A protocol to identify and minimise selection and information bias in abattoir surveys estimating prevalence, using Fasciola hepatica as an example.