Family Medicine Notes

So I'm having dinner tonight with this group of happy medical students and they are bummin out about their EBM course and they ask me if I really use "EBM" and I say yes and then I do this little strep throat thing like this:

I see an adult in the office with a sore throat.  No kids.  No contact with kids

So I assume pre-test probability of about 5% Yeh - I made it up.  I'm probably close.

And the sensitivity of the test I'm doing is 85% (possibly better -- but we'll be safe)

Specificty is very good - 99%

So let's do the poor man's version that I use to explain to the patient:

You have a 5% chance of having this BEFORE I do the test - so after the test you have a 15% chance of the 5% chance of having strep throat.

.15 x .05 = .0075 ... let's round it up to 1%.

"Mrs Jones - you have a 1% chance that you have strep throat after this negative rapid-strep. Go home and drink warm tea with honey and you will feel better soon." (ok .. go find the evidence for THAT!)

---------

Students get confused.  "That's too easy" they seem to say. "We have to do all of those crazy calculations."

No - just understand.  I say.  Do the math and you'll get the same result here.

So here goes.

Given my number above - we need to calculate the Negative Likelihood Ratio - which is:

Negative Likelihood=(1-SENS)/SPEC

So:  (1-.85)/.99 = 0.15  

So our negative Likelhood Ratio is .15 = 15%

Now we need to relate this to our patient.  For this - the EBM disciples use the Fagan Nomogram which really uses the formula:

Pos-test odds =  Likelihood Ratio x pre-test odds

So ... Post-test odds = 15% x 5%

which is the same as

15 x .05 = .0075 ... let's round it up to 1%.

Look familar?