USE AND VALIDATION OF A NOVEL MARKOV CHAIN MONTE CARLO METHOD OF ANALYSIS FOR FAECAL EGG COUNT REDUCTION TEST DATA

摘要

Faecal Egg Count Reduction Test (FECRT) data are frequently characterised by high variability, small sample sizes and frequent zero observations. Accurate analysis of the data therefore depends on the use of appropriate statistical techniques. Analyses of simulated FECRT data by methods based on calculation of the empirical mean and variance, non-parametric bootstrapping, and a novel Markov chain Monte Carlo (MCMC) method are compared. The MCMC method consistently outperformed the other methods, independently of the distribution from which the data were generated. Notional 95% confidence intervals produced by non-parametric bootstrapping contained the correct value between only 84% and 90% of the time, compared to between 92% and 97% of the time for the MCMC method. Computationally intensive parametric techniques such as MCMC should therefore be used for analysis of these kinds of data in order to avoid making erroneous inference about the true efficacy of anthelmintics in the field.