Decision support for trypanosomiasis control: An example using a geographic information system in eastern Zambia

In many african countries where government resources for livestock disease control are declining and donor aid toward the control of tsetse-transmitted trypanosomiasis is less forthcoming, there is an increasing need to identify areas where intervention is most likely to be technically, economically and environmentally sustainable. Activities must be focused in these areas so that the maximum benefits are obtained from limited resources. This paper provides an example of how geographic information systems can be used to identify areas of high priority, and presents a decision-support framework through which this can be achieved. Digital coverage were generated for six environmental variables: 1) cattle density, 2) human population density, 3) land designation, 4)relative arable potential, 5) crop use intensity and 6) proximity to existing control operations. The distribution of tsetse in the area was predicted using a multivariate (maximum likelihood) analysis of areas of known presence and absence and a series of environmental data. Experienced veterinarians and biologists working in the region established criteria weights for the input variables and the data were integrated using weighted linear combination to prioritise part of the common fly belt in Zambia for trypanosomiasis control. The results of this exercise, estimates of the errors involved and the implications that follow for achieving sustainable tsetse control are discussed. Through this exercise it was possible to make recommendations to the Regional Tsetse and Trypanosomiasis Control Programme (RTTCP), Zambia, regarding the optimal location of a community-based control programme in the Eastern Province.