TABLE 5.

Results of the multivariable logistic regression analyses for associations with a diagnosis of AKI‐L vs AKI‐nL as dependent variable

Variable (reference)	β	SE	OR	95% CI of OR	Wald‐P
L4‐vaccination model
Intercept	1.70	0.28	5.493	3.175‐9.506	<.001
L4‐vaccination status (L0)
L4+	−2.19	0.35	0.112	0.056‐0.222	<.001
L4−	−2.40	0.43	0.090	0.039‐0.211	<.001
Age	−0.13	0.03	0.877	0.824‐0.934	<.001
Neuter status (intact)
Neutered	−0.48	0.25	0.619	0.381‐1.005	.05
73.4% dogs classified correctly; ROC‐AUC, 0.774 Pearson chi‐square 69.1; 12 outliers on Pearson residual analysis

Variable (reference)	β	SE	OR	95% CI of OR	Wald‐P
L2‐vaccination model
Intercept	0.72	0.64	2.046	0.588‐7.120	.26
L2‐vaccination status (L0)
L2+	0.73	0.65	2.078	0.582‐7.419	.26
L2−	0.04	0.66	1.037	0.286‐3.757	.96
Age	−0.10	0.04	0.905	0.839‐0.976	.009
Neuter status (intact)
Neutered	−0.85	0.29	0.426	0.239‐0.759	.004
61.2% dogs correctly classified; ROC‐AUC, 0.690
Pearson chi‐square 119.3; 1 outlier on Pearson residual analysis

Note: These analyses include all 368 dogs with complete vaccination information for the L4 model and 273 dogs for the L2 model. Additional diagnostics conducted on both models include the linearity check for the continuous variable age. The variable checks confirmed the inclusion of the model variables except for the L2‐vaccination status in the L2‐model; none of the variable interactions was retained.

Abbreviations: β, regression coefficient; OR, odds ratio; 95% CI, 95% confidence interval; ROC‐AUC, area under the curve of the receiver operating characteristic curve.