The use of zero inflated modelling to identify high and low gastrointestinal nematode egg shedders in groups of horses

Jell, K., Barnes, T. S., Seddon, J. M., Coleman, G. T., Morton, J. M. and Kotze, A. (2008). The use of zero inflated modelling to identify high and low gastrointestinal nematode egg shedders in groups of horses. In: Proceedings of Australian Equine Science Symposium (AESS). Australian Equine Science Symposium, Gold Coast, Australia, (51-51). 4-6 June 2008.


Author Jell, K.
Barnes, T. S.
Seddon, J. M.
Coleman, G. T.
Morton, J. M.
Kotze, A.
Title of paper The use of zero inflated modelling to identify high and low gastrointestinal nematode egg shedders in groups of horses
Conference name Australian Equine Science Symposium
Conference location Gold Coast, Australia
Conference dates 4-6 June 2008
Proceedings title Proceedings of Australian Equine Science Symposium (AESS)
Publication Year 2008
Sub-type Oral presentation
Volume 2
Start page 51
End page 51
Total pages 1
Language eng
Abstract/Summary Anthelmintic resistance is widespread among cyathostomins infecting horses. The ability to identify the level of an individual horse’s immunity to cyathostomins might facilitate the development of novel approaches to worm control in this host. Horses on nine properties in south east Queensland were recruited for a study to test the hypothesis that certain individuals consistently contribute disproportionately to pasture contamination with these parasites. Faecal samples from approximately 56 horses were collected monthly for 12 months. As the FEC data were heavily skewed to the right and contained a high proportion of zero counts (48.5%), a zero-inflated negative binomial model was used to assess the associations between explanatory variables and FEC. Various explanatory variables (Table 1) were fitted separately in a univariable model in both the negative binomial and logistic parts of the model. Selected variables were progressively included in a multivariable model and retained in the final model when the likelihood ratio test returned a p-value < 0.05. 95% confidence intervals (CI) were calculated using the Huber/White estimate of variance to account for clustering by horse. All analyses were undertaken Stata Version 10 (StataCorp, College Station, TX, USA, 2007). The count and odds ratios for the final model are shown in Table 1. Treatment with an anthelmintic in the previous month and two months previously, breed and season all remained in the logistic part of the final model, whereas anthelminthic treatment 2 months previously, sex and breed all remained in the negative binomial component.
Subjects 0707 Veterinary Sciences
Q-Index Code EX
Q-Index Status Provisional Code
Institutional Status Unknown

Document type: Conference Paper
Collection: School of Veterinary Science Publications
 
Versions
Version Filter Type
Citation counts: Google Scholar Search Google Scholar
Created: Tue, 16 Mar 2010, 00:52:20 EST by Laura McTaggart on behalf of Faculty Of Nat Resources, Agric & Veterinary Sc