Print

Print


That is a good question. While I can only guess at the number of 'true 
negatives', I am not that interested in the NPV. The NPV (for bacteremia) 
was greater than 95% when the kiddo was carried in the door.

I don't have journal quality data on this but it is my sense that these 3 
misses would make the sensitivity of the Rochester criteria (for bacteremia) 
pretty weak in our neck of the woods. At 30K visits a year, we only have 
about 5 'true positives' a year in this age group (looking at bacteremia 
only). This is one of the ways that these criteria can be a little 
misleading and limit their external validity, IMHO. Bacteremia gets lumped 
in the sensitivity calculation with UTI, and the Rochester criteria are very 
good at picking up UTI. So if 80% of your SBI is pyelo, than this will tend 
to overinflate the sensitivity of the criteria. Moreover, if GBS bacteremia 
is under-represented in a particular study, and pneumococcal bacteremia is 
over-represented, this may have an impact as well (since pneumococcal 
bacteremia is more likely to cause leukocytosis).


Jay 

For more information, send mail to [log in to unmask] with the message: info PED-EM-L
The URL for the PED-EM-L Web Page is:
  http://www.brown.edu/Administration/Emergency_Medicine/ped-em-l.html