Table 3.

Results of multivariate logistic regression model for “antibiotic prescription”. Factors significantly associated with antibiotic prescription at consultation according to multivariate logistic regression (N = 1454)

Variables	Number of patients N Percentage (%) N = 1454	Odds Ratio	P	95 % Confidence Interval
Temperature above 38 °C (for at least one day)	659 (45.26 %)	1.42	0.019	1.06–1.90
Symptom duration longer than 7 days	421 (28.91 %)	3.94	<0.001	2.83–5.48
Headache while bending forward	181 (12.43 %)	1.93	0.003	1.26–2.95
Redness of the throat	556 (38.19 %)	2.26	<0.001	1.63–3.12
Tonsillar exudates	178 (12.23 %)	14.72	<0.001	7.68–28.18
Enlargement of cervical and/or submandibular lymph nodes	260 (17.86 %)	2.74	<0.001	1.81–4.13
Expectoration of purulent sputum	203 (13.94 %)	2.22	0.001	1.36–3.63
Thick purulent nasal discharge	173 (11.88 %)	3.87	<0.001	2.43–6.15
Auscultation sounds: wheezing or rhonchi	181 (12.43 %)	13.03	<0.001	7.14–23.8
Auscultation sounds: rales or crepitations	99 (6.80 %)	26.90	<0.001	9.00–80.35
Patient factor 1: starting antibiotics on his/her own	111 (7.62 %)	4.11	<0.001	2.33–7.27
Patient factor 2: direct request for antibiotics	247 (16.96 %)	1.87	0.002	1.27–2.77
Factors that decreased the likelihood of antibiotic prescription
Patient factor 9: direct request not to prescribe	80 (5.49 %)	0.34	0.001	0.18–0.64
Sore throat	919 (63.12 %)	0.58	0.001	0.42–0.80
Rhinorrhoea	787 (54.05 %)	0.36	<0.001	0.27–0.48