Household based treatment of drinking water with flocculant-disinfectant for preventing diarrhoea in areas with turbid source water in rural western Kenya : cluster randomised controlled trial.

Crump, J.A., Otieno, P.O., Slutsker, L., Keswick, B.H., Rosen, D.H., Hoekstra, R.M., Vulule, J.M. and Luby, S.P. (2005) BMJ, 331 (7515); 478.

It is estimated that there are 2.2 million deaths a year caused by diarrhoeal diseases worldwide. The importance of household based water treatment to reduce diarrhoeal disease in low income countries which lack access to clean water is increasingly being recognised.

A study was undertaken in rural western Kenya where source waters are both heavily faecally contaminated and very turbid. The study was conducted over 20 weeks. It was hypothesised that children less than 2 years of age living in family compounds (restricted plots containing several houses of family members) that received a new flocculant-disinfectant would have less episodes of diarrhoea than children in compounds using sodium hypochlorite (a widely used household based disinfectant). The use of both of these interventions was compared with the usual water handling practices to assess the effect of water treatment on diarrhoea among household residents of all ages and to assess the relative suitability of the two interventions for treating highly turbid water.

There were 605 family compounds identified from 49 villages near Lake Victoria in Siaya and Bondo Districts in western Kenya with at least one child aged less than 2 years. Family compounds were randomly assigned to either flocculant-disinfectant (201 families), sodium hypochlorite (203 families) or a control group (201 families). Flocculant-disinfectant is a new technology using single use sachets which work quickly on small volumes of water by aggregating and facilitating the removal of suspended organic matter, bacteria, viruses, parasites and heavy metals in treated water. One packet contains enough calcium hypochlorite to leave a residual chlorine concentration of 3.5 mg/L in 10L of demineralised water. The sodium hypochlorite treatment used 1 % sodium hypochlorite manufactured commercially. The control group continued their usual water collection, treatment and storage practices. These practices included turbidity mitigation by setting and decanting, cloth filtration and treatment with alum to modestly reduce turbidity.

Field workers visited participants weekly and used a questionnaire to record the presence or absence of diarrhoea and any deaths during the seven days since the last visit. A questionnaire was also used to assess the mothers' knowledge of and attitudes towards the interventions during the 5 th and 15 th week of the study. Field workers collected samples of stored drinking water during the baselines survey and during unannounced visits every four weeks to measure free chlorine concentrations and turbidity as well as samples of source water to measure turbidity. The concentration of Escherichia coli was measured during the baseline survey and the 10 th week of the study from stored drinking water and from source water.

When children under 2 years old were compared with those in the control group, the absolute difference in prevalence of diarrhoea was 25% less for the flocculant-disinfectant users (95% CI -40 to -5) and 17% less in those using the sodium hypochlorite (-34 to 4). When all age groups were compared with the control group, the absolute difference in prevalence was reduced by 19% in the flocculant-disinfectant users (- 34 to -2 ) and by 26% in those using the sodium hypochlorite (- 39 to -9 ). It was found that there were significantly less deaths in the intervention groups than in the control group (relative risk of death 0.58, P=0.036). E. coli concentrations less than 1 CFU/100 ml were found in 14% of water samples from the control group compared with 82% in the flocculant-disinfectant users and 78% in the sodium hypochlorite users. The mean turbidity was found to be 8 NTU in the flocculant-disinfectant users compared with 55 NTU in the other two groups.

This study showed that flocculant-disinfectant was well suited to areas with highly contaminated and turbid water and provided a significant health benefit by reducing the risk of diarrhoea particularly among infants and children.