Description

Cowling et al reported two models for evaluating a patient with Severe Acute Respiratory Syndrome (SARS). The full model requires access to radiology and laboratory testing. The authors are from the University of Hong Kong, Mount Sinai Hospital (Toronto) and the Canadian SARS Research Network.


Patient selection: severe acute respiratory syndrome (SARS)

 

Outcome: mortality

 

Parameters:

(1) age in years

(2) sex

(3) number of comorbid conditions (asthma, COPD, cardiovascular disease, cerebrovascular disease, cancer, diabetes mellitus, chronic renal disease, chronic liver disease)

(4) chest radiograph

(5) oxygen saturation on room air in percent

(6) serum LDH relative to the upper limit of normal

(7) absolute neutrophil count relative to the upper and lower limit of normal

(8) platelet count relative to the lower limit of normal (thrombocytopenia)

(9) absolute lymphocyte count relative to the lower limit of normal (lymphopenia)

 

Parameter

Finding

Points

age in years

< 40 years

0

 

40 to 49 years

12

 

50 to 59 years

16

 

60 to 69 years

25

 

>= 70 years

34

sex

female

0

 

male

4

comorbid conditions

0

0

 

1

3

 

>= 2

10

chest X-ray

normal

0

 

abnormal

5

oxygen saturation

>= 95%

0

 

< 95%

3

serum LDH

<= ULN

0

 

> ULN

7

absolute neutrophil count

< LLN

-6

 

LLN to ULN

0

 

> ULN

10

platelet count

>= LLN

0

 

< LLN

3

absolute lymphocyte count

>= LLN

0

 

< LLN

-1

 

total score =

= SUM(points for all 9 parameters)

 

Interpretation:

minimum score: -7

maximum score: 76

The higher the score the greater the risk of death.

 

Total Score

Mortality Rate

<= 0

< 1%

1 to 10

2%

11 to 20

5%

21 to 30

13%

31 to 40

27%

41 to 45

43%

46 to 50

54%

51 to 55

66%

56 to 60

76%

61 to 70

87%

> 70

> 91%

 

Performance:

The area under the ROC curve is 0.88 in the derivation cohort and 0.89 in the validation cohort.


To read more or access our algorithms and calculators, please log in or register.