Biostats Review

  • /Reviewed by: Amy Rontal, MD
  • One thing students always fear, be it on the USMLE, COMLEX, or Shelf exams, is biostatistics.

    As a tutor, I have found that most students and medical schools are not spending nearly enough time teaching their students key statistical concepts and helping them learn to apply biostats when interpreting research.

    But fear not: As we go through the new school year, I will be posting detailed biostatistics breakdowns with the hope of better preparing anyone reading this blog. Let’s get started.


    The sensitivity of a test refers to its’ ability to correctly identify patients with a specific disease. Tests with high sensitivity (i.e. ELISA for HIV) are important for conditions where early diagnosis is imperative for successful treatment. A test with high sensitivity will have a low number of false negatives.

    When calculating sensitivity (or specificity), it is always important to start by constructing your 2×2 table.

    Positive Disease Negative Disease
    Positive Test Result True Positive False Positive
    Negative Test Result False Negative True Negative

    From here, you must commit the following formula to memory:

    Sensitivity = True Positives/ (True Positives + False Negatives)

    Now, let’s try a real example:

    A new blood test has been created that will help diagnosis ovarian cancer. A study of 200 patients is performed; 100 with ovarian cancer and 100 without. Using this new test, 90 patients test positive for ovarian cancer with further workup showing that 80 of these patients with positive tests truly had ovarian cancer. What is the sensitivity of this new test?

    A) 25%
    B) 50%
    C) 75%
    D) 80%
    E) 90%

    Ovarian Cancer No Ovarian Cancer
    Test Positive 80 10
    Test Negative 20 90

    Sensitivity = 80/ (80 + 20) = 80%

    If you have any questions, please let us know. Otherwise, stay tuned—we’ll be back with more Biostat practice soon.