There is a growing consensus among practitioners involved in the research and control of trypanosomiasis that the people affected by this disease should be actively involved in the implementation of control plans (Laveissière et al., 1990; Ssennyonga, 1998). This involvement can only be successful if it is based on a sound understanding of the communitys social and cultural settings. Ssennyonga (1998) suggested the collection of four types of baseline information necessary to understand the social and cultural settings of the community that will participate in T&T control:
local modes of adaptation and spatial dimension of production systems; and
information on indigenous knowledge and management of T&T.
It is important to have reliable knowledge of the basic demographic structure and trends of a target population participating in a T&T control programme. This involves, when necessary, taking a census of populations and livestock-holding households at intervals. The information is particularly useful when projecting the population and stock size in relation to the evolution of trypanosomiasis risk. Kamuanga and colleagues (2001b) have shown evidence of a relationship between the variations in the size of the human population and the number of cattle-holding households in response to changes in the risk of trypanosomiasis in a settlement of pastoralists in Burkina Faso.
The need for a better understanding of the social organization of production beyond knowledge of lineage linkages is of prime importance. At this stage, collaboration between economists and socio-anthropologists is needed for the design of support systems. It is the social scientists job to anticipate the emergence of incompatible organizational forms (Goodell, 1984) and help minimize the social costs of participation in T&T control operations. On the other hand, economists should use such conceptual tools as unit of production, consumption groups, level of accumulation, and role and status within households - tools commonly used by social scientists - because these are helpful in capturing the structure and dynamics of family groups as production units. Land-tenure disputes and farmer-herder conflicts can also be anticipated with a sound understating of local struggles over values, claims to status, power and scarce resources (e.g. those arising from the use of land cleared of tsetse) and their articulation into conflicts of interests between social groups (Hagberg, 1998). Thus T&T will affect nomadic pastoralists differently from settled agropastoralists and crop farmers.
Information on the location of homesteads, grazing areas and watering points is essential in order to identify risk zones at the level of the village or small region. In general, new techniques for modelling of T&T risk, using GIS and a variety of other information systems to gather data on vegetation, soil quality, water points and production systems, are becoming key factors in the identification of high-risk zones.
An important aspect of the baseline data is the information on indigenous knowledge and management of T&T from the perspectives of the beneficiaries that can reveal the size of gaps in knowledge. In several parts of Africa, local explanation of trypanosomiasis causality is embedded in beliefs. These can be in the form of superstitions, magic or witchcraft, myths, taboos and religion. The common explanation of trypanosomiasis epidemics in many study sites was witchcraft (Swallow and Woudyalew, 1994; Mwangi, Swallow and Roderick, 1998; Echessah et al., 1997). A study in Burkina Faso indicated that trypanosomiasis was unknown to most Fulani herders before they settled in the southern region. Indeed, on average they only identified 3.7 out of the eight most important symptoms of trypanosomiasis (Kamuanga et al., 1997). In Côte dIvoire the results of a survey have shown that only 12 percent of respondents could establish a link between tsetse fly and the disease; most respondents related trypanosomiasis to mosquito bites and witchcraft (Dagnogo, Yapi and Koné, 1997).
The next question one should ask is how to collect the required qualitative information. Several approaches are available in the form of participatory rapid appraisal techniques including key informant interviews, community profile analysis, focus groups, participatory mapping and social drama (Swallow and Woudyalew, 1994; Echessah et al., 1997; Mwangi, Swallow and Roderick, 1998). Formal household surveys using a structured questionnaire are also frequently used. In all case situations, however, it is important to provide a time frame for delivering the results and interpretation of baseline data on people affected by T&T in order to make them available at the very time of design and implementation of tsetse control with community participation.
Quantitative surveys are instituted to ascertain the significance of socio-economic and behavioural variables identified in the preliminary qualitative stage. Cross-sectional surveys should be initiated to incorporate a willingness to pay assessment using the technique of contingent valuation to elicit respondents information on the quantity of resources that they are willing to commit to a particular activity. The contingent valuation technique was initially developed in the United States of America to value environmental public goods (Mitchell and Carson, 1989). The best-documented use of this technique to date is its application to water use in sub-Saharan Africa (Whittington, Lauria and Mu, 1991; Boadu, 1992). Swallow and Woudyalew (1994), Kamara and Echessah (1994) and Kamuanga and colleagues (1997) have used the technique to evaluate the willingness of livestock farmers in study sites in Ethiopia, Kenya and Burkina Faso to contribute to tsetse control using traps and targets. Thus, contingent valuation can be used to determine the quantity of time and labour that people may be willing to contribute to community-based T&T programmes and it can provide a comparative database from which actual contributions of resources can be measured (Kamuanga et al., 2001a). It can also be used to examine household characteristics that influence decisions to contribute financial and/or time resources to T&T programmes and how changes in these characteristics will affect future demand for the goods and services offered (Whittington et al., 1990).