Evaluation of the pathoscreen technique as a screening tool for measuring fecal contamination in point-of-use water treatment systems

Over the last 5 years, point-of-use water filters have been introduced into rural communities in the Dominican Republic to provide a reliable source of purified water. These low-cost, family-based systems consist of two five gallon buckets, a ceramic filter, and a spigot. An important aspect for the sustainability of the program is household visitation after a filter has been in operation several months. Filters are inspected, problems are identified, and water samples are taken to be tested for the presence of fecal bacteria by membrane filtration. Homes are revisited to show the owners bacteria cultures of their source and filtered water. This process is effective, but expensive. In this research, a qualitative technique, the pathoscreen method, which is 5 times less expensive, was evaluated as a screening tool. Positive indication of contamination by pathoscreen would be followed by membrane filtration as a confirmatory test.

Duplicate samples obtained from 185 filters were tested by both methods. Using E. coli counts greater than 5 cfu/100 mL as an indication of contamination, only 8% of the samples were determined to be contaminated. As a screening tool the pathoscreen method had sensitivity and specificity of 79%, a negative predictive value of 98%, and a positive predictive value of 23%. Because of the low sensitivity (i.e. the pathoscreen method misses 20 % of the samples determined to be positive by membrane filtration), the pathoscreen method is not an effective way to screen samples.