Questionnaire survey of detrimental fur animal epidemic necrotic pyoderma in Finland

Table 5 Multivariable logistic regression analyses of significant risk factors for fur animal epidemic necrotic pyoderma (FENP)

Model	Cases (n)/	Risk factors	OR (95% CI)	Goodness-of-fit statistics
Model	Controls (n)	Risk factors	OR (95% CI)	Test	Value	P
Model 1	88/118	Farm type
		Mink farm vs fox farm	1.3 (0.5–3.1)	McFadden’s R²	0.147
		Mixed farm vs fox farm	3.8 (1.9–7.6)	Cox-Snell R²	0.182
		Imported from Denmark	6.0 (1.6–22.8)	Pearson	0.721	0.608
		Imported from Poland	7.2 (1.4–37.3)
Model 2	42/23	Access by birds	4.6 (1.2–16.8)	McFadden’s R²	0.188
		Nipple	8.4 (2.0–35.0)	Cox-Snell R²	0.217
				Pearson	0.411	0.675
Model 3	42/34	Imports	5.3 (1.6–18.0)	McFadden’s R²	0.241
		Access by wildlife	13.6 (1.5–121.0)	Cox-Snell R²	0.282
		Size of the farm > 750 vs ≤ 750 mink	3.1 (1.0–9.0)	Pearson	0.561	0.847
Model 4	65/90	Mink farm vs fox farm	4.5 (2.1–9.4)	McFadden’s R²	0.179
		Access of wildlife	2.3 (1.0–5.4)	Cox-Snell R²	0.216
		Nipple	3.3 (1.6–7.0)	Pearson	0.658	0.621

Model 1 all farms, Model 2 mixed farms Model 3 farms with mink Model 4 farms with foxes
Number (n) of case farms and control farms in the model and odds ratios (OR) of the variables included in the model. In all of the goodness-of-fit tests, a test value of 1 indicates a particular well-fitting model; a Pearson’s value <0.05 indicates which model should be rejected. The variable “farm type” has three categories: mink, fox and mixed farms where fox farms serve as the reference group

ISSN: 1751-0147