StudyRareStudyRare

Positive and Negative Predictive Value

Log in to star

Last updated May 28, 2026

Log in to add personal notes on this page.

Predictive values answer the clinically relevant question: given a test result, what is the probability that the patient actually has (or does not have) the condition? Unlike sensitivity and specificity, predictive values depend on disease prevalence.

  • Positive Predictive Value (PPV) = TP / (TP + FP) = probability of disease given a positive test
  • Negative Predictive Value (NPV) = TN / (TN + FN) = probability of no disease given a negative test
  • PPV increases as prevalence increases (more true positives relative to false positives)
  • PPV decreases as prevalence decreases (false positives overwhelm true positives)
  • NPV is generally high when prevalence is low (most negatives are true negatives)

This is why screening tests perform differently in high-risk vs low-risk populations.

Example 1: NIPT in high-risk vs low-risk

High-risk population (prevalence 1/100):

T21+ (100)T21- (9,900)
Screen+9950
Screen-19,850

PPV = 99/149 = 66%

Low-risk population (prevalence 1/1,000):

T21+ (10)T21- (9,990)
Screen+9.950
Screen-0.19,940

PPV = 9.9/59.9 = 17%

Same test, same sensitivity/specificity, but PPV drops from 66% to 17% because prevalence is lower.

Example 2: Why confirmatory testing is needed

Problem: A 25-year-old with no risk factors has a positive NIPT for trisomy 18. Prevalence at her age is approximately 1/5,000. Test sensitivity is 96%, specificity is 99.5%. What is the PPV?

Solution:

  • In 100,000 pregnancies: 20 with T18, 99,980 without
  • True positives: 20 x 0.96 = 19.2
  • False positives: 99,980 x 0.005 = 499.9
  • PPV = 19.2 / (19.2 + 499.9) = 3.7%
  • Over 96% of positive screens at this prevalence are false positives
  • This is why diagnostic confirmation (amniocentesis/CVS) is essential after a positive screen
  • A positive screening test in a low-prevalence population is more likely to be a false positive than a true positive
  • Diagnostic tests (amnio, CVS) have near-100% sensitivity and specificity, so PPV approaches 100%
  • When counseling a patient about a positive screen, always contextualize the PPV based on their prior risk

"Prevalence Powers PPV": PPV is driven by prevalence. Low prevalence = low PPV, even with an excellent test. Always consider the prior probability (age, family history, ultrasound findings) when interpreting a positive screen.