Skip to main content

Table 1 Eligible patient characteristics

From: Improving prediction accuracy of hospital arrival vital signs using a multi-output machine learning model: a retrospective study of JSAS-registry data

Variables

 

Overall

n = 10478

Patient demographics

 Age (years), median (IQR)

 

70 (55, 80)

 Sex (male), n (%)

 

7266 (69)

Underlying condition, n (%)

 Trauma

 

4553 (43)

 Cerebrovascular diseases

 

2144 (20)

 Other internal causes

 

1632 (16)

 Cardiovascular diseases

 

990 (9)

 Other external causes

 

565 (5)

 Unknown

 

594 (6)

Vital sign (n (%), median (IQR))

 GEMS at contact

HR (n = 9515)

82 (70, 97)

 

SBP (n = 9330)

141 (118, 165)

 

DBP (n = 8945)

84 (70, 99)

 

RR (n = 8383)

20 (18, 24)

 

GCS (n = 6756)

15 (13, 15)

 HEMS at contact

HR (n = 10191)

81 (70, 94)

 

SBP (n = 10027)

141 (120, 163)

 

DBP (n = 9852)

83 (70, 97)

 

RR (n = 9647)

20 (18, 24)

 

GCS (n = 10364)

15 (13, 15)

 HEMS final measurement

HR

80 (69, 94)

 

SBP

139 (120, 158)

 

DBP

82 (70, 93)

 

RR

20 (17, 23)

 

GCS

15 (13, 15)

Duration of intervention

 GEMS (min), median (IQR)

 

22 (17, 28)

 HEMS (min), median (IQR)

 

26 (21, 32)

  1. IQR, interquartile range; GEMS, ground emergency medical service; HEMS, helicopter emergency medical service; HR, heart rate; SBP, systolic blood pressure; DBP, diastolic blood pressure; RR, respiratory rate; GCS, Glasgow coma scale