TABLE 1

Clinical variables selected by logistic regression models to assess intensive care unit (ICU) admission and hospital mortality prediction

ICU admissionHospital mortality
Logistic regression variables OR (95% CI)p-valueLogistic regression variables OR (95% CI)p-value
Age1.62 (1.43–1.83)<0.0011.68 (1.51–1.87)<0.001
FiO24.10 (3.55–4.73)<0.0014.32 (3.75–4.97)<0.001
Systolic blood pressure1.20 (1.05–1.38)0.0071.20 (1.05–1.38)0.007
Diastolic blood pressure0.80 (0.70–0.93)0.0030.80 (0.70–0.93)0.003
SO20.84 (0.76–0.94)0.0020.82 (0.74–0.91)<0.001
Glasgow Coma Scale0.60 (0.53–0.69)<0.0010.61 (0.54–0.69)<0.001
Sex1.42 (1.28–1.59)<0.0011.44 (1.29–1.60)<0.001
Dyspnoea1.42 (1.28–1.58)<0.0011.50 (1.35–1.66)<0.001
Obesity1.42 (1.28–1.58)<0.0011.43 (1.28–1.59)<0.001
Vomiting/nausea#1.00 (0.90–1.11)0.88
Abdominal pain1.08 (0.98–1.20)0.111.05 (0.94–1.16)0.34
Chronic kidney disease#1.20 (1.08–1.33)<0.001
Conjunctivitis#1.00 (0.90–1.12)0.91
Arterial hypertension1.17 (1.05–1.32)0.0051.21 (1.08–1.35)0.001
Skin ulcers#0.95 (0.86–1.06)0.43
Diabetes mellitus#,¶1.22 (1.10–1.36)<0.001

FiO2: fraction of inspired oxygen; SO2: oxygen saturation. #: missing values were variables not selected for the predictive model analysis due to statistical significance or biological plausibility; : noncomplicated diabetes.