Associations between hair trace mineral concentrations and the occurrence of treponeme-associated hoof disease in elk (Cervus canadensis)

Table 1 Ten hypotheses of mineral influences on treponeme-associated hoof disease (TAHD) in elk. We evaluated multiple models to address independent a priori hypotheses and post hoc models on hoof health minerals and toxicants’ influence on TAHD diagnosis and severity of disease (i.e., “Status” and “Grade_Severity”, respectively) using the dataset without outliers. Model formulas are in italics beside a shorthand for their motivating hypotheses

Model	Formula
A priori hypotheses
A. Hoof health minerals	Status ~ Co + Cu + Mo + Se + Zn
B. Hoof health minerals with sex adjustment	Status ~ Co + Cu + Mo + Se + Sex + Zn
C. Hoof health minerals with “Longitude” ^a adjustment	Status ~ Co + Cu + Mo + Long. + Se + Zn
D. Toxicant hypothesis	Status ~ As + Co + Cr + Pb
E. Toxicants with “Longitude” adjustment	Status ~ As + Co + Cr + Long.+ Pb
F. Severity of disease and hoof health minerals	Grade_Severity ~ Co + Cu + Mo + Se + Zn
G. Severity of disease affected by toxicants and “Longitude”	Grade_Severity ~ As + Co + Cr + Long.+ Pb
Post hoc models
H. Interaction^b of “Longitude” and Se	Status ~ Co + Cu + Mo + Se * Long. + Zn
I. Se with toxicants and “Longitude” adjustment	Status ~ As + Co + Cr + Long.+ Pb + Se
J. Cr (as possible toxicant) exacerbating Se deficiency	Status ~ Co + Cu + Mo + Se * Cr + Zn
K. Co removed and substituted with Mn from Model C	Status ~ Cu + Long. + Mo + Mn + Se + Zn

^aLongitude category = “west” or “east” assigned from available location information (see Methods)
^bInteraction term was applied with releveling longitude under two sub-models: H1 and H2

ISSN: 1746-6148