A novel spatial and stochastic model to evaluate the within and between farm transmission of classical swine fever virus: II validation of the model.

Abstract

A new, recently published, stochastic and spatial model for the evaluation of classical swine fever virus (CSFV) spread into Spain has been validated by using several methods. Internal validity, sensitivity analysis, validation using historical data, comparison with other models and experiments on data validity were used to evaluate the overall reliability and consistency of the model. More than 100 modifications in input data and parameters were evaluated. Outputs were obtained after 1000 iterations for each new scenario of the model. As a result, the model was shown to be consistent, being the probability of infection by local spread, the time from infectious to clinical signs state, the probability of detection based on clinical signs at day t after detection of the index case outside the control and surveillance zones and the maximum number of farms to be depopulated at day t the parameters that have more influence (>10% of change) on the magnitude and duration of the epidemic. The combination of a within- and between-farm spread model was also shown to give significantly different results than using a purely between-farm spread model. Methods and results presented here were intended to be useful to better understand and apply the model, to identify key parameters for which it will be critical to have good estimates and to provide better support for prevention and control of future CSFV outbreaks.