Skip to main content

Table 3 Evidence level and validation of scoring systems to predict in-patient mortality in the medical admission unit

From: Risk scoring systems for adults admitted to the emergency department: a systematic review

Scoring system

Level of evidence1

Validated in a new population at same site as developed

Validated in a new population at an external location

Discrimination (ability to identify patients at risk), AUROC

Calibration (agreement between predicted and observed risk), Chi-square

REMS[21]

3

•

•

0.852 (+/- 0.014)

487.3 (p < 0.0001)

RAPS[17]

3

-

•

NS

NS

GS[18]

4

-

-

0.81 (95% CI: 0.78-0.84)

NS

HOTEL[15]

3

•

-

0.865 (0.793-0.937)

Validation: 0.854 (0.746-0.962)

1.49 (p = 0.83)

SCS[14]

3

•

-

0.858 (SE 0.009)

Validation: 0.856 (SE 0.013)

NS

WPS[20]

3

•

-

0.74 (0.74-0.77)

p = 0.119

RLD[13]

3

•

-

2

2

TTS[16]

2

•

•3

0.657-0.782

NS

EWS[19]

2

•

-

0.68 (0.65-0.71)

NS

ALT[12]

3

•

-

0.904

4

  1. NS = not specified
  2. 1 according to McGinn et al[8]
  3. 2 only specified for validation cohorts
  4. 3 several publications validate TTS
  5. 4 mentioned to be good, but not specified