Statistical inference for ectoparasiticide efficacy in animal trials

Abstract

In controlled animal trials of ectoparasiticides the e cacy of treatments is esti- mated based on the number of surviving parasites with which experimental animals have been infected. Guidelines for the conduct and analysis of animal trials pub- lished by regulatory authorities require that the e cacy of the test treatment (as determined by the Abbott formula) should be at least 90%, for the test treatment to be declared e cacious. This decision rule, therefore, is simply based on a point estimate of e cacy and does not take into account the precision of the estimate; speci cally, proper statistical inference on the e cacy of the test treatment in question is not required. As a consequence, the Type I error probability of falsely declaring a non-e cacious product to be e cacious can be overin ated. In the proposed research project we investigate the use of appropriate statistical decision rules for the e cacy which control the Type I error at a speci ed low level, say 5%. The statistical model for the data assumes a beta-binomial distribution which can accommodate the binomial overdispersion typically associated with such data. A Bayesian approach for implementing the analysis of ectoparasiticide e cacy data is explored.