Table 3.

Statistically significant associations between probabilities and attributes used to structure scenario trees. Attributes tested: production type, age, gender, herd size, herd-health scheme membership, role on farm

Probability of veterinary contact given variation in listed attribute	P value	Description	Accepted/rejected	Reason for acceptance/rejection
On death (recorded) of an animal with or without herd-health scheme	0·08	Members of a herd-health scheme showed an increased probability.	Accepted	—
On ED syndrome ± dairy production	0·029	Dairy farms showed a lower probability.	Accepted	—
On ED syndrome ± herd health scheme within dairy farms	0·051	Dairy farms showed a greater probability when members of a herd-health scheme.	Accepted	Significant in multivariate analysis when controlling for production type
On ED syndrome with age of farmer	0·059	Younger farmers less probability to report	Rejected	Insignificant when controlling for production type in multivariate analysis
On ND syndrome ± dairy production	<0·001	Dairy farms showed a lower probability than non-dairy	Accepted	—
On ND syndrome with large herd size	0·036	Lower probability with large herd size	Rejected	Large herd size associated with dairy production (P < 0·001). Insignificant when controlling for production type
On ND ± routine veterinary visits	0·041	Lower probability with routine veterinary visits	Rejected	Routine visits associated with production type (P < 0·001). Insignificant when controlling for production type
On RD ± routine veterinary visits	0·021	Those units with routine veterinary visits showed a greater probability	Accepted	—

ED, enteric disease; ND, neurological disease; RD, reproductive disease.