Table 1 Baseline characteristics and outcomes of the cohorts.

From: Accurate prediction of hypoglycemia and hyperglycemia using machine learning in critically ill patients

 

n = 8,853

(1,350,097 records)

Age

67 (52–76)

Males, n (%)

5,238 (59.2)

Height

161.3 (153.5–168.0)

Body weight

58.4 (49.5–68.4)

BMI

22.5 (19.7–25.5)

SOFA score

6 (3–9)

Admission route, n (%)

 Emergency room

2,986 (33.7)

 Operating room

1,782 (20.1)

 General ward

1,319 (14.9)

 Others/unknown

3,690 (41.7)

Diagnosis on admission, n (%)

 Elective surgery

1,518 (17.1)

 Heart diseases

854 (9.6)

 Sepsis/septic shock

724 (8.2)

 Respiratory failure

706 (8.0)

 Post cardiac arrest syndrome

535 (6.0)

 Trauma

435 (4.9)

 Stroke

285 (3.2)

 Others/unknown

4,720 (53.3)

 Blood glucose levels

141 (118–174)

 Experienced hypoglycemia at least once in ICU, n (%)

1,942 (21.9)

 Experienced hyperglycemia at least once in ICU, n (%)

4,930 (55.7)

 Insulin use in the ICU, n (%)

2,841 (32.1)

Outcomes

 ICU mortality, n (%)

665 (7.5)

 ICU stay (days)

7 (3–14)

  1. SOFA: sequential organ failure assessment; ICU: intensive care unit.
  2. Continuous variables are presented as medians (interquartile ranges), and categorical variables are presented as counts and frequencies (%).