Fit of the zero-inflated negative binomial model to analyze fecal egg counts

Authors: GUNES, HILAL - University Of Nebraska; HOWARD, REKA - University Of Nebraska; FUDOLIG, MIGUEL - University Of Nevada Las Vegas, Las Vegas, Nv; Burke, Joan; LEWIS, RONALD - University Of Nebraska

Submitted to: American Society of Animal Science
Publication Type: Abstract Only
Publication Acceptance Date: 6/10/2023
Publication Date: 11/6/2023
Citation: Gunes, H.Y., Howard, R., Fudolig, M., Burke, J.M., Lewis, R. 2023. Fit of the zero-inflated negative binomial model to analyze fecal egg counts. American Society of Animal Science. https://doi.org/10.1093/jas/skad281.626.
DOI: https://doi.org/10.1093/jas/skad281.626

Interpretive Summary: Fecal egg count (FEC) is used as an indicator of parasite infection level in sheep and often used in genetic evalutions to select for parasite resistance in an animal. FEC does not have a normal distribution due to an excess in zero counts. This could be due to true zero FEC if an animal is resistant to parasites, or false zero FEC if the animal has not been exposed to parasites. A mathematical model was explored to determine if true and false zero FEC could be differentiated. The model failed to differentiate between true and false zero FEC, possibly needing an accurate measure of anemia such as packed cell volume.

Technical Abstract: Fecal egg count (FEC) is used as an indicator of parasite infection level in sheep. Its distribution is non-Gaussian and typically overdispersed, often with an excess in zero counts. Quantifying reasons for inflated zero counts can be difficult. Our objective was to assess a potential zero-inflation problem with FEC resulting from variation in infection with gastrointestinal nematodes using a generalized linear model approach. The zero-inflated negative binomial (ZINB) model is deemed a useful technique to analyze count data with excess zeros and overdispersion. Through a mixture of count and binomial processes, ZINB models may delineate ‘true’ zeros—in this case, animals resistant to parasitism and thereby with zero FEC—from ‘false’ zeros—animals’ never or minimally exposed to a parasite challenge. Separating true and false zeros should allow a more robust characterization of parasite resistance. Two datasets on Katahdin sheep, a hair breed know to express resistance to gastrointestinal nematodes, were investigated: a smaller set (n = 3,048) with FEC and FAMACHA (FAM) scores, a subjective measure of anemia based on the color of the ocular mucous membrane; and a larger set (n = 14,405) with FEC and a contemporary group (CG) designation, assigned based on the animal’s flock, birth year, management group, and FEC recording date. Among animals with FAM recorded, 14 % had scores indicative of at least border line anemia. Amongst the 410 CG, 22% had mean FEC more than 500 eggs/g, a threshold value routinely used to indicate a substantial infection level with Haemonchus contortus, a blood-sucking nematode. For each dataset, the ZINB models were fitted using R (pscl package) and SAS software; FEC was the response variable and either FEC or CG was the explanatory variable, as relevant. Despite evidence of parasite challenge, for neither dataset nor software package could false and true zeros be delineated. The estimated proportion of false zeros was equal to the proportion observed. Either all zeros coincided with no infection, which seems unlikely in Katahdins, or neither FAM nor CG provided sufficient information to distinguish resistant from uninfected individuals. Alternative or additional explanatory variables, such as packed cell volume or immunoglobulins indicative of parasitic infection, may be necessary to separate true from false zero FEC in sheep challenged with gastrointestinal nematodes using a ZINB model.