USMLE Forums banner

1 - 3 of 3 Posts

·
Registered
Joined
·
741 Posts
Discussion Starter · #1 · (Edited)
The 2x2 Table in biostatistics and epidemiology also referred to as Punnett Square is the most important piece of information that the student has to learn by heart in order to answer many questions on both USMLE Step 1 and CK exams.

Here's the general structure:
2x2table.JPG

Sensitivity = A/A+C (best way to understand it is to remember that a highly sensitive test has less false negative results)
Specificity = D/B+D (best way to understand it is to remember that a highly specific test test has less false positive results)
Positive Predictive Value = A/A+B
Negative Predictive Value = D/D+C
Odds Ratio = AD/BC (see http://www.usmle-forums.com/usmle-step-1-mnemonics/645-odds-ratio.html)
Accuracy = A+D / A+B+C+D or True rates divided by all

Note that in USMLE they may simply rotate the table and you have to be careful to locate the A cell which is a positive test in a diseased person.
This 2X2 table is not only for disease/test, it can be disease/effect/side effects/outcome versus drug/exposure/risk factor ....etc.

Examples:

  • Patient had a positive HIV test comes and ask you what is the chance that his result is true? Answer, give him the positive predictive value.
  • Lady looking for pregnancy had a negative pregnancy test and asks you how correct is her result? Answer, give her the negative predictive value.
  • In developing a screening test what will you be looking for sensitivity or specificity? Answer, sensitivity as we need less false negative results.
 

·
Registered
Joined
·
2 Posts
iPhone Epi calculator

Great new free iPhone epidemiology calculator- "iEpidemiology"

Now available in the Apple App store
 
1 - 3 of 3 Posts
Top