Preserving new anthelmintics: A simple method for estimating faecal egg count reduction test (FECRT) confidence limits when efficacy and/or nematode aggregation is high

Abstract

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.Page 1 As it has been 30 years since a new anthelmintic class was released, it is appropriate 22 to review management practices aimed at slowing the development of anthelmintic 23 resistance to all drug classes. Recommendations to delay anthelmintic resistance and 24 provide "refugia" are reviewed and a simulation model used to find optimum 25 treatment strategies that maintain nematode control. Simulated Australian conditions 26 indicated that a common successful low-risk treatment program was a rapid rotation 27 between a "triple-combination" product (benzimidazole + levamisole + abamectin) 28 and a new high-efficacy drug (monepantel). Where Haemonchus contortus was a 29 threat, moxidectin was required at critical times because of its persistent activity 30 against this parasite. Leaving up to 4% of adult sheep untreated provided sufficient 31 "refugia" for non-selected worms to reduce the risk of selecting for anthelmintic 32 resistance without compromising nematode control. 33For a new anthelmintic, efficacy estimated by faecal egg count reduction (FECR) is 34 likely to be at or close to 100%, however using current methods the 95% confidence 35 limits (CL) for 100% are incorrectly determined as 100%. The fewer eggs counted 36 pre-treatment, the more likely an estimate of 100% will occur, particularly if the true 37 efficacy is >90%. A novel way to determine the lower-CL (LCL) for 100% efficacy is 38 to reframe FECR as a binomial proportion, i.e. define: n and x as the total number of 39 eggs counted (rather than eggs per gram of faeces) for all pre-treatment and post-40 treatment animals, respectively; p the proportion of resistant eggs is p=x/n and 41 percent efficacy is 100*(1-p) (assuming equal treatment group sizes and detection 42 levels, pre-and post-treatment). The LCL is approximated from the cumulative 43 inverse beta distribution by: 95%LCL=100*(1-(BETAINV(0.975,x+1,n-x+1))). This 44 method is simpler than the current method, independent of the number of animals 45Page 3 of 47 A c c e p t e d M a n u s c r i p t 3 tested, and demonstrates that for 100% efficacy at least 37 eggs (not eggs per gram) 46 need to be counted pre-treatment before the LCL can exceed 90%. When nematode 47 aggregation is high, this method can be usefully applied to efficacy estimates lower 48 than 100%, and in this case the 95% upper-CL (UCL) can be estimated by: 49 95%UCL=100*(1-(BETAINV(0.025,x+1,n-x+1))), with the LCL approximated as 50 described above. A simulation study to estimate the precision and accuracy of this 51 method found that the more conservative 99%CL...

Evaluating faecal egg count reduction using a specifically designed package “eggCounts” in R and a user friendly web interface

“…In this case the UI of the percentage reduction are 98.0% to 100%. The method previously described by Dobson et al (2012) attempts to address the problem of uncertainty intervals for such observations with a mean of zero epg, an issue earlier techniques such as the RESO method cannot address (Excel version of RESO available at http://sydney.edu.au/vetscience/sheepwormcontrol/software/FECR4.xls).…”

“…The proposed solution of Dobson et al (2012) uses the raw observed egg count and the inverse beta distribution. However this method is unable to generate rational UI for the FECR when the efficacy is poor or if the observed post-treatment egg counts are higher than pre-treatment counts.…”

“…This is clearly incorrect as zero epg observed following treatment does not prove there are no eggs; it merely indicates that no eggs were seen. This problem is discussed further in Dobson et al (2012).…”

Erratum to: Comparison of calculation methods used for the determination of anthelmintic resistance in sheep in a temperate continental climate

“…The timeliness and relevance of our results are indicated by the recent literature published on the topic (Denwood et al 2010;Levecke et al 2011;Dobson et al 2012;McKenna 2013), highlighting the urgency to reach a consensus as to which calculation method should be used. Shortly after the World Association for the Advancement of Veterinary Parasitology (WAAVP) endorsed FECR 4 as the standardized method of calculation (Coles et al 1992), a second set of WAAVP-endorsed guidelines, recommending the use of geometric means, were published by Wood et al (1995).…”

“…Shortly after the World Association for the Advancement of Veterinary Parasitology (WAAVP) endorsed FECR 4 as the standardized method of calculation (Coles et al 1992), a second set of WAAVP-endorsed guidelines, recommending the use of geometric means, were published by Wood et al (1995). The latter guidelines were criticized by McKenna (1997) and Van Wyk and Groeneveld (1997), and since that time there has been an ongoing discussion as to which method is more appropriate (Denwood et al 2010;Dobson et al 2012).…”