Li et al reported a model for predicting in-hospital mortality for a patient in the intensive care unit (ICU). This can help to identify a patient who may benefit from more aggressive management. The authors are from multiple hospitals in CHina.
Patient selection: adult in the ICU
Outcome: in-hospital mortality
Parameters:
(1) serum lactate
(2) neutrophil to lymphocyte ratio (NLR)
(3) Acute Physiology Score (APS), from 0 to 71
(4) Charlson Comorbidity Index (CCI)
(5) surgery
Parameter
Finding
Points
serum lactate
0.106 * (lactate)
NLR
0.026 * (NLR)
APS
0.101 * (APS)
CCI
0.306 * (CCI)
surgery
none
2.085
elective surgery
0
emergency surgery
1.296
X =
= SUM(points for all 5 parameters) - 5.06
probability of in-hospital mortality =
= 1 / (1 + EXP((-1) * X))
Performance:
• The area under the ROC curve is 0.84.
To read more or access our algorithms and calculators, please log in or register.