Comparison of methods for estimating drug coverage for filariasis elimination, Leogane Commune, Haiti

Abstract

In the global effort to eliminate lymphatic filariasis, annual mass treatments are conducted with diethylcarbamazine (DEC) or ivermectin, combined with albendazole. The success of this strategy depends on achieving high levels of drug coverage, which reduces the number of persons with circulating microfilariae so that transmission of the parasite is interrupted. Because resources are often limited, a simple, inexpensive, and reliable method to estimate drug coverage is needed. During the period December 2000 to February 2001, three methods were used to assess drug coverage in Leogane Commune, Haiti: a probability survey using a cluster sample design (n = 1421 persons); a distribution-point survey based on a convenience sample of houses near the distribution points (n = 4341 persons); and a survey based on a convenience sample of primary schools (n = 5036 children). The coverage estimations were 71.3% (95% CI 66.7-75.9), 73.6% (95% CI 70.1-77.0), and 77.8% (95% CI 73.5-82.1), respectively. Survey costs for the probability, distribution point, and school surveys were US$2217, US$979, and US$312, respectively. The 2 convenience sampling methods provided point estimates of drug coverage that were similar to those of the probability survey. These methods may have a role for monitoring drug treatment coverage between less frequent, but more costly, probability sample surveys.