A deep learning model for predicting multidrug-resistant organism infection in critically ill patients

Table 3 Multivariable logistic analysis for MDRO infection

Predictors	β	SE	Wald χ²	OR	95% CI	P value
(Intercept)	− 6.81	0.62	− 10.91	0.001	0.000–0.003	< 0.001
APACHE II	0.06	0.02	3.05	1.06	1.02–1.110	0.002
Quantity of antibiotics (categories)	0.59	0.22	2.72	1.81	1.18–2.78	0.002
Chronic lung disease	0.70	0.35	2.03	2.02	1.02–3.97	0.040
Hypoproteinemia, n (%)	1.28	0.54	2.37	3.59	1.24–10.35	0.010
Invasive operation before ICU, n (%)	0.79	0.32	2.46	2.20	1.17–4.11	0.010
Antibiotics use before ICU, n (%)	1.08	0.32	3.40	2.95	1.58–5.53	< 0.001
Length of hospitalization (days)	0.03	0.01	4.23	1.04	1.02–1.05	< 0.001
Length of ICU stay (days)	0.02	0.01	1.98	1.02	1.00–1.05	0.040
Long-term bed rest, n (%)	1.30	0.38	3.42	3.69	1.80–8.12	< 0.001

MDRO: multidrug-resistant organism; ICU: intensive care unit; APACHE II: acute physiology and chronic health evaluation II

ISSN: 2052-0492