Skip to main content

Table 3 Comparison of clinical characteristics and adverse outcomes between the non-AI and AI groups in new ED patients with hyperglycemic crises between December 1, 2019 and April 30, 2021

From: Using artificial intelligence to predict adverse outcomes in emergency department patients with hyperglycemic crises in real time

Variable

Overall

n = 271

Non-AI

n = 253

AI

n = 18

p-value

Age (years)

69.6 ± 16.6

70.0 ± 16.8

64.4 ± 13.5

0.113

Age subgroup (%)

   

0.090*

 20 − 34

4.4

4.7

0

 

 35 − 49

8.1

7.5

16.7

 

 50 − 64

18.8

17.8

33.3

 

 65 − 74

21.4

20.9

27.8

 

  ≥ 75

47.2

49.0

22.2

 

Sex (%)

 Female

45.0

45.8

33.3

0.432

 Male

55.0

54.2

66.7

 

Body mass index (kg/m2)

22.7 ± 4.6

22.7 ± 4.6

22.9 ± 4.7

0.849

Asian BMI level subgroup (%)

  < 18.5

15.9

15.0

27.8

0.267

 18.5 − 22.9

41.7

43.01

22.2

 

 23 − 24.9

17.0

17.0

16.7

 

  ≥ 25

25.5

24.9

33.3

 

Vital signs at triage

 Glasgow coma scale

11.7 ± 3.9

11.6 ± 3.9

13.67 ± 2.4

0.002

 Systolic blood pressure (mmHg)

101.6 ± 23.0

102.3 ± 23.1

91.4 ± 19.8

0.037

 Heart rate (beats/min)

137.8 ± 36.8

137.9 ± 36.1

135.9 ± 46.9

0.859

 Respiratory rate (breaths/min)

20.1 ± 4.7

20.3 ± 4.7

18.0 ± 3.5

0.018

 Body temperature (°C)

36.6 ± 0.9

36.6 ± 0.9

36.6 ± 0.5

0.893

Bedridden (%)

66.8

66.8

66.7

0.805

Nasogastric tub feeding (%)

14.0

14.2

11.1

 > 0.999

Past histories (%)

 Hypertension

62.7

64.0

44.4

0.159

 Hyperlipidemia

37.3

37.2

38.9

0.916

 Cerebrovascular accident

33.6

34.8

16.7

0.189

 Malignancy

15.1

15.4

11.1

 > 0.999

 Chronic kidney disease

22.9

22.9

22.2

 > 0.999

Laboratory data

 Blood urea nitrogen (mg/dL)

28.2 ± 15.2

28.2 ± 15.7

28.8 ± 7.7

0.753

 White blood cell count (103/µL)

11.6 ± 6.4

11.6 ± 6.5

12.6 ± 5.5

0.440

 Serum creatinine (mg/dL)

2.0 ± 1.7

2.0 ± 1.6

2.7 ± 2.3

0.234

 Hemoglobin (g/dL)

12.5 ± 3.0

12.5 ± 3.0

13.0 ± 2.6

0.391

 Glucose (mg/dL)

416.8 ± 367.3

407.7 ± 363.9

544.1 ± 401.1

0.177

 hs-CRP (mg/L)

50.4 ± 83.8

51.9 ± 85.7

29.0 ± 47.4

0.076

Concomitant infection (%)

62.0

62.8

50.0

0.405

PHD score

2.4 ± 1.4

2.4 ± 1.4

2.4 ± 1.3

0.127

PHD risk class (%)

 Low risk (Score 0–2)

50.9

49.8

66.7

0.366

 Intermediate risk (Score 3)

29.2

29.6

22.2

 

 High risk (Score ≥ 4)

19.9

20.6

11.1

 

Outcomes < 1 month (%)

    

 Sepsis or septic shock

37.6

37.5

38.9

0.890

 ICU admission

19.2

19.8

11.1

0.540

 All-cause mortality

14.8

15.0

11.1

 > 0.999

  1. Data are presented as % or mean ± SD. The independent t-test was used to analyze continuous variables, while the Chi-Square test was utilized to examine categorical variables
  2. AI Artificial intelligence, ED Emergency department, ICU Intensive care unit, BMI Body mass index, hs-CRP High sensitivity C-reactive protein, PHD Predicting the hyperglycemic crisis death, SD Standard deviation
  3. *Because the number of an AI group in the age category “20–34” is 0, we only conducted the test for the other four age subgroups