Prehospital intravenous fentanyl administered by ambulance personnel: a cluster-randomised comparison of two treatment protocols

Table 3 Odds ratios for differences in effect and safety measures

	Odds ratio (95% CI)		P-value
	Crude	Adjusted^a	P-value
NRS ≤ 3 at hospital arrival:
Standard	1 (reference)	1 (reference)	0.001
Liberal	1.42 (1.13–1.80)	1.47 (1.17–1.84)	0.001
NRS decrease (relative) during transport:
Standard	1 (reference)	1 (reference)	0.14
Liberal	1.15 (0.90–1.46)	1.18 (0.95–1.48)	0.14
GCS < 15:
Standard	1 (reference)	1 (reference)	0.76
Liberal	0.87 (0.27–2.80)	0.84 (0.27–2.62)	0.76
Respiratory rate < 10/min:
Standard	1 (reference)	1 (reference)	0.85
Liberal	1.07 (0.37–3.09)	1.11 (0.40–3.09)	0.85
Oxygen saturation < 90%:
Standard	1 (reference)	1 (reference)	0.15
Liberal	1.19 (0.96–1.47)	1.17 (0.95–1.44)	0.15
MAP < 70 mmHg and decrease ≥ 10 mmHg:
Standard	1 (reference)	1 (reference)	0.96
Liberal	1.08 (0.74–1.57)	1.01 (0.70–1.47)	0.96

Abbreviations: CI confidence interval, NRS numeric rating scale, GCS Glasgow Coma Scale, MAP mean arterial pressure, EMT emergency medical technician, PM paramedic
^a: Adjusted for age, sex, comorbidity, etiology (injury yes/no), time in ambulance, station size, ambulance personnel experience (EMT or PM) and mean dose administered by ambulance personnel. For vital parameters, the highest cluster level (ambulance station) has been removed in the adjusted analyses to ensure better model fit

ISSN: 1757-7241