B. Levecke, R.J. Dobson, N. Speybroeck, J. Vercruysse, J. Charlier

The faecal egg count reduction test (FECRT) is the method of choice to monitor anthelmintic efficacy against gastro-intestinal nematodes in livestock. Guidelines on how to conduct a FECRT are made available by the World Association for the Advancement of Veterinary Parasitology (WAAVP). Since the publication of these guidelines in the early 1990 s, some limitations have been noted, including (i) the ignorance of host-parasite interactions that depend on animal and parasite species, (ii) their feasibility under field conditions, (iii) appropriateness of study design, and (iv) the high detection limit of the recommended faecal egg count (FEC) method. Therefore, the objective of the present study was to empirically assess the impact of the level of excretion and aggregation of FEC, sample size and detection limit of the FEC method on the sensitivity and specificity of the FECRT to detect reduced efficacy (<90% or <95%) and to develop recommendations for surveys on anthelmintic resistance. A simulation study was performed in which the FECRT (based on the arithmetic mean of grouped FEC of the same animals before and after drug administration) was conducted under varying conditions of mean FEC, aggregation of FEC (inversely correlated with k), sample size, detection limit and 'true' drug efficacies. Classification trees were built to explore the impact of the above factors on the sensitivity and specificity of detecting a truly reduced efficacy. For a reduced-efficacy threshold of 90%, most combinations resulted in a reliable detection of reduced and normal efficacy. For the reduced-efficacy threshold of 95% however, unreliable FECRT results were found when sample sizes <15 were combined with highly aggregated FEC (k=0.25) and detection limits ≥ 5 EPG or when combined with detection limits ≥ 15 EPG. Overall, an increase in sample size and mean preDA FEC, and a decrease in detection limit improved the diagnostic accuracy. FECRT remained inconclusive under any evaluated condition for drug efficacies ranging from 87.5% to 92.5% for a reduced-efficacy-threshold of 90% and from 92.5% to 97.5% for a threshold of 95%. The results highlight that (i) the interpretation of this FECRT is affected by a complex interplay of factors, including the level of excretion and aggregation of FEC and (ii) the diagnostic value of FECRT to detect small reductions in efficacy is limited. This study, therefore, provides a framework allowing researchers to adapt their study design according to a wide range of field conditions, while ensuring a good diagnostic performance of the FECRT.