Skip to main content
. 2020 Aug 18;7:493. doi: 10.3389/fvets.2020.00493

Table 6.

Results of the multivariable logistic model and linear regression of the variables associated with the high Pet Attitude Scale (PAS).

Variables Model 1 Model 2
OR 95% CI P-value B 95% CI P-value
BCS
    Overweight 2.68 1.43–5.01 0.002 5.28 2.00–8.53 0.002
    Normal 2.09 1.34–3.26 0.001 5.58 3.22–7.94 <0.001
    Underweight
Pet's medical history over a year
    Disease 2.36 1.57–3.55 <0.001 6.38 4.25–8.50 <0.001
    Vaccination 2.10 1.12–3.94 0.021 6.22 2.75–9.29 <0.001
    None
Buying the dog
    Yes 0.60 0.42–0.85 0.005 −3.85 −4.74 to −0.97 0.003
    No
Size
    Small (≤10 kg) 1.78 1.09–2.90 0.022
    Large (>10 kg)
Frequently visiting the veterinarian 1.08 1.03–1.13 0.002 0.39 0.21–0.57 <0.001
Spending more time with the dog 1.23 1.16–1.32 <0.001 1.32 1.09–1.56 <0.001
Keeping other species in the house −4.27 −6.38 to −2.16 <0.001
R2 0.279 0.274

Model 1 is the result of a multivariable logistic regression in the high-PAS group where the dependent variable was considered as a categorized variable based on the mean value of 90. Model 2 is the result of a multiple linear regression in the high-PAS group where the dependent variable was considered as a continuous variable. In model 1, the Nagelkerke R2 value was estimated, and in model 2, the adjusted R2 value was estimated.

OR, odds ratios; B, unstandardized beta; CI, confidence interval.