Question

CA-125 is a marker for the screening of ovarian cancer. To characterize this test, histopathological confirmation of ovarian cancer was done in a cohort of patients. 60 of 100 women who tested positive had ovarian cancer, and 20 of 100 women who tested negative had ovarian cancer. What is the negative predictive value of this test?

• A. 20/100
• B. 40/100
• C. 60/100
• D. 80/100

Hint:

NPV uses only the test-negative column: true negatives over all negatives.

Answer 1

The Correct Option is D

Stay in the test-negative column only, because NPV is a column statistic, not a row one.
The 100 women who tested negative are your whole denominator for NPV. Of them, 20 actually had cancer (false negatives) and 80 did not (true negatives). NPV asks: of everyone who screens negative, what fraction is genuinely disease-free?
So $NPV = \frac{TN}{TN + FN} = \frac{80}{100}$, which is 80 percent. Do not be lured by 60/100, that is positive predictive value from the test-positive column. The answer is 80/100. Ref: Park PSM, screening test characteristics.