Skip to main content

Table 1 Inputs adopted in the model specification

From: Neural networks and hospital length of stay: an application to support healthcare management with national benchmarks and thresholds

Variables

DIEM (%)

Internal medicine (%)

Cardiology (%)

Emergency medicine (%)

Geriatrics (%)

Respiratory diseases (%)

Neurology (%)

Oncology (%)

Age

 18–40 (years)

4.72

2.30

1.46

4.83

0.00

1.95

6.33

1.61

 41–65 (years)

24.57

15.27

29.12

19.93

0.23

21.84

25.18

42.97

 66–75 (years)

21.88

18.46

32.36

20.64

6.54

25.59

19.42

31.53

 > 75 (years)

48.84

63.97

37.06

54.60

93.24

50.63

49.06

23.90

Presence of cancer

15.73

10.23

1.14

5.00

6.20

20.17

8.35

94.98

Sex (% male)

55.55

50.26

65.69

55.59

39.80

61.34

54.24

56.43

Transfers (every 100 hospitalizations)

0.09

0.2

0.16

0.04

0.06

0.09

0.08

Time slot

 Morning

33.20

16.50

74.10

12.70

19.50

31.30

34.20

57.20

 Afternoon

48.80

60.60

16.70

68.40

56.80

54.80

43.70

33.50

 Night

18.10

22.90

9.20

18.90

23.70

13.90

22.00

9.20

Day

 Monday

13.90

13.20

14.50

13.80

12.40

12.20

12.50

17.10

 Tuesday

16.30

14.00

19.80

15.90

13.90

18.20

14.20

17.70

 Wednesday

16.60

16.10

17.30

14.30

15.30

16.10

17.00

16.10

 Thursday

15.80

14.20

20.20

15.50

15.90

15.30

16.40

16.70

 Friday

16.30

15.90

17.80

15.10

15.60

17.50

14.20

16.70

 Saturday

11.70

14.90

4.10

13.20

15.10

11.80

13.50

8.80

 Sunday

10.15

11.70

6.30

12.20

11.80

8.80

12.10

7.00

Urgent hospitalization (direct)

12.8

2.40

4.00

3.40

1.80

21.30

12.40

18.50

Urgent hospitalization (from ER)

69.0

93.60

24.70

96.20

96.60

62.40

86.00

42.00

Planned hospitalization

18.2

4.10

71.40

0.40

1.60

16.30

1.60

39.60

  1. The variable “transfers” is used exclusively in the micro-analysis (i.e., with the most representative medical units)