Biostatistics & Epidemiology USMLE Practice Questions
Study design, statistical analysis, and public health.
4 free biostatistics & epidemiology practice questions
Pick an answer and check it. After each question you get the same explanation StepGenie students see: the stem decoded clue by clue, the reasoning chain with its infographic, why every wrong option is wrong, and the trap the question was built around. Step 1 questions test mechanism; Step 2 CK questions ask for the next step.
A recent epidemiologic study conducted in several Scandinavian towns evaluated the incidence and 1-year case fatality of ischemic stroke. The annual incidence of ischemic stroke was 60 cases per 2,000 people, and the 1-year case fatality rate was 20%. A town in southern Sweden with a population of 20,000 would like to estimate the expected number of deaths from ischemic stroke in 1 year based on these data. Which of the following is the best estimate of the annual mortality rate due to ischemic stroke per 20,000 people?
Correct answer · B
120 people
Decoding the stem
Step-by-step reasoning
Step 1 of 5
Incidence in the study towns: 60 cases per 2,000 people per year → 60/2000 = 0.03 = 3% annual incidence.
Step 2 of 5
Scale incidence to a town of 20,000: 3% of 20,000 = 0.03 × 20,000 = 600 new ischemic strokes per year.
Step 3 of 5
Case fatality rate is 20% → 20% of those 600 strokes will die within 1 year.
Step 4 of 5
Calculate deaths: 0.20 × 600 = 120 deaths per year from ischemic stroke.
Step 5 of 5
Thus, the annual mortality rate due to ischemic stroke per 20,000 people is best estimated as 120 people.
Why not the others?
A. 12 people
This implies a much lower mortality. It would correspond to 20% of only 60 strokes (not scaled to 20,000), BUT the town has 20,000 people, so strokes scale up to 600.
C. 400 people
This suggests a 2/3 fatality of 600 strokes or misusing incidence; BUT case fatality is only 20%, so deaths must be far fewer than total strokes.
D. 60 people
This is 10% of 600 or 100% of 60; BUT the case fatality is 20%, so deaths must be 20% of 600, not 10% or 100% of 60.
E. 600 people
This equals the total incident stroke cases (600), implying 100% case fatality, BUT the study clearly states a 20% 1‑year case fatality rate.
Exam trap
Students often forget to first scale incidence to the new population (20,000) and/or to then apply the 20% case fatality; some pick 600 (incidence) or 60 (original numerator) instead of deaths.
The bottom line
Ischemic stroke deaths = incidence × case fatality. Here: 60/2000 × 20% scaled to 20,000 → 120 annual deaths.
A genetic population study is conducted to evaluate the penetrance of an inherited disorder characterized by impaired iron metabolism that primarily affects the liver. Affected individuals commonly develop diabetes mellitus and bronze skin pigmentation. Among 120 individuals with a family history of this disease who undergo genetic screening, 40 are found to carry the disease-causing genotype. Of these, only 10 exhibit clinical manifestations of the disorder. What is the probability that individuals with the disease-producing genotype will develop the disease phenotype?
Correct answer · B
25%
Decoding the stem
Step-by-step reasoning
Step 1 of 6
The described disease (iron overload, liver involvement, diabetes, bronze skin) corresponds to hereditary hemochromatosis.
Step 2 of 6
The question is not asking about disease mechanism but about genetic penetrance: probability that a genotype manifests as a phenotype.
Step 3 of 6
Penetrance is calculated among carriers of the disease genotype only.
Step 4 of 6
Here, 40 people have the disease-causing genotype (denominator).
Step 5 of 6
Of those 40, only 10 show clinical disease (numerator).
Step 6 of 6
Penetrance = 10 ÷ 40 = 0.25 = 25%.
Why not the others?
A. 3%
This implies 3 affected out of 100 genotype carriers, which does not match 10/40. Penetrance must be calculated from the given genotype-phenotype counts.
C. 4%
This implies 4 affected out of 100 carriers, BUT the data show 10 affected among 40 carriers; using 120 total screened instead of 40 confuses penetrance with prevalence in the screened group.
D. 40%
This would suggest 16 affected out of 40 carriers. The stem explicitly states only 10 have manifestations; overestimates penetrance.
E. 0.4%
This misplaces the decimal and likely confuses total screened (120) with carriers (40). Penetrance is among carriers only: 10/40, not 10/12000.
Exam trap
Students often confuse penetrance (affected among genotype carriers) with prevalence in the entire screened population. Only use the 40 carriers as the denominator.
The bottom line
Hemochromatosis: incomplete penetrance; penetrance = affected with phenotype / total with genotype = 10/40 = 25%.
A novel PET radiotracer is being evaluated for its utility in diagnosing Alzheimer disease (AD). A total of 1,000 patients are enrolled, with 500 patients having AD and 500 serving as controls. Among patients with AD, 400 have a positive PET result, whereas 50 control patients also have a positive result. What is the positive predictive value (PPV) of this novel PET examination?
Correct answer · C
400 / (400+50)
Decoding the stem
Step-by-step reasoning
Step 1 of 6
Identify disease-positive group: 500 patients with Alzheimer disease (AD).
Step 2 of 6
Of those with AD, 400 test positive → True positives (TP) = 400; therefore False negatives (FN) = 500 - 400 = 100.
Step 3 of 6
Among 500 controls (no AD), 50 test positive → False positives (FP) = 50; therefore True negatives (TN) = 500 - 50 = 450.
Step 4 of 6
Recall formula: PPV = TP / (TP + FP), the probability of having disease given a positive test.
Step 5 of 6
Plug in numbers: PPV = 400 / (400 + 50) = 400 / 450.
Step 6 of 6
Match with answer choice: 400 / (400+50).
Why not the others?
A. 400 / (400+100)
This option uses TP / (TP + FN), which is the formula for sensitivity, not PPV. PPV must include false positives, not false negatives.
B. 450 / (450 + 50)
This implies TP = 450, FP = 50, which is incorrect. 450 is the number of true negatives, not true positives.
D. 400 / (400 + 150)
This adds extra 100 to FP (50), effectively using TP / (TP + FP + FN). That is not any standard test characteristic formula.
E. 450 / (450 + 100)
This uses TN / (TN + FN), which is not a standard measure (neither specificity nor NPV). Specificity is TN / (TN + FP); NPV is TN / (TN + FN).
Exam trap
Students often confuse PPV with sensitivity or NPV. The question stresses a positive test, so think: among positives, how many truly have disease → TP / (TP + FP).
The bottom line
Positive predictive value (PPV) = true positives / (true positives + false positives). Here: 400 / (400 + 50).
An epidemiologist evaluates nationwide health surveillance data after several large clinics report a substantial increase in the prevalence of primary biliary cholangitis (PBC) over the last 20 years. Analysis reveals that the incidence of PBC has remained stable during this period. Which of the following is the most plausible explanation for the increased prevalence of PBC?
Correct answer · B
Improved quality of care for PBC
Decoding the stem
Step-by-step reasoning
Step 1 of 5
Prevalence ≈ Incidence × Disease Duration. If incidence is stable, any change in prevalence must come from a change in average disease duration.
Step 2 of 5
Disease duration increases when patients live longer with the disease (reduced mortality or slower progression).
Step 3 of 5
Improved quality of care (better treatment, earlier management of complications) leads to longer survival of PBC patients.
Step 4 of 5
With the same number of new cases (stable incidence) but more people staying alive with PBC, the number of existing cases at any time (prevalence) increases.
Step 5 of 5
Therefore, improved quality of care is the most plausible explanation for higher prevalence with unchanged incidence.
Why not the others?
A. Increased exposure to environmental risk factors for PBC
This option implies more new cases (↑incidence), which should raise both incidence and prevalence, BUT the stem explicitly states incidence is stable, contradicting this.
C. Increased availability of diagnostic testing for PBC
Better testing increases case detection, often seen as an apparent rise in incidence first, BUT this vignette says incidence has not increased.
D. Increased average age of the population at risk for PBC
An older at-risk population usually leads to more new cases of a chronic disease (↑incidence), BUT again, the stem says incidence remained stable.
E. Increased awareness of PBC among clinicians
More clinician awareness typically causes more diagnoses and an apparent ↑incidence, BUT the question explicitly notes no change in incidence.
Exam trap
Students often see '↑prevalence' and assume '↑incidence' (more risk factors, testing, or awareness). Here, the stem locks incidence as stable, forcing you to think of the other determinant of prevalence: disease duration/survival, which points to improved quality of care.
The bottom line
Primary biliary cholangitis prevalence can rise despite stable incidence when patients live longer due to improved quality of care (reduced mortality).
Source: StepGenie question bank. Questions are written by StepGenie, not taken from NBME or USMLE materials, and are chosen for this page by how students actually performed on them (first-attempt accuracy on StepGenie). Selection last refreshed September 5, 2026.
Keep going on Biostatistics & Epidemiology
A free account includes 5 Visual QBank questions a day in practice blocks you build by subject and organ system, plus 10 Pattern Mastery questions a day.
About Biostatistics & Epidemiology
Essential biostatistics and epidemiology for evidence-based medicine. Master study designs, statistical tests, sensitivity/specificity calculations, and interpretation of medical literature.
Key Topics Covered
Why Study Biostatistics & Epidemiology with StepGenie?
- AI-Powered Explanations: Get personalized, Socratic explanations when you're stuck on complex biostatistics & epidemiology concepts.
- Adaptive Learning: Questions automatically adjust to target your weak areas in biostatistics & epidemiology.
- Spaced Repetition Flashcards: Reinforce high-yield biostatistics & epidemiology facts with our FSRS-powered flashcard system.
- Pattern Recognition: Master classic biostatistics & epidemiology presentations through keyword recognition drills.
Related Articles
Q-Bank Breakdown: Kaplan-Meier curves — Why Every Answer Choice Matters
Kaplan-Meier questions reward careful reading: steps mark events, tick marks mark censoring, and the best comparison of two curves is the log-rank test.
Q-Bank Breakdown: Cohort vs case-control — Why Every Answer Choice Matters
Case-control questions hinge on who gets selected first. If investigators start with disease and look backward for exposure, the answer is odds ratio.
Q-Bank Breakdown: Meta-analysis — Why Every Answer Choice Matters
Meta-analysis means statistically pooling studies, not just reviewing them. Learn the forest plot clues, line of no effect, and the classic study-design mix-ups.
Biostatistics & Epidemiology study guides
Cross-Sectional vs Ecological Studies for USMLE Step 1
Cross-sectional means an individual-level snapshot tied to prevalence but not causality. Ecological means group-level averages, with ecological fallacy as the classic trap.
Intention-to-Treat Analysis: ITT vs Per-Protocol
Intention-to-treat means analysing everyone in the group they were assigned to, even after dropout or crossover. That preserves randomization and curbs bias.
Types of Bias: Selection, Recall, Lead-Time & Length-Time
Bias questions get easier when you match the stem to the distortion: who got selected, who remembers differently, or why screening only looks better.
Type I vs Type II Error: Alpha, Beta, Power & Examples
Type I means rejecting a true null; Type II means missing a real effect. Tie each to alpha, beta, power, and classic screening-test examples fast.
Biostatistics & Epidemiology — Frequently Asked Questions
What is the difference between sensitivity and specificity?
Sensitivity is the proportion of people with disease who test positive, so a highly sensitive test with a negative result rules disease out. Specificity is the proportion of people without disease who test negative, so a highly specific test with a positive result rules disease in. Both are properties of the test and do not change with disease prevalence.
How does prevalence affect predictive values?
Positive predictive value rises as prevalence rises, and negative predictive value falls. This is why a screening test performing well in a high-risk population produces many false positives when applied to a low-risk one. Sensitivity and specificity stay fixed, which makes the distinction between them and predictive values a favourite exam point.
When do you use an odds ratio versus a relative risk?
Relative risk requires knowing incidence, so it applies to cohort studies and randomised trials where subjects are followed forward. Case-control studies begin with outcome and look backward, so incidence is unknown and the odds ratio is used. When the outcome is rare, the odds ratio approximates the relative risk closely.
What does a p-value actually mean?
It is the probability of observing a result at least as extreme as the one obtained, assuming the null hypothesis is true. It is not the probability that the null hypothesis is true, and it says nothing about effect size or clinical importance. A confidence interval is more informative because it shows both the precision and the magnitude of the effect.
Are the biostatistics & epidemiology practice questions on this page free?
Yes. The 4 questions above come from StepGenie's question bank and can be answered here without an account, with the full explanation shown after each one. From the link under the set you can try three more Biostatistics & Epidemiology questions in the dashboard preview, also without an account; a free account is what lets you choose the system and keep going.