Economics of Brucella ovis control in sheep: computerized decision-tree analysis

The epidemiology and economics of Brucella ovis control in a hypothetical, commercial sheep flock (100 rams and 2,500 ewes) were investigated. The investigation consisted of an epidemiologic simulation model, reported in a companion paper, and a decision-tree analysis, reported here. It was predicted from the simulation model that B ovis could be eradicated successfully in 2 test periods (less than 1 year) from a flock by using intensive screening and culling. A computerized decision-tree program was used to determine the economically optimal control strategy among several alternatives. Two versions of the program were used to determine the optimal alternative, based on minimizing the expected monetary loss (deterministic) and minimizing the associated risk (stochastic). The economically optimal alternative was to screen the rams by means of palpation, semen testing, and ELISA prior to the mating season. Rams positive to any test were culled. After the mating season was completed, the optimal action was to use ELISA for the remaining rams and to cull all that were ELISA positive. The cost of this alternative was approximately $6,150, or less than one half the annual cost of a vaccination program ($12,800) or no program ($13,550). Continuing palpation and semen testing were considered worthwhile on the basis of detecting new cases of B ovis infection and in maintaining high flock fertility. Similarly, the cost of annual use of ELISA was small (approximately $100), compared with the potential cost of not detecting a new case of B ovis infection.