Which statement about PPV and NPV with respect to prevalence is true?

Which statement about PPV and NPV with respect to prevalence is true?

• A. They depend on test sensitivity only
• B. They depend on test specificity only
• C. They depend on disease prevalence
• D. They are independent of prevalence

Answer: (C)

Predictive values depend on how common the disease is in the population. Positive predictive value (the chance a person with a positive test truly has the disease) increases as prevalence increases, because more of the positives come from people who actually have the disease. Negative predictive value (the chance a person with a negative test truly does not have the disease) decreases as prevalence increases, since a higher proportion of negatives may be false negatives when the disease is more common. The math makes this clear: PPV = (sensitivity × prevalence) / [(sensitivity × prevalence) + ((1 − specificity) × (1 − prevalence))], and NPV = (specificity × (1 − prevalence)) / [(specificity × (1 − prevalence)) + ((1 − sensitivity) × prevalence)]. Both expressions include prevalence, showing that predictive values change with how common the disease is. In contrast, sensitivity and specificity describe the test's performance independent of how common the disease is, so statements claiming dependence on sensitivity alone or independence from prevalence are not accurate.

Predictive values depend on how common the disease is in the population. Positive predictive value (the chance a person with a positive test truly has the disease) increases as prevalence increases, because more of the positives come from people who actually have the disease. Negative predictive value (the chance a person with a negative test truly does not have the disease) decreases as prevalence increases, since a higher proportion of negatives may be false negatives when the disease is more common.

The math makes this clear: PPV = (sensitivity × prevalence) / [(sensitivity × prevalence) + ((1 − specificity) × (1 − prevalence))], and NPV = (specificity × (1 − prevalence)) / [(specificity × (1 − prevalence)) + ((1 − sensitivity) × prevalence)]. Both expressions include prevalence, showing that predictive values change with how common the disease is.

In contrast, sensitivity and specificity describe the test's performance independent of how common the disease is, so statements claiming dependence on sensitivity alone or independence from prevalence are not accurate.