Integrated cost-benefit analysis of tsetse control and herd productivity to inform control programs for animal African trypanosomiasis.

Animal African trypanosomiasis (AAT) and its tsetse vector are responsible for annual losses estimated in billions of US dollars ($). Recent years have seen the implementation of a series of multinational interventions. However, actors of AAT control face complex resource allocation decisions due to the geographical range of AAT, diversity of ecological and livestock systems, and range of control methods available.The study presented here integrates an existing tsetse abundance model with a bio-economic herd model that captures local production characteristics as well as heterogeneities in AAT incidence and breed. These models were used to predict the impact of tsetse elimination on the net value of cattle production in the districts of Mambwe, in Zambia, and Faro et Déo in Cameroon. The net value of cattle production under the current situation was used as a baseline, and compared with alternative publicly funded control programmes. In Zambia, the current baseline is AAT control implemented privately by cattle owners (Scenario Z0). In Cameroon, the baseline (Scenario C0) is a small-scale publicly funded tsetse control programme and privately funded control at farm level. The model was run for 10 years, using a discount rate of 5%.Compared to Scenario C0, benefit-cost ratios (BCR) of 4.5 (4.4-4.7) for Scenario C1 (tsetse suppression using insecticide treatment of cattle (ITC) and traps + maintenance with ITC barrier), and 3.8 (3.6-4.0) for Scenario C2 (tsetse suppression using ITC and traps + maintenance with barrier of targets), were estimated in Cameroon. For Zambia, the benefit-cost ratio calculated for Scenarios Z1 (targets, ITC barrier), Z2 (targets, barrier traps), Z3 (aerial spraying, ITC barrier), and Z4 (aerial spraying, barrier traps) were 2.3 (1.8 - 2.7), 2.0 (1.6-2.4), 2.8 (2.3-3.3) and 2.5 (2.0-2.9), respectively. Sensitivity analysis showed that the profitability of the projects is relatively resistant to variations in the costs of the interventions and their technical efficiency.It is envisioned that the methodologies presented here will be useful for the evaluation and design of existing and future control programmes, ensuring they have tangible benefits in the communities they are targeting.