Ebell et al used classification and regression tree (CART) analysis to evaluate patients after in-hospital cardiopulmonary arrest. One model reported was supervised. The authors are from the University of Georgia, Duke University, and the Johns Hopkins University.
Patient selection: in-hospital cardiopulmonary arrest
Outcome: survival to discharge with good or fair neurological status (CPC 1)
High probability of survival: 17-40%
Average probability of survival: 10-15%
Low probability of survival: 5-7%
Very low probability of survival: 2-3%
Parameters:
(1) Cerebral Performance Category (CPC) from 1 to 5
(2) admitted from home or transferred from another hospital
(3) mechanical ventilation
(4) age
(5) sepsis
(6) metastatic cancer or hematologic malignancy
(7) acute myocardial infarction
Good cerebral function (CPC 1)
|
Sepsis
|
Age
|
Cancer
|
Mechanical Ventilation
|
AMI
|
Survival
|
|
Yes
|
NA
|
NA
|
NA
|
NA
|
10.4%
|
|
No
|
>= 72
|
NA
|
NA
|
NA
|
17.2%
|
|
No
|
< 72
|
Yes
|
NA
|
NA
|
12%
|
|
No
|
< 72
|
No
|
Yes
|
NA
|
14.9%
|
|
No
|
< 72
|
No
|
No
|
Yes
|
39.5%
|
|
No
|
< 72
|
No
|
No
|
No
|
28.3%
|
Impaired cerebral function (CPC 2-5)
|
Admitted from
|
CPC
|
Mechanical Ventilation
|
Age
|
Survival
|
|
Yes
|
2-5
|
NA
|
NA
|
2.3%
|
|
No
|
3-5
|
NA
|
NA
|
2.2%
|
|
No
|
2
|
Yes
|
NA
|
5.5%
|
|
No
|
2
|
No
|
>= 80
|
6.2%
|
|
No
|
2
|
No
|
< 80
|
11.4%
|
