Summary

This case study describes how tsetse control interventions were planned to see whether trypanosome prevalence in village cattle could be reduced in an area in south western Ethiopia where there was widespread drug resistance to all available trypanocidal drugs. The study was organised as a longitudinal monitoring study conducted over periods without tsetse control followed by periods with tsetse control. Two methods were applied: (1) a set of insecticide-impregnated black-cloth targets that attract and kill tsetse (see photograph), and (2) an insecticidal pour-on applied to the backs of cattle once a month. The latter method involved participation by farmers who decided when to bring cattle for treatment and which animals to bring.

The case study describes how it was impossible to study two separate areas simultaneously for this type of research, one area with and one without tsetse control. It shows instead how researchers were required to design a 'before-after' study with an intervention following a period without control.

Such trials are beset with the problems of confounding, however, with some of the variation occurring from one year to the next being associated with changes unconnected with tsetse control.

Interpretation of the results of such studies therefore needs to be made with caution.