Challen et al developed models for evaluating patients in the Emergency Department (ED). One model predicts short-term mortality rate for the patient. The authors are from the University of Sheffield.
Patient selection: patient in the ED
Outcome: 7-day mortality rate
Parameters:
(1) Glasgow Coma Score (GCS)
(2) age in years
(3) respiratory rate in breaths per minute
(4) diastolic blood pressure in mm Hg
(5) oxygen saturation risk group
(6) body temperature in °C
(7) respiratory disease
|
Parameter |
Finding |
Points |
|
GCS |
15 |
0 |
|
|
13 or 14 |
5.311 |
|
|
3 to 12 |
29.372 |
|
age in years |
< 50 years of age |
0 |
|
|
50 to 69 years of age |
0.608 |
|
|
70 to 85 years of age |
0.348 |
|
|
> 85 years of age |
0.924 |
|
respiratory rate |
< 19 breaths per minute |
0 |
|
|
19 to 23 |
1.753 |
|
|
> 23 |
2.791 |
|
diastolic blood pressure |
> 90 mm Hg |
5.204 |
|
|
65 to 90 mm Hg |
0 |
|
|
< 65 |
3.172 |
|
oxygen saturation |
low risk (99 to 100% on room air) |
0 |
|
|
moderate risk (95 to 98% on room air or > 95% on supplemental oxygen) |
0.932 |
|
|
high risk (< 95% on room air or <=95% on supplemental oxygen) |
3.202 |
|
body temperature |
>= 36°C |
0 |
|
|
< 36°C |
4.714 |
|
respiratory disease |
absent |
0 |
|
|
present |
12.355 |
|
respiratory disease and temperature |
temperature < 36°C AND respiratory disease |
7.466 |
|
|
other |
0 |
X =
= SUM(points for all 7 parameters)
probability of a potential to benefit =
= 1 / (1 + EXP((-1) * X))
Interpretation:
• The minimum value of X is 0 which occurs when all of the findings are normal. This gives a probability of 50% which would seems high.
• The maximum value of X is 41.391 which has a probability of 100% which seems high.
• Many logistic regression models subtract a constant. None is specified in the paper or supplemental files.
Performance:
• The area under the ROC curve was 0.85 in the derivation set and 0.74 in the validation set.
Specialty: Surgery, orthopedic, Emergency Medicine, Critical Care, Surgery, general
