de la Varga-Martinez et al reported a model for predicting the risk of delirium for a patient undergoing cardiac surgery. This can help to identify a patient who may require closer monitoring and more aggressive management. The authors are from multiple institutions in Spain.
Patient selection: cardiac surgery
Parameters from the preoperative assessment:
(1) impairment of cognitive function
(2) age in years
(3) level of physical activity
(4) insomnia
|
Parameter
|
Finding
|
Points
|
|
cognitive impairment
|
none
|
0
|
|
|
possible
|
2.294
|
|
|
present
|
1.648
|
|
age in years
|
<= 65 years
|
0
|
|
|
> 65 years
|
1.108
|
|
level of physical activity
|
moderate to high
|
0
|
|
|
low
|
1.1010
|
|
insomnia
|
none
|
0
|
|
|
not requiring medication
|
0
|
|
|
requiring medication
|
1.107
|
value of X =
= SUM(points for all of the parameters) - 4.092)
In the paper, the risk of delirium is reported by the following equation:
probability of delirium =
= 1 / (1 + EXP(X))
However, the standard logistic regression model has the form.
probability of delirium =
= 1 / (1 + EXP((-1) * X))
The results of the second equation are more plausible. The 2 scores are related as the second is equal to 1 minus the first.
Performance:
• The area under the ROC curve was 0.83 in the derivation and 0.79 in the validation cohorts.
