The use and performance of BioSand filters in the Artibonite Valley of Haiti: a field study of 107 households.

INTRODUCTION: Approximately one billion people world-wide lack access to adequate amounts of safe water. Most are in developing countries, especially in rapidly expanding urban fringes, poor rural areas, and indigenous communities.
METHODS: In February and March 2005, a field study of 107 households was conducted to evaluate the use and performance of the Manz BioSand filter in the Artibonite Valley of Haiti. Approximately 2000 filters had been installed in this area over the preceding 5 years by the staff in Community Development at Hospital Albert Schweitzer, Deschappelle, Haiti. Interviews, observations, and water samplings were carried-out by two teams of Haitian enumerators, each consisting of a nurse and a filter technician. Water analyses were performed by Haitian lab technicians using the membrane filtration method to determine Escherichia coli counts. The enumerators and the lab technicians completed a 2 week training program before beginning the study; they worked under the direct supervision of the primary investigator. Laboratory quality was monitored by running 10% blank and 10% duplicate samples.
RESULTS: The households contained an average of 5.4 persons. Filters had been in use for an average of 2.5 years, and participants were generally satisfied with their filter's performance. Shallow, hand-dug wells provided the only source of water for 61% of the households, with 26% using water piped from springs or deep wells, and 13% having access to both. Only 3% had plumbing in their homes. Source water from shallow wells contained an average of 234 E. coli cfu/100 mL. Piped sources averaged 195 E. coli cfu/100 mL. Of the source water samples 26% contained 0-10 E. coli cfu/100 mL. Of the filtered water samples 97% contained 0-10 E. coli cfu/100 mL (80% with 0 cfu/100 mL, and 17% with 1-10 cfu/100 mL). Overall bacterial removal efficiency for the filters was calculated to be 98.5%. Turbidity decreased from an average of 6.2 NTU in source water samples to 0.9 NTU in the filtered water. None of the households treated the water after filtering; 91% used the filtered water only for drinking. No problems related to filter construction were observed; 13% were found to have significantly decreased flow rates (all restored by cleaning the filter). Recontamination was found to occur, with only 3% of the samples from the filters' spouts containing >10 E. coli cfu/100 mL and 22% of the stored filtered water samples at point-of-use containing >10 cfu/100 mL.
CONCLUSION: The Manz BioSand filters are an attractive option for supplying water treatment to family units in rural areas of poorly developed countries.