Description

Little et al evaluated various ways to identify a patient with a urinary tract infection. One approach is to use clinical findings alone without any laboratory data. The evaluation was performed as part of the NIHR Health Technology Assessment Programme.


 

Clinical parameters:

(1) urine cloudiness

(2) urine smell

(3) moderate or severe dysuria

(4) moderate or severe nocturia

(5) any hematuria

Parameter

Points

absent

0

present

1

 

4 item clinical score =

= SUM(points for the first 4 items)

 

5 item clinical score =

= SUM(points for all 5 items)

 

Interpretation:

• minimum score: 0

• maximum score: 4 for 4 item score; 5 for 5 item score

• The higher the score the more likely that the patient has a urinary tract infection.

 

Performance:

• A score >= 1 in the 4 item score had a sensitivity of 92% and specificity of 31%.

• A score >= 2 in the 4 item score had a sensitivity of 65% and specificity of 69%.

• A score >= 2 in the 5 item score had a sensitivity of 75% and specificity of 61%.

• None of these findings are very good. One strategy is to do a urine culture on anyone with a score >= 1, but two thirds of the cultures would be "unnecessary".

 


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