TABLE 4.

Multivariable mixed effects logistic model findings exploring odds of anti‐inflammatory prescription against a range of clinical signs and presence of prescription of other pharmaceutical classes. Individual practice variance was <0.01 (SD < 0.01), individual site variance was 0.38 (SD 0.62), and individual animal variance was 0.55 (SD 0.74). Univariable analyses are summarised in Supplementary material, Table 2

Variable	Beta	SE a	OR b (CI c )	p value
Intercept	‐3.02	0.18	–	–
		Clinical signs
Unknown/blank narrative	‐0.49	0.26	0.62 (0.37–1.03)	0.065
Abnormal faeces or other GIT	−0.39	0.19	0.68 (0.47–0.98)	0.039
Check‐up	−0.40	0.16	0.67 (0.49–0.92)	0.014
Musculoskeletal	2.00	0.18	7.38 (5.20–10.49)	<0.001
Postural	1.60	0.35	4.96 (2.47–9.94)	<0.001
Skin	0.49	0.15	1.62 (1.21–2.19)	0.001
Swollen head	1.61	0.43	5.03 (2.19–11.57)	<0.001
Trauma	1.03	0.19	2.79 (1.91–4.08)	<0.001
Wasting	−0.40	0.16	0.67 (0.49–0.92)	0.014
		Pharmaceutical prescription
Antimicrobial	1.60	0.13	4.93 (3.82–6.37)	<0.001
Antimycotic	1.80	0.80	6.07 (1.27–28.89)	0.024
Endectocide	‐0.77	0.41	0.46 (0.21–1.03)	0.059
Neurological	1.54	0.53	4.65 (1.65–13.12)	0.004
Replacement agent	1.05	0.37	2.87 (1.38–5.96)	0.005

^aStandard error.

^bOdds ratio.

^c95% confidence interval.