The predicted and observed decline in onchocerciasis infection during 14 years of successful control of Simulium spp. In West Africa
Since it began in 1975, the strategy of the Onchocerciasis Control Programme in West Africa (OCP) has been to reduce transmission of the parasite Onchocerca volvulus to insignificant levels by means of vector control, and to maintain these levels for sufficiently long to allow the initial reservoir of the parasite to fall so low that vector control can be safely interrupted [1[.
The period of successful vector control required depends primarily on the duration of onchocerciasis infection, which is a function of the reproductive lifespan of the adult O. volvulus and the longevity of their microfilariae. In 1975, little was known about the duration of onchocerciasis infection. The only relevant information was from East Africa, where cross-sectional skin snip surveys in three foci showed that active onchocerciasis infection still persisted 9-11 years after the elimination of the vector, while there was no active infection in another focus 18 years after interruption of transmission . As a result, vector control in the OCP area was originally planned to last for a period of 20 years .
Since 1975, epidemiological follow-up surveys have been undertaken in selected indicator villages from all the major river basins in the OCP area. The objective of these surveys was to evaluate the impact of vector control on transmission and on disease, and to document the decline in infection levels in the human population. For long-term planning and funding of OCP, it was important to arrive at accurate estimates of the reproductive lifespan of O. volvulus and to predict how many years of vector control were still needed before the parasite reservoir would fall to an insignificant level. Epidemiological modeling was used to evaluate the data. The first step was the development of a force-infection model for onchocerciasis  that facilitated a better understanding of age-specific epidemiological trends and resulted in new statistical methods for the comparative analysis of the epidemiological data. The use of this model led also to a preliminary estimate of 11 years for the average duration of onchocerciasis infection.
The force-of-infection model was based on certain simplifications, notably the assumption that the duration of onchocerciasis infection was constant, which made it inappropriate for the prediction of the final decline in the parasite reservoir. A more sophisticated model, based on the simulation of individual life-histories of human hosts and adult female worms, was therefore developed. The first predictions with this host-parasite model were made in 1985, and these indicated that onchocerciasis infection should be virtually eliminated 15 years after interruption of transmission. (a) Since then, the model has been used to analyse and evaluate various epidemiological trends [5-7].
From December 1988 to February 1989, OCP carried out epidemiological evaluations in indicator villages from the original programme area, which had been under vector control for 12-14 years. Because of thepredictions made by the host-parasite model, it was expected that the results would provide important information on the impact of vector control on the parasite reservoir and on the reproductive lifespan of O. volvulus. Here, we report the trends observed for the intensity and prevalence of infection during 12-14 years of successful vector control and compare them with the predictions that were made in 1985 using the host-parasite model.
Materials and methods
Skin snip surveys
In all epidemiological surveys in the OCP area, two skin snips are taken from the iliac crest of each person covered. The snips are incubated in distilled water for 30 minuteS, subsequently examined microscopically for the presence of 0. …