Table 5.

Demographic, clinical, laboratory, and chest X-ray data that demonstrated predictive value for death and the most invasive respiratory support employed (none, oxygen mask, continuous positive airway pressure/noninvasive ventilation or invasive ventilation) in 340 patients with confirmed COVID-19

	Predictors	Odds ratio	95% CI	p
Death	Age, years	1.16	[1.11–1.22]	< 0.001
	SpO2, % *	0.96	[0.92–1.01]	0.115
	PaO2/FiO2 ratio*	0.99	[0.98–1.00]	0.002
	Cardiovascular disease**	3.21	[1.28–8.39]	0.014
	Brixia score [18]	1.19	[1.06–1.34]	0.003
Respiratory support	Age, years	0.96	[0.94–0.98]	0.001
	SpO2, % *	0.96	[0.92–0.99]	0.008
	PaO2/FiO2 ratio *	0.99	[0.99–1.00]	< 0.001
	Rheumatic pathology	3.22	[1.05–9.89]	0.041
	% of lung involvement	1.02	[1.01–1.03]	0.001

*Within 24 h of ED presentation. **Including coronary heart disease, cerebrovascular disease, heart failure, and peripheral vascular disease. SpO₂ reported only in cases with FiO₂ = 0.21. Odds ratios, 95% CI, and p values were computed by logistic (death) and ordinal logistic (respiratory support) regression models, using stepwise model selection technique. Only patients with no missing data (n = 210) were included in the models. ED, emergency department; PaO₂/FiO₂ ratio, ratio of partial pressure of oxygen to fraction of inspired oxygen; SpO₂, oxygen saturation