Skip to main content
. 2022 Jul 25;22:194. doi: 10.1186/s12911-022-01934-2

Table 2.

Encounter characteristics of the training, validation, and testing cohorts

Characteristics Training cohort
(N = 303)
Validation cohort
(N = 101)
Testing cohort
(N = 104)
p-value train versus Test* p-value validation versus test* p-value train + validation versus test*
Number of unique patients N (%) 288 (95.0) 96 (95.0) 97 (93.3)
Primary outcome (N, (%))
Mortality 43 (14.2) 6 (5.9) 11 (10.6) 0.18 0.12  < 0.0001
Demographics
Age in years Mean (std) 56.6 (16.6) 56.6 (15.6) 53.4 (14.2) 0.012 0.028 0.009
Female N (%) 147 (48.5) 50 (49.5) 56 (53.8) 0.18 0.27 0.18
Race/ethnicity (N, (%)) 0.63 0.95 0.76
Black 137 (45.2) 51 (50.5) 49 (47.1)
Hispanic 36 (11.9) 13 (12.9) 16 (15.4)
Other, non- hispanic 112 (37.0) 30 (29.7) 32 (30.7)
White 18 (5.9) 7 (6.9) 7 (6.7)
Mean (std) of the number of laboratory measurements per encounter
636 (786) 510 (663) 531 (972) 0.078 0.228 0.090
Mean (std) vital signs measurements per encounter
999 (1540) 765 (1344) 802 (1971) 0.026 0.12 0.030
Comorbidities 0.81 0.69 0.81
Mean (std) comorbidities per encounter 1.0 (1.1) 1.0 (1.1) 0.9 (0.9)
Hypertension N (%) 128 (42.2) 43 (42.6) 37 (35.6)
Diabetes N (%) 89 (29.4) 32 (31.7) 30 (28.8)
Heart disease N (%) 12 (3.9) 1 (1.0) 2 (1.9)
COPD N (%) 3 (1.0) 0 (0.0) 1 (1.0)
Stroke N (%) 1 (0.3) 0 (0.0) 0 (0.0)
Cerebrovascular disease N (%) 0 (0.0) 2 (2.0) 0 (0.0)
Cancer N (%) 4 (1.3) 2 (2.0) 1 (1.0)
Respiratory problems N (%) 44 (14.5) 12 (11.9) 15 (14.4)
Chronic kidney disease N (%) 28 (9.2) 11 (10.9) 6 (5.7)
Tuberculosis N (%) 3 (1.0) 1 (1.0) 3 (2.9)

Bold indicates p-value < 0.05

Significance was set at 0.05

Patients older than 89 have been clipped to age 90

*Continuous variables were compared using a t-test and categorical variables were compared using a Chi-square test