QUICKmedicine app

likelihood ratio ( LR )

FREE subscriptions for doctors and students... click here
You have 3 open access pages.

 
   

  • this is defined as the ratio of the probability of a test result among patients with a target disorder to the probability of that same test result among patients who do not have the target disorder
  • the likelihood ratio for a positive test is calculated as:
    sensitivity / (1-specificity)
  • the likelihood ratio of a negative result is calculated as:
    (1-sensitivity)/specificity

Note that sensitivity = (a/[a+c]) and

specificity = (d/[b+d])

where:

  target disorder present target disorder absent
test result positive

a

b

test result negative

c

d

 

 

Links: