Skip to main content

Table 1 Patients’ characteristics

From: Machine learning-based prediction models for accidental hypothermia patients

Variables Development cohort Validation cohort
(N = 288) (N = 244)
Men 144 (50.0%) 126 (51.6%)
Age, years 79 (69–87) 79 (64–87)
 < 60 37 (12.8%) 47 (19.3%)
 60–69 35 (12.2%) 37 (15.2%)
 70–79 75 (26.0%) 48 (19.7%)
 ≥ 80 140 (48.6%) 117 (48.0%)
Activities of daily living
 Disturbance 96 (33.3%) 66 (27.0%)
Comorbidity
 Cardiovascular diseases 126 (43.8%) 111 (45.5%)
 Neurological diseases 53 (18.4%) 40 (16.4%)
 Endocrine diseases 83 (28.8%) 47 (19.3%)
 Psychiatric diseases 55 (19.1%) 63 (25.8%)
 Malignant diseases 12 (4.2%) 4 (1.6%)
 Dementia 57 (19.8%) 51 (20.9%)
 Other 56 (19.4%) 38 (15.6%)
External and minimally invasive rewarming
 Warm intravenous fluid 223 (77.4%) 168 (68.9%)
 Forced warm air 80 (27.8%) 4 (1.6%)
 Warm environment, warm blanket 242 (84.0%) 222 (91.0%)
 Other 23 (8.0%) 15 (6.1%)
Active internal rewarming
 Lavage 29 (10.1%) 15 (6.1%)
 CHDF 4 (1.4%) 17 (7.0%)
 VV-ECMO 0 (0%) 2 (0.8%)
 VA-ECMO 3 (1.0%) 17 (7%)
In-hospital mortality 64 (22.2%) 66 (27.0%)
  1. Categorical variables: n (%), continuous variables: median [interquartile range]
  2. CHDF Continuous hemodiafiltration, VV-ECMO Veno-venous extracorporeal membrane oxygenation, V-A ECMO Veno-arterial membrane oxygenation