SIMULATION-MODEL OF PARATUBERCULOSIS CONTROL IN A DAIRY-HERD

A simple simulation model of the control of paratuberculosis in a dairy herd was developed using a spreadsheet program. Reed-Frost methods were used to calculate new infections. Age-specific culling rates for infected animals (indexed to the annual herd replacement rate) were used as Markov chain transition probabilities for calculating culling probabilities. The predictive value model was used as the means of determining the probability that diagnostic tests would correctly identify infected animals for culling. This deterministic probability model was validated with previously published data from a 5 year field study on the control of paratuberculosis in an 80 cow dairy herd. Sensitivity analysis of the model variables was performed. Calf-management techniques that reduce the number of effective cow-calf contacts (k3) decreased the prevalence of paratuberculosis in a herd at a rate in direct relation to the value of k3. Test-and-cull control measures also decreased the prevalence of disease in the herd; test sensitivity was the sole determinant of the effectiveness of this control strategy. Use of both calf-management and test-and-cull methods provided the quickest means of controlling paratuberculosis (which is consistent with reports in the literature). The model provides a simple method of predicting the rate at which paratuberculosis can be controlled or eliminated from a dairy herd and has the potential of being used for the economic evaluation of control programs. Because of the model's flexibility, it can be used to examine a variety of strategies for specific herds.