Nguyen and Graviss reported a score for predicting mortality for a patient with both tuberculosis and diabetes. This can help to identify a patient who may benefit from more aggressive management. The authors are from Houston Methodist Research Institute.
Patient selection: tuberculosis and diabetes, receiving antituberculous therapy, United States resident
Parameters:
(1) age in years
(2) location of birth
(3) homelessness
(4) intravenous drug use
(5) chronic kidney failure
(6) TB meningitis
(7) miliary tuberculosis
(8) AFB sputum smear positive
(9) HIV status
Parameter |
Finding |
Points |
age in years |
< 65 years |
0 |
|
>= 65 years |
16 |
location of birth |
USA |
5 |
|
other |
0 |
homeless |
no |
0 |
|
yes |
11 |
IV drug use |
no |
0 |
|
yes |
20 |
chronic kidney failure |
no |
0 |
|
yes |
20 |
TB meningitis |
no |
0 |
|
yes |
20 |
miliary TB |
no |
0 |
|
yes |
13 |
AFB sputum smear |
negative |
0 |
|
positive |
6 |
HIV status |
negative |
0 |
|
positive |
24 |
total score =
= SUM(points for all of the parameters)
Interpretation:
• minimum score: 0
• maximum score: 135
Score |
Mortality Risk |
<= 11 |
low |
12 to 21 |
medium |
>= 22 |
high |
value of X =
= (0.1 * (score)) - 4
probability of mortality =
= 1 / (1 + EXP((-1) * X))
Performance:
• The area under the ROC curve is 0.82.
Specialty: Infectious Diseases