Table 5

Performance of admission prediction models in the older population

StudyModel nameAdmission, N (%)Derivation AUC (95% CI)Validation methodValidation AUCCalibration methodCalibration
Brouns et al26MTS4223 (59.4)External0.74 (0.73–0.75)
Di Bari et al20ISAR558 (34)External0.65 (0.62–0.68)
Di Bari et al20SC558 (34)External0.63 (0.60–0.65)
Grossmann et al27ESI250 (48.8)External0.741 (0.734–0.747)
Lucke et al25Own model older patients1817 (43.8)0.81 (0.79 to 0.82)External0.77 (0.75–0.79)Calibration plot, GOF testp>0.05
Salvi et al21ISAR626 (30)External0.68 (0,66–0.70)
Salvi et al21TRST626 (30)External0.66 (0.64–0.69)
  • Empty cells mean that specific characteristics were not tested.

  • AUC, area under the curve; ESI, emergency severity index; GOF, goodness of fit; ISAR, identification of seniors at risk; MTS, Manchester triage system; SC, silver code; TRST, triage risk screening tool.