USMLE Step 1 practice question: Positive predictive value and the effect of prevalence (2 × 2 table)
USMLE Step 1BiostatisticsBiostatistics
What is the probability that a primary care patient with a positive assay result has celiac disease?
- 10%
- 25%
- 50%
- 75%
- 90%
Show answer and explanation
Correct answer: C. 50%
Step 1 – Build the 2 × 2 table for the referral study (N = 1,000; 250 with disease, 750 without disease): • True positives (TP) = 225; false negatives (FN) = 250 − 225 = 25 • False positives (FP) = 75; true negatives (TN) = 750 − 75 = 675
Sensitivity = TP / (TP + FN) = 225 / 250 = 90%. Specificity = TN / (TN + FP) = 675 / 750 = 90%. (In the referral population, PPV = 225 / (225 + 75) = 225 / 300 = 75%, and NPV = 675 / (675 + 25) = 675 / 700 ≈ 96%.)
Step 2 – Apply the same sensitivity and specificity to the primary care population (prevalence 10%). Using a hypothetical 1,000 patients: • With disease: 100 → TP = 0.90 × 100 = 90; FN = 10 • Without disease: 900 → TN = 0.90 × 900 = 810; FP = 900 − 810 = 90
PPV = TP / (TP + FP) = 90 / (90 + 90) = 90 / 180 = 50%. (NPV = 810 / (810 + 10) = 810 / 820 ≈ 99%.)
Key concept: Sensitivity and specificity are intrinsic properties of a test and do not change with prevalence (assuming a similar spectrum of disease). Predictive values, however, depend on prevalence: as prevalence decreases, PPV decreases (false positives become more numerous relative to true positives) and NPV increases. Thus the same assay with 90% sensitivity and 90% specificity has a PPV of 75% in a referral center (prevalence 25%) but only 50% in primary care (prevalence 10%). The question asks for the probability of disease given a positive result, which is the definition of PPV.
High-yield summary: PPV = TP / (TP + FP); ↓prevalence → ↓PPV and ↑NPV; sensitivity and specificity are unaffected by prevalence.
Why the other options are wrong
- A. 10% — 10% is the prevalence (pretest probability) of celiac disease in primary care. A positive result raises the probability above the pretest value; PPV = 90 / 180 = 50%.
- B. 25% — 25% is the prevalence in the referral study (250 / 1,000), which is neither the pretest probability in primary care nor a predictive value.
- D. 75% — 75% is the PPV in the referral population (225 / 300), where prevalence was higher. PPV falls when the same test is applied to a lower-prevalence population.
- E. 90% — 90% is the sensitivity (225 / 250) and also the specificity (675 / 750) of the assay. Sensitivity is the probability of a positive test given disease, not the probability of disease given a positive test.
Key point
Positive predictive value (TP / [TP + FP]) depends on disease prevalence: applying a test with fixed sensitivity and specificity to a lower-prevalence population lowers PPV and raises NPV.
Memory hook
PPV follows Prevalence — the P's rise and fall together; NPV moves opposite. Sensitivity and specificity don't budge. Rarer disease → more false positives.
15 free bank questions a day, exam-length timed blocks, spaced repetition and 3D anatomy — in English and Russian.
Play free