The 10 studies included six randomized controlled trials [4,32–36], two controlled before/after studies [37,38], one interrupted-time series , and one historically controlled trial . Primary participants were PLHIV for eight studies; two studies examined young children born to HIV-positive mothers whose HIV status was not fully ascertained [36,40]. Two studies reported outcomes for all members of the household in addition to the primary participants (PLHIV or children born to HIV-positive mothers) [4,36]. With the exception of Sorvillo et al. that examined filtration at the water treatment plant level, the intervention allocation occurred at the household level. Three studies were carried out in the United States and seven in sub-Saharan Africa. Although the US-based studies had predominantly male participants (75–98%), females constituted 74–100% of participants in the sub-Saharan Africa studies. The follow-up period ranged from 16 weeks  to 8 years . The studies covered 12 690 participants with HIV/AIDS (primarily from 10 988 in the Sorvillo et al. study) and 591 children born to HIV-positive mothers, totalling 13 281 individuals. The studies were published from 1994 to 2012, with one study under review for publication at the time of this review.
Except for one study that examined a handwashing intervention , all study interventions consisted of measures to improve drinking water quality. One study was of a filtration addition to a water treatment plant , and eight studies were of household water treatment interventions, including five filtration studies [two ceramic pot filters [34,35], two LifeStraw Family filters [36,38], one filter combined with ultraviolet disinfection  and three household chlorination studies [4,39,40]. Four studies included safe water storage containers as part of the intervention [4,36,39,40]; additionally, the two ceramic pot filter studies integrated water storage as part of the device [34,35]. In one study, the intervention was a combination of a long-lasting insecticide-treated bednet and LifeStraw Family filter ; therefore the outcomes cannot be separated for the two interventions. One study examined the combined effect of cotrimoxazole prophylaxis and household chlorination after examining chlorination alone for 5 months ; only the results for chlorination alone are included in this review.
Intervention adherence is characterized as correct, consistent and sustained use, also referred to as compliance . Studies varied in whether and how they assessed participant adherence to the WASH intervention. Three studies assessed adherence based on participant self-reports [33,36,38], two studies of household chlorination reported on chlorine residual levels [39,40], and three studies compared microbial water quality in control and intervention arms [4,35,36]. For reported adherence, household filtration was reported in the intervention group in 93% of households in the Walson et al. study and 96% in the Peletz et al. study. In the Huang and Zhou study, handwashing was more frequently reported in the intensive handwashing intervention group compared with the control (seven vs. four times a day, P < 0.05). Chlorine residual was present in 50–80% of intervention households  or 80–92% of intervention households . Microbial water quality was significantly improved in intervention households compared with control households (P < 0.001) in Lule et al. and Peletz et al. studies; this comparison is not evaluated statistically in Potgieter et al. Neither Colford nor Sorvillo et al. studies reported on adherence; although it was assumed high as Colford attached the water treatment intervention to the main faucet of the household and Sorvillo et al. examined filtration at the water treatment plant.
Outcomes measures and effect estimates
Nine studies examined diarrhoea-related morbidity and results were reported by eight (results not reported in Potgieter et al. 2010) (Fig. 2). With the exception of the Harris study that used clinic visits for diarrhoea, studies with diarrhoea as an outcome relied on self-reports by participants. For the case definition of diarrhoea, six studies used the WHO definition (≥three loose stools per day), one study used ‘highly credible gastrointestinal illness’ which counts vomiting or abdominal cramps as well as diarrhoea  and one study did not provide a definition .
Households were visited or called periodically to assess self-reported diarrhoea-related outcomes. Pictorial diarrhoea diaries or health logs were used in four studies wherein participants were instructed to record health outcomes daily [32–35]. Household interviews were used in four studies, ranging from weekly [4,36,39] to quarterly visits , where participants were asked about diarrhoea in the preceding time period.
Diarrhoea outcomes were reported as rate ratios [4,34], risk ratios [32,38,39], longitudinal prevalence ratio , difference in absolute diarrhoea episodes per year , or difference in absolute clinic visits per month . Although some studies presented adjusted ratios [4,32], we used crude ratios in our analysis to avoid pooling ratios that were adjusted for different factors; for the Lule et al. study we chose to use diarrhoea episodes as the main outcome though days with diarrhoea were also reported.
All studies reporting on diarrhoea found some reduction in morbidity, ranging from 17  to 77% , although the 25% reduction reported by Colford was not statistically significant (Fig. 2). The single handwashing study  reported a reduction of 58% (RR = 0.42, 95% CI: 0.33–0.55). The pooled reduction from the water quality interventions was 43% (RR = 0.57, 95% CI 0.38–0.86) (Fig. 2). However, there was substantial heterogeneity of the water quality studies (probability of heterogeneity, χ 2 = P < 0.001) and 95.0% consistency (I 2, P < 0.001).
Only the Barzilay et al. study stratified diarrhoea by intervention adherence; diarrhoea was significantly reduced among high frequency chlorination users (46% reduction, P = 0.04) but not among low-frequency users (15% reduction, P = 0.47) . Only two studies reported results for all members of households with an HIV-positive individual. For household members in the Lule et al. study, there were borderline significant reductions in diarrhoea episodes (adjusted RR = 0.80, 95% CI 0.64–1.0, P = 0.047) and days with diarrhoea (adjusted RR = 0.74, 95% CI: 0.54–1.01, P = 0.055) . For household members in the Peletz et al. study, there was a significant reduction in diarrhoea (LPR = 0.46, 95% CI:0.30–0.70, P < 0.001) but not persistent diarrhoea (≥14 days) (LPR = 0.75, 95% CI: 0.37–1.53, P = 0.43) .
Two studies examined cryptosporidiosis as a primary outcome; one household ceramic filter study in South Africa  and one study including filtration at a water treatment plant in the United States . Cryptosporidiosis was verified by stool samples  or records from the national AIDS surveillance for all PLHIV .
Abebe et al. found no significant difference in cryptosporidiosis prevalence at the end of the study (7% household filtration group vs. 22% control, P = 0.11), although they did find a significant reduction between baseline and final visits in the intervention group (25% reduction, P = 0.02) and not in the control (4% reduction, P = 0.74) . Sorvillo et al. found no effect from water treatment plant filtration. Although prevalence declined by 20% (from 4.2 to 3.4%) after the filtration addition, prevalence also declined by 47% (from 6.2 to 3.3%) in a neighbouring area that had not changed their water treatment plant technology during the same time period . In the Lule et al. and Huang and Zhou studies, participants with diarrhoea were tested for Cryptosporidium spp. in addition to other pathogens as secondary outcomes, no significant difference was found between intervention and control groups [4,33].
Three studies examined other enteric infections of PLHIV as secondary outcomes, by collecting stool samples of participants with diarrhoea [4,33], or using a new bio-wipes technique . Lule et al. tested for hookworms, Strongyloides stercoralis, enterotoxigenic Escherichia coli, enteropathogenic E. coli, Aeromonas spp., Shigella spp., Salmonella spp., Campylobacter spp., Vibrio cholerae, and Pleisiomonas spp. Huang and Zhou tested participants with diarrhoea for Shigella spp., Campylobacter spp., enteroaggregative E. coli, Clostridium difficile, Yersinia enterocolitica, Salmonella spp., human cytomegalovirus, adenovirus, norovirus, rotavirus, Giardia lamblia, Entamoeba histolytica, and Microsporidium. In the study by Potgieter et al., samples were tested for pathogenic E. coli (five types), Shigella flexneri, and Salmonella typhimurium. None of the studies assessed exposure for enteric infections specifically.
The Lule et al. and Huang and Zhou studies reported no significant differences between intervention and control groups in rates of the infection among PLHIV except in the case of G.lamblia in the Huang and Zhou  study (2 vs. 6%, P < 0.05). For HIV-negative members of the household in the Lule et al. study, the intervention group had lower rates of hookworm than the control (27 vs. 40%, P = 0.0138) and Shigella species (1 vs. 5%, P = 0.0292) . In the Potgieter et al. study, results were not stratified by intervention group for PLHIV.
Progression of HIV/AIDS was examined in three studies. In the Walson et al. study where disease progression was the primary outcome, individuals receiving bednets and water filters were 27% less likely to reach the endpoint of CD4+ cell count of less than 350 cells/μl after controlling for baseline CD4+ cell counts (HR = 0.73, 95% CI: 0.57–0.95, P = 0.02) . CD4+ cell decline was significantly lower in the intervention group (−54 vs. −70 cells/μl per year, P = 0.03) .
The Lule et al. and Potgieter et al. studies reported on the impact of household water treatment on progression of HIV/AIDS, although this was not the primary outcome in either study. Lule et al.  found household chlorination did not impact viral load, although diarrhoea episodes were significantly associated with viral load and HIV viral load increased by 0.40 log10 per person-year for PLHIV using household chlorination compared with 0.71 log10 per person-year in control. Potgieter et al.  found that household filtration did not significantly impact changes in CD4+ cell counts (P = 0.344).
Only one study examined nutrition, measured as weight-for-age z-scores (WAZ) for children less than 2 years born to HIV-positive mothers . This study found no impact of household filtration on mean WAZ scores (−1.21 vs. −1.24, respectively, P = 0.92) as a secondary outcome .
None of the reviewed studies reported on the impact of WASH interventions on environmental enteropathy.
Two studies reported mortality, although in neither case was it the primary outcome. Walson et al. reported that participants that received household filters and bednets were significantly less likely to die as a result of nontraumatic death or reach CD4+ cell counts of less than 350 cells/μl during the surveillance period (HR 0.74, 95% CI: 0.58–0.95, P = 0.02) . Peletz et al.  reported fewer deaths of children less than 2 years in the filtration intervention group, but these results were not significant (RR = 0.56; 95% CI: 0.13–2.37, P = 0.43).
Methodological quality of included studies
Methodological quality of studies and assessment criteria details are summarized in Table 2. The intervention was allocated randomly in six studies; the others compared the intervention group to a separate control group [37,38,40] and/or the same group before they received the intervention [37,39]. We classified two studies as blinded: Colford by design and the Sorvillo et al. study by virtue of the fact that participants were not likely to be aware of the filtration addition to the water treatment plant. Neither blinded study found significant health effects, although both were conducted in the United States where water supplies are generally of good quality. The methodological quality criteria was completely met for only one RCT (Colford study) and one nonrandomized controlled trial (Harris study). Four studies did not report on all criteria evaluated. Reported (subjective) outcomes were primary outcomes in five studies with nonblinded interventions, suggesting the potential of reporting bias. All studies reported on all outcomes with the exception of the Potgieter et al. study; results on diarrhoeal disease were not available at the time of this review.
We reviewed water, sanitation and hygiene interventions to prevent disease among PLHIV. Ten studies covering 13 281 individuals in six countries met the review's eligibility criteria. Nine assessed water quality interventions and one study assessed a handwashing intervention. Significantly, we identified no studies that assessed the impact of water supply or sanitation, two of the most fundamental WASH interventions.
Evidence from the eight studies that reported on diarrhoea suggests that water quality interventions and handwashing interventions may be protective among PLHIV. Notably, however, all but one of such studies relied on self-reported diarrhoea and the only blinded study found no statistically significant result. Thus, we cannot rule out the possibility that the effect is exaggerated .
All seven water quality studies that reported on diarrhoeal disease were of household water treatment, a water quality intervention reported to be effective in preventing diarrhoea [43,44]. Our pooled estimate of effect of 43% is within the range of estimates observed for household water treatment interventions, suggesting that the level of effectiveness among PLHIV is comparable to general populations. Although most interventions consisted of household water treatment, they included a variety of filtration and chlorination approaches that have different levels of efficacy against important opportunistic agents for PLHIV ; for example, chlorination does not inactivate Cryptosporidium spp. , a pathogen of particular concern for PLHIV .
Pooled estimates of the impact of the intervention on diarrhoea should be interpreted with caution due to important differences in the studies. Populations varied in terms of demographics, access to sanitation, water supplies, hygiene practices, viral load, access to antiretroviral drugs, and other factors. Adherence with household water treatment, a major factor affecting exposure and potential health impact , also varied among studies or was not measured at all. Differences in study design, case definitions, and the method of diarrhoea assessment limited the potential utility of pooled estimates of effect.
The one study evaluating a handwashing intervention among PLHIV reported the intervention to be effective against diarrhoea . The 58% reduction in risk exceeds the pooled estimate of a previous systematic review not focused on PLHIV (four trials, IRR 0.68, 95% CI: 0.52–0.90) ; however, it is not possible to conclude from this single study whether handwashing is more effective among PLHIV compared with the general population.
Despite this evidence of effectiveness on reported diarrhoea, results on other objective outcomes provide only limited evidence of a protective effect. The studies reporting on cryptosporidiosis and other enteric infections generally lacked significant findings, although studies could have been underpowered. Reductions in disease progression and nontraumatic death were reported in only one study , although as the intervention included provision of insecticide-treated nets, it is not possible to ascribe these results solely to the WASH (water filter) component.
Most studies included in this review presented issues of methodological quality. Four of the 10 studies employed a nonrandomized study design; populations assessed in different geographical locations and/or at different time periods may limit study population comparability. In addition to the issues noted above concerning self-reported outcomes (e.g. diarrhoeal disease) in nonblinded trials, some studies reporting on diarrhoea used longer diarrhoea recall periods that may be unreliable . The one blinded study that met all criteria of methodological quality did not report the water quality intervention to be protective against diarrhoea . However, this study was conducted in the United States wherein the potential impact may have been reduced due to generally higher levels of water quality. Most of the studies included in the review were small in scale with three studies had fewer than 100 HIV-positive participants. Results from pooling multiple small-scale studies should be interpreted with caution. Furthermore, the included studies had relatively short follow-up periods, a factor that has been shown to exaggerate the effect of WASH interventions .
In conclusion, the evidence of the health impact of WASH interventions among PLHIV is limited and mixed. Future studies should examine the impact of improved water supply and sanitation, two of the most fundamental of WASH interventions. Although blinding of most WASH interventions may be impossible, assessments should employ study designs and objective outcomes that minimize the risk of bias. They should also carefully measure compliance as a possible effect modifier and track the impact of the intervention on reducing pathogen exposure, a necessary condition for achieving health benefits from WASH interventions.
R.P. and T.C. designed the study. R.P. conducted all searches (except the Chinese databases), reviewed results for eligibility, extracted data from included studies, conducted the analysis, and assessed quality of the included studies. K.C. conducted database searches of the Chinese databases and reviewed subsequent results for eligibility. T.M. reviewed database results for study eligibility. M.H. and M.E. reviewed gray literature for eligibility. TC contacted agencies for grey literature, resolved any disagreements on eligibility, and helped with quality assessment. R.P. and T.C. wrote the paper. All authors contributed to drafts of this report and interpreted findings.
We would like to thank those that provided unpublished data and draft articles, especially Abebe, Harris, Potgieter, and Walson. Additionally, we thank Christian Jasper for providing information from her initial database searches on WASH and HIV integration.
This work was funded by Unilever, Ltd., University of North Carolina at Chapel Hill Water Institute and the University of North Carolina at Chapel Hill Institute for Global Health & Infectious Diseases.
Conflicts of interest
We declare that we have no conflicts of interest.
1. UNAIDS. Together we will end AIDS
. (UNAIDS) JUNPoHA, 2012.
2. Liu L, Johnson HL, Cousens S, Perin J, Scott S, Lawn JE, et al. Global, regional, and national causes of child mortality: an updated systematic analysis for 2010 with time trends since 2000
3. Mermin J, Lule J, Ekwaru JP, Malamba S, Downing R, Ransom R, et al. Effect of co-trimoxazole prophylaxis on morbidity, mortality, CD4-cell count, and viral load in HIV infection in rural Uganda
4. Lule JR, Mermin J, Ekwaru JP, Malamba S, Downing R, Ransom R, et al. Effect of home-based water chlorination and safe storage on diarrhea among persons with human immunodeficiency virus in Uganda
. Am J Trop Med Hyg
5. Stark D, Barratt JL, van Hal S, Marriott D, Harkness J, Ellis JT. Clinical significance of enteric protozoa in the immunosuppressed human population
. Clin Microbiol Rev
6. CAWST, Mahin T, Peletz R. Waterborne Pathogen Infection Rates for People Living with HIV/AIDS: A CAWST Literature Summary
7. Tumwine JK, Kekitiinwa A, Bakeera-Kitaka S, Ndeezi G, Downing R, Feng X, et al. Cryptosporidiosis and microsporidiosis in ugandan children with persistent diarrhea with and without concurrent infection with the human immunodeficiency virus
. Am J Trop Med Hyg
8. Carcamo C, Hooton T, Wener MH, Weiss NS, Gilman R, Arevalo J, et al. Etiologies and manifestations of persistent diarrhea in adults with HIV-1 infection: a case-control study in Lima, Peru
. J Infect Dis
9. Amadi B, Kelly P, Mwiya M, Mulwazi E, Sianongo S, Changwe F, et al. Intestinal and systemic infection, HIV, and mortality in zambian children with persistent diarrhea and malnutrition
. J Pediatr Gastroenterol Nutr
10. Kurniawan A, Karyadi T, Dwintasari SW, Sari IP, Yunihastuti E, Djauzi S, et al. Intestinal parasitic infections in HIV/AIDS patients presenting with diarrhoea in Jakarta, Indonesia
. Trans R Soc Trop Med Hyg
11. Mellors JW, Munoz A, Giorgi JV, Margolick JB, Tassoni CJ, Gupta P, et al. Plasma viral load and CD4+ lymphocytes as prognostic markers of HIV-1 infection
. Ann Intern Med
12. Humphrey JH. Child undernutrition, tropical enteropathy, toilets, and handwashing
13. Kelly P, Davies SE, Mandanda B, Veitch A, McPhail G, Zulu I, et al. Enteropathy in Zambians with HIV related diarrhoea: regression modelling of potential determinants of mucosal damage
14. Black RE, Morris SS, Bryce J. Where and why are 10 million children dying every year?
15. WHOInvesting in health research and development
. Geneva:World Health Organization; 1996.
16. Filteau S. The HIV-exposed, uninfected African child
. Trop Med Int Health
17. Peletz R, Simuyandi M, Sarenje K, Baisley K, Kelly P, Filteau S, et al. Drinking water quality, feeding practices, and diarrhea among children under 2 years of HIV-positive mothers in peri-urban Zambia
. Am J Trop Med Hyg
18. Fischer Walker CL, Friberg IK, Binkin N, Young M, Walker N, Fontaine O, et al. Scaling up diarrhea prevention and treatment interventions: a Lives Saved Tool analysis
. PLoS Med
19. WHO/UNICEFProgress on Drinking Water and Sanitation 2012 Update. Joint Monitoring Programme for Water Supply and Sanitation
. Geneva, Switzerland:WHO/UNICEF; 2012.
20. WHOEssential Prevention and Care Interventions for Adults and Adolescents Living with HIV in Resource-Limited Settings
. Geneva, Switzerland:World Health Organization; 2008.
21. US GovernmentReauthorizing PEPFAR. The United States President's Emergency Plan for AIDS Relief
. Washington, DC:US Government; 2008.
22. 110th Congress USG, Lantos T, Henry J. 110th Congress USGHyde United States Global Leadership Against AIDS, Tuberculosis, and Malaria Reauthorization Act of 2008
23. USAID, USAIDProgramming Guidance for Integrating Water, Sanitation, and Hygiene Improvement into HIV/AIDS Programs to Reduce Diarrhea Morbidity
24. USAID, CDC, USAID, CDCProgramming Water, Sanitation and Hygiene (WASH) Activities in U.S. Government Country Operational Plans (COPs): A Toolkit for FY2009 Planning
25. Water and Sanitation Program, WSPWater, Sanitation, and Hygiene for People Living with HIV and AIDS. Field Note
26. Water Supply & Sanitation Collaborative Council, Water Supply & Sanitation Collaborative CouncilHIV/AIDS & WASH. WSSCC Reference Note
27. Collaboration TC, Higgins J, Green S. Cochrane Handbook for Systematic Reviews of Interventions: Version 5.1.0
28. Baker SJ, Mathan VI. Tropical enteropathy and tropical sprue
. Am J Clin Nutr
29. CEPaOoC, Cochrane CollaborationRisk of bias. In: EPOC Author Resources
30. Sterne JA, Sutton AJ, Ioannidis JP, Terrin N, Jones DR, Lau J, et al. Recommendations for examining and interpreting funnel plot asymmetry in meta-analyses of randomised controlled trials
31. Rothman KJ, Greenland S, Lash TL. Modern epidemiology
. 3rd ed.Philadelphia:Wolters Kluwer Health/Lippincott Williams & Wilkins; 2008.
32. Colford J. A pilot randomized, controlled trial of an in-home drinking water intervention among HIV+ persons
. J Water Health
33. Huang DB, Zhou J. Effect of intensive handwashing in the prevention of diarrhoeal illness among patients with AIDS: a randomized controlled study
. J Med Microbiol
34. Abebe LS, Narkiewicz S, Oyandedel-Craver V, Conaway M, Singo A, Amidou S, et al.Ceramic water filters impregnated with silver nanoparticles as a point-of-use water-treatment intervention for HIV-positive individuals in limpopo province, South Africa: a pilot study of technological performance and human health benefits
. Submitted, 2012.
35. Potgieter N, Koekemoer R, Jagals P. Impacts of the provision of water, sanitation, hygiene and home based care services to HIV and AIDS infected people. WRC Report No KV 209/08
. Gezina, South Africa:Water Research Comission; 2008.
36. Peletz R, Simunyama M, Sarenje K, Baisley K, Filteau S, Kelly P, et al. Assessing water filtration and safe storage in households with young children of HIV-positive mothers: a randomized, controlled trial in Zambia
. PLoS One
37. Sorvillo F, Lieb LE, Nahlen B, Miller J, Mascola L, Ash LR. Municipal drinking water and cryptosporidiosis among persons with AIDS in Los Angeles County
. Epidemiol Infect
38. Walson J, Sangare L, Singa B, Naulikha J, Piper B, Yuhas K, et al. Evaluation of long-lasting insecticide-treated bed nets and a point-of-use water filter on HIV-1 disease progression in Kenya
39. Barzilay EJ, Aghoghovbia TS, Blanton EM, Akinpelumi AA, Coldiron ME, Akinfolayan O, et al. Diarrhea prevention in people living with HIV: an evaluation of a point-of-use water quality intervention in Lagos, Nigeria
. AIDS Care
40. Harris JR, Greene SK, Thomas TK, Ndivo R, Okanda J, Masaba R, et al. Effect of a point-of-use water treatment and safe water storage intervention on diarrhea in infants of HIV-infected mothers
. J Infect Dis
41. Brown J, Clasen T. High adherence is necessary to realize health gains from water quality interventions
. PLoS One
42. Wood L, Egger M, Gluud LL, Schulz KF, Juni P, Altman DG, et al. Empirical evidence of bias in treatment effect estimates in controlled trials with different interventions and outcomes: meta-epidemiological study
43. Clasen T, Schmidt WP, Rabie T, Roberts I, Cairncross S. Interventions to improve water quality for preventing diarrhoea: systematic review and meta-analysis
44. Waddington H, Snilstveit B. Effectiveness and sustainability of water, sanitation, and hygiene interventions in combating diarrhoea
. J Dev Effect
45. Korich DG, Mead JR, Madore MS, Sinclair NA, Sterling CR. Effects of ozone, chlorine dioxide, chlorine, and monochloramine on Cryptosporidium parvum oocyst viability
. Appl Environ Microbiol
46. Tzipori S, Widmer G. A hundred-year retrospective on cryptosporidiosis
. Trends Parasitol
47. Ejemot RI, Ehiri JE, Meremikwu MM, Critchley JA. Hand washing for preventing diarrhoea
. Cochrane Database Syst Rev
48. Schmidt WP, Arnold BF, Boisson S, Genser B, Luby SP, Barreto ML, et al. Epidemiological methods in diarrhoea studies--an update
. Int J Epidemiol
49. Hunter PR. Household water treatment in developing countries: comparing different intervention types using meta-regression
. Environ Sci Technol
Keywords:© 2013 Lippincott Williams & Wilkins, Inc.
AIDS; HIV; hygiene; sanitation; systematic review; water