Assessing water filtration and safe storage in households with young children of HIV-positive mothers: a randomized, controlled trial in Zambia

Abstract

Background: Unsafe drinking water presents a particular threat to people living with HIV/AIDS (PLHIV) due to the increased risk of opportunistic infections, diarrhea-associated malabsorption of essential nutrients, and increased exposure to untreated water for children of HIV-positive mothers who use replacement feeding to reduce the risk of HIV transmission. This population may particularly benefit from an intervention to improve water quality in the home.

Methods and findings: We conducted a 12-month randomized, controlled field trial in Zambia among 120 households with children <2 years (100 with HIV-positive mothers and 20 with HIV-negative mothers to reduce stigma of participation) to assess a high-performance water filter and jerry cans for safe storage. Households were followed up monthly to assess use, drinking water quality (thermotolerant coliforms (TTC), an indicator of fecal contamination) and reported diarrhea (7-day recall) among children <2 years and all members of the household. Because previous attempts to blind the filter have been unsuccessful, we also assessed weight-for-age Z-scores (WAZ) as an objective measure of diarrhea impact. Filter use was high, with 96% (596/620) of household visits meeting the criteria for users. The quality of water stored in intervention households was significantly better than in control households (3 vs. 181 TTC/100 mL, respectively, p<0.001). The intervention was associated with reductions in the longitudinal prevalence of reported diarrhea of 53% among children <2 years (LPR=0.47, 95% CI: 0.30-0.73, p=0.001) and 54% among all household members (LPR=0.46, 95% CI: 0.30-0.70, p<0.001). While reduced WAZ was associated with reported diarrhea (-0.26; 95% CI: -0.37 to -0.14, p<0.001), there was no difference in WAZ between intervention and control groups.

Conclusion: In this population living with HIV/AIDS, a water filter combined with safe storage was used correctly and consistently, was highly effective in improving drinking water quality, and was protective against diarrhea.

Trial registration: Clinicaltrials.gov NCT01116908.

Figure 1. Recruitment flow diagram.

Figure 2. Water quality testing results.

Unfiltered water is for all households; filtered and stored filtered is only for the intervention arm.

Figure 3. Longitudinal prevalence of diarrhea in children <2 years and all household members.

Data for July 2010 are grouped with August 2010, due to follow-up visits commencing the final week of July.

Figure 4. Water quality and diarrhea in children <2 years.

Water quality is of stored drinking water (stored filtered water for intervention households and unfiltered water for control households). If intervention households did not have stored filtered water available, it was assumed they were drinking unfiltered water. Both analyses are adjusted for age; adjusting for trial arm is examined separately due to the partial collinearity between trial arm and water quality. Predicted probabilities of diarrhea are from unadjusted and adjusted binomial regression models with log link functions and robust standard errors with GEE to account for repeated measures. Error bars represent 95% confidence intervals. Unadjusted model coefficients: ln(diarrhea prevalence) = −1.25+0.186(log₁₀TTC)+−0.0991(child's age). Adjusted model coefficients: ln(diarrhea prevalence) = −0.868+0.0825(log₁₀ water quality)+−0.0990(child's age)+−0.506 (trial arm).