Subscribe to E-Bulletin Donate to CEGA

Household Water Treatment and Handwashing in Guatemala

Development Challenge

To promote overall health and to reduce the prevalence of diarrhea among young children in developing countries, health programs promote low-cost, behavior-change interventions to improve water quality at the point of use and to increase the practice of handwashing with soap.  Prior studies conducted at the time of intervention indicate that these measures may reduce diarrhea and acute respiratory infections amongst children under the age of five by 25 to 40 percent.1,2 However, few studies have examined the effectiveness of such programs after the conclusion of intervention activities.  This study contributes to the literature by evaluating the lasting health and behavioral effects of a combined water treatment and handwashing intervention in Guatemala six months after the project ended.

Context

Between 2003 and 2006, Caritas and Catholic Relief Services implemented a three-year household water treatment and handwashing campaign in approximately 90 villages across rural eastern Guatemala.  Caritas technicians recruited and trained community-based health promoters from each village.  These trained health promoters then visited households with children aged three years or under or with pregnant mothers to promote handwashing with soap and home water treatment, including boiling, solar disinfection, and chlorination using bleach.  These visits occurred monthly or bi-monthly and lasted for approximately 30 minutes.  A survey conducted at the end of the intervention showed that 70 percent of participating households self-reported regular use of some water treatment method.

Evaluation Strategy

To evaluate the campaign, researchers matched intervention villages with a group of comparison villages from the same region.  Using village-level data from a 2002 census, researchers used propensity score matching to identify 15 village pairs that were comparable demographically and by household characteristics including water and sanitation facilities.  The researchers randomly sampled 20 households from each of the 30 villages that had at least one child under age five and that had lived in the village since at least 2003 when the intervention began.  In the end, surveys were administered to 600 households containing 929 children under age five between April and June 2007.

The primary health outcomes of interest were diarrhea, respiratory infections, and child growth.  The survey measured diarrhea and respiratory infections for each child by asking caretakers to report any illness during the previous 48 hours.  Interviewers determined child growth using anthropometric techniques to measure height, weight, and mid-upper arm circumference. 

To measure household water treatment practices, respondents were asked to report their use of a treatment method during the previous seven days.  To confirm the practice of water treatment, researchers checked among those who self-reported use of a treatment method for the presence of materials used to treat water and for treated water in the home.  In addition, interviewers collected water samples from a random selection of intervention and comparison households for quality testing.  Finally, to measure handwashing practices, interviewers asked mothers when they washed their hands at critical times over the previous 24 hours.  They also discreetly spot-checked hygiene and water storage. 

Results and Policy Implications

Overall, the study found minimal sustained water treatment and positive handwashing behavior.  Six months after the end of the combined water treatment and handwashing intervention, confirmed water treatment was higher in intervention villages (8.7 percent) than in comparison villages (3.3 percent).  These figures are considerably lower than self-reported use of water treatment methods (33 percent vs. 21 percent, respectively), and represent a marked decline from the 70 percent estimate from the survey conducted at the end of the intervention.3 There was no difference between intervention and comparison villages in the quality of the tested water samples – nearly all contained E. coli

Intervention and comparison villages did not differ in self-reported handwashing behavior or spot-check observations of hygiene and water storage.  Intervention and comparison villages did not differ in diarrhea, respiratory infections, or anthropometric weight and height measures among children under age five.  The researchers note that the absence of child health impacts is consistent with the observed modest change in water treatment behavior, no detectable change in handwashing behavior, and highly contaminated living environments.

The researchers acknowledge various study limitations, including that only 49 percent of households in intervention villages reported participating in the intervention, which may have led to an underestimate of the intervention’s true effect.  However, when comparing only the subset of participating intervention households with non-participants, the conclusions remain the same.  This evaluation highlights the challenge of sustaining behavior-change water treatment and handwashing interventions.  The authors indicate that future research could more conclusively evaluate the sustainability of household water treatment programs by measuring outcomes at baseline, endline, and at points of time after the intervention concludes, since without an endline measurement it is impossible to differentiate between a scenario of high uptake / low sustainability and a scenario of low uptake. 

Timeline

2007

Ejemot, Regina I, Ehiri, John E, Meremikwu, Martin M, and Julia A Critchley, (2008), “Hand washing for preventing diarrhoea.” Cochrane Database Syst Rev, Issue 1.

Rabie, Tamer, and Valerie Curtis, (2006), “Handwashing and risk of respiratory infections: a quantitative systematic review.” Trop Med Int Health, Vol. 11, 258–67.

Among only those households that reported participation in the intervention, 37 percent report use of a water treatment method six months later.