An evaluation was conducted in 2007 on biosand filters that were installed in Posoltega, Nicaragua in 1999 and 2004.The objectives were to characterize the condition and use of filters eight and three years after installation, determine filter performance of those filters still in use, and identify determinants of successful long-term use and performance. Methods consisted of household identification, user questionnaires, and water quality testing. Of the 234 filters installed, only 24 were found to still be in use. Average log reductions were 1.73 (98%) for total coliforms, 1.36 (96%) for Escherichia Coli, and 0.91 (88%) for turbidity. Statistically significant effects were detected for the magnitude of the contamination of source water, the peak hydraulic loading rate, and the standing depth of water over the filter media. Questionnaire results indicated user training on filter maintenance could improve the peak hydraulic loading rate and hence filter performance. The low rate of sustained use (10%) is an indication of failed implementation, and is attributable to structural failure, particularly cracking of the concrete filters from 2004. Nonetheless, this evaluation demonstrated the biosand filter technology to be robust since those filters still in use were performing as expected three and eight years postimplementation.

A field evaluation of biosand filters, a method of household water treatment, was conducted in Posoltega, Nicaragua. The evaluation consisted of water quality testing for Escherichia coli and total coliforms using membrane filtration (MF) with m-coliBlue24 growth media, and SolarCult dipslides. For each sample tested, the MF procedure was applied to two volumes differing by an order of magnitude to capture the expected range of microbial contamination. A follow-up laboratory quality assurance/quality control procedure was undertaken to validate the field methods. The average coefficient of variation was 51%, with no statistically significant difference between the field and laboratory data. The MF procedure used with m-coliBlue24 produced useful reproducible results for the microbial concentrations encountered in the field. Dipslide results were found to correlate poorly to the membrane filtration results from the field, and should not be used to test drinking water due to the high limit of detection. They may still be useful for preliminary testing of highly contaminated source water as part of a household water treatment project. Dipslides may also be a useful tool for local health representatives to promote safe water practices within the community since they are simple to use and provide a visual indicator of water quality.