority of households in the control sample used drinking water from an improved source (e.g., piped water to the household, a community standpipe, or protected spring) within a 30-minute round trip including waiting time. For sanitation, the counterfactual scenario was defined as “improved” if the majority of controls had a sewer connection to the home or an improved pit latrine was used by a single household. Where insufficient information was reported about the counterfactual scenario to categorise baseline water supply or sanitation use, the figures were imputed from online data provided by the JMP for the relevant country, year, and location.
Eligible studies were RCTs and nonrandomised studies of interventions (NRSI) promoting access to or use of WASH technologies to households in L&MICs in endemic disease circumstances. We included new or improved water supplies, drinking water treatment and storage, sanitation, and hygiene technologies, including those enabling or promoting handwashing at key times and other beneficial household practices (e.g., the washing of food, clothing, and fomites). We excluded trial arms with a major non-WASH component (e.g., nutrition interventions). We classified WASH interventions according to the “main WASH” technology provided, which was either water supply, water treatment and storage, sanitation, or hygiene technologies provided or promoted alone, or multiple combinations of WASH technologies. It was also possible to characterise interventions by whether they provided any improvements in water supply, water treatment, sanitation, and/or hygiene alone or in combination with others, which we refer to as “any WASH.” This was due to problems in clearly identifying all the components of an intervention. For example, a debate among practitioners suggested that hand hygiene messaging is usually incorporated in community-led total sanitation (CLTS) [ 26 ].
This review was registered with Prospero under registration number CRD42020210694 and is reported following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guideline ( S1 PRISMA Checklist ). A full description of the procedures followed for searches, study inclusion, outcomes data collection, analysis, and reporting is presented in the published protocol [ 24 ]. Searches for literature were done as part of an evidence and gap map [ 10 ]. Studies selected were published at any time until March 2020. Eleven academic databases and trial registries (e.g., Cochrane, Econlit, Medline, OpenTrials, Scholar, Web of Science) and sources of nonpeer-reviewed literature including databases and organisational repositories were searched (e.g., 3ie Repositories, J-PAL, IRC International Water and Sanitation Center, UNICEF, the World Bank, and the regional development banks). We used reference snowballing, including bibliographic backreferencing and forward citation tracking of studies and existing reviews. As a measure to reduce publication bias, studies published in any format were eligible, and searches done of repositories of this information. As a measure to avoid language bias, studies published in English, French, Spanish, and Portuguese were included, and searches done of repositories of this information. A priority search algorithm based on machine learning was used in filtering studies at title and abstract stage using EPPI-reviewer software [ 25 ]. Selection of studies was done by two authors working independently.
Results
Impacts of WASH on diarrhoea mortality (review question 2) The meta-analysis of diarrhoea mortality in childhood indicated WASH provision and promotion lead to a significant reduction in the odds of death due to diarrhoea by 45% (OR = 0.55, 95% CI = 0.35, 0.84; 10 estimates) (Fig 9). Six of the studies were assessed as being at “high risk of bias” [33,38,44,46,50,65], and three were at “moderate risk” [32,60,64]. The relatively high degree of absolute and relative heterogeneity in findings (I-squared = 43%, tau-squared = 0.15) suggested additional analysis was needed of factors that could explain the variation across study contexts. PPT PowerPoint slide
PNG larger image
TIFF original image Download: Fig 9. Effects on diarrhoea mortality in childhood of WASH interventions.
https://doi.org/10.1371/journal.pmed.1004215.g009 One of those factors is the degree of movement up the WASH ladders. We tested this hypothesis in moderator analysis according to the type of water supply and sanitation facilities used in the counterfactual group. When the WASH interventions were provided when counterfactuals were using no or unimproved sanitation and water supplies and, therefore, exposed to very high risk of environmental contamination by pathogens, there was an estimated 69% reduction in diarrhoea mortality in childhood (OR = 0.31, 95% CI = 0.16, 0.60, I-squared = 17%, 4 estimates). But for interventions provided in circumstances when most people already had access to improved sanitation, there was only a 22% reduction in odds of mortality (OR = 0.78, 95% CI = 0.62, 0.98, I-squared = 0%, 6 estimates) (Fig 9). The impacts of WASH interventions on childhood diarrhoea mortality were significantly greater (p < 0.01) when counterfactual groups lacked access to improved water supply and sanitation—and most people were therefore using unimproved facilities, or none at all and openly defaecating—than when most people in counterfactual groups were using improved facilities. The largest effects on diarrhoea mortality were from studies of multiple WASH technologies: two contained a component that aimed to provide latrines to all households in intervention communities [32,33], and two involved water supply improvements [33] or hygiene promotion when water supplies were already improved [38]. With regard to the two studies of latrine provision or promotion to whole communities, both were provided alongside hygiene promotion, but only in Côte d’Ivoire was the water supply also improved [33]. In the case of Mali [32], hygiene promotion was given as part of CLTS when water supplies were limited. Another longitudinal follow-up study of an RCT of hygiene improvement, which was rated at “high risk of bias,” was conducted among communities where some households had access to running water for only 2 hours each week [65], suggesting these households had limited opportunities for adherence to improved hygiene practices. Few studies of household water treatment in endemic circumstances have reported diarrhoea mortality outcomes. Among the studies examining household water treatment, only one was of an approach that has been found to reduce diarrhoea morbidity; the study was of filtration [60], and it found large but statistically insignificant impacts in children from immunocompromised populations (HIV–positive mothers). The other was a study of chlorine provision alongside safe storage and hygiene education [38]. Meta-regression analysis suggested interventions providing community-wide sanitation, and hygiene promotion in circumstances when water supplies were improved were associated with significantly larger impacts on diarrhoea mortality (Table D in S1 Annex).
[END]
---
[1] Url:
https://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.1004215
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/
