Skip to content

Biostatistics and epidemiology medical question bank

Practise every biostatistics and epidemiology topic with explained SBAs.

Recall Engine has 800+ biostatistics and epidemiology questions across 40+ topics. Every question explains each option.

questions
800+
topics
40+
topic notes
40+
Start free

7 days of the library free plus 50 credits.

Biostatistics and epidemiology by body system

Biostatistics and epidemiology has 40+ topics. Each section shows how many questions it has and every topic it covers.

Population health and evidence

800+ questions across 40+ topics. It also has 40+ topic notes.

  • Assessment of measuring instruments
  • Binary probability: addition and multiplication rules
  • Case-control studies
  • Cohort studies
  • Communicable disease transmission
  • Critical appraisal: study designs
  • Data collection tools
  • Demography
  • Descriptive statistics and summarising data
  • Descriptive statistics: summary measures and graphs
  • Diagnostic test accuracy
  • Diagnostic tests: probability and risk
  • Disease frequency: prevalence and incidence
  • Disease surveillance
  • Distributions and probability
  • Epidemiology: risks, rates, and ratios
  • Epidemiology: study design and bias
  • Estimation: standard error and confidence intervals
  • Evidence-based health care
  • Health indicators: morbidity, mortality, and fertility
  • Hypothesis testing in practice: errors, power, and test choice
  • Introduction to research
  • Mortality rates and death certificates
  • Normal distribution: z-scores
  • Outbreak investigation
  • Randomised controlled trials: design
  • Research questions, hypotheses, and protocols
  • Sampling inference: parameter estimation
  • Sampling methods
  • Scientific writing, awareness campaigns, and referencing
  • Screening tests
  • Significance testing
  • Significance tests: p values and test choice
  • SPSS: chi-square, Fisher exact, and McNemar tests
  • SPSS: data entry, coding, and recoding
  • SPSS: descriptive statistics and graphs
  • Study designs: observational and interventional
  • Systematic reviews and meta-analysis: forest plots
  • Treatment uncertainty and evidence
  • Trial endpoints, designs, and follow-up
  • Trial results: effect size and precision
  • Variation in coronary heart disease

Biostatistics and epidemiology sample questions

These 5 are published biostatistics and epidemiology questions shown exactly as students see them. Open an answer to see why it is right and why every other option is wrong.

Question 1 · Biostatistics and epidemiology, Preclinical

A new drug is to be tested for a condition for which a standard treatment is already available. Which comparison group should the trial use?

  1. A. Physiotherapy
  2. B. A newer experimental drug
  3. C. An identical placebo with no active ingredient
  4. D. The current standard treatment
  5. E. Surgery
Show the answer and explanation

Answer: D. The current standard treatment

A placebo is often used for comparison, but if a standard treatment is available this should be used as the control, so the trial compares the intervention with the standard of care. A control group is the comparison group receiving standard treatment or placebo.

Why the other options are wrong

  • A. Physiotherapy: Physiotherapy is the control group when surgery is the experimental intervention.
  • B. A newer experimental drug: A new drug is the experimental intervention, with the old drug as the control group.
  • C. An identical placebo with no active ingredient: If a standard treatment is available it should be used as the control, and it is not ethical to use placebo in many circumstances.
  • E. Surgery: Surgery is the experimental intervention when physiotherapy is the control group.

Key point: When a standard treatment is available, it should be used as the control group.

Question 2 · Biostatistics and epidemiology, Preclinical

Investigators with a small number of cases add controls to gain statistical power. Up to how many controls per case is it efficient to use?

  1. A. 1
  2. B. 2
  3. C. 3
  4. D. 4
  5. E. 5
Show the answer and explanation

Answer: C. 3

The control group should be at least equal in size to the case group, and an extra control for each case adds statistical power, especially with few cases. Use of more than 3 controls for each case is not efficient.

Why the other options are wrong

  • A. 1: The size of the control group should be at least equal to the size of the case group.
  • B. 2: The use of an extra control for each case adds statistical power, especially with a small number of cases.
  • D. 4: Use of more than 3 controls for each case is not efficient.
  • E. 5: Use of more than 3 controls for each case is not efficient.

Key point: Use at least as many controls as cases, and more than 3 controls per case is not efficient.

Question 3 · Biostatistics and epidemiology, Preclinical

A school health programme measures systolic blood pressure in 2,000 healthy secondary-school students. Values follow a normal distribution with mean 112 mmHg and SD 9 mmHg. The school doctor wants whole-SD limits that contain about 95 percent of students to use as a local reference range. Which limits contain about 95 percent of the students?

  1. A. 103 to 121 mmHg
  2. B. 85 to 139 mmHg
  3. C. 97 to 127 mmHg
  4. D. 112 to 130 mmHg
  5. E. 94 to 130 mmHg
Show the answer and explanation

Answer: E. 94 to 130 mmHg

For a normal distribution about 68 percent of values lie within 1 SD of the mean, about 95 percent (95.4 percent) within 2 SD, and 99.7 percent within 3 SD. Two SD = 2 x 9 = 18 mmHg, so the limits are 112 - 18 = 94 mmHg and 112 + 18 = 130 mmHg. The exact 95 percent limits use 1.96 SD (17.6 mmHg): 94.4 to 129.6 mmHg, which round to the same range.

Why the other options are wrong

  • A. 103 to 121 mmHg: These limits are the mean plus or minus 1 SD (112 - 9 and 112 + 9), which contain only about 68 percent of students.
  • B. 85 to 139 mmHg: These limits are the mean plus or minus 3 SD (112 - 27 and 112 + 27), which contain about 99.7 percent, wider than the range requested.
  • C. 97 to 127 mmHg: These limits come from plus or minus 1.645 SD (about 14.8 mmHg), which contains 90 percent of a normal distribution, not 95 percent.
  • D. 112 to 130 mmHg: This covers only the upper half, from the mean to 2 SD above it, which holds about 47.7 percent of students.

Key point: In a normal distribution about 68, 95, and 99.7 percent of values lie within 1, 2, and 3 SD of the mean, so a 95 percent reference range is the mean plus or minus 2 SD (exactly 1.96 SD).

References

Question 4 · Biostatistics and epidemiology, Preclinical

Brucellosis notifications in a governorate tripled this quarter compared with the same quarter of the past three years. The governorate laboratory switched in January from a tube agglutination test to a more sensitive ELISA and now reports every positive result. Hospital admissions for brucellosis, the case definition, and the number of reporting units are unchanged. What is the most likely explanation for the rise?

  1. A. A change in the brucellosis case definition
  2. B. A seasonal peak linked to the lambing and kidding period
  3. C. A reporting artefact caused by the new diagnostic test
  4. D. A common-source outbreak from contaminated dairy products
  5. E. A propagated outbreak spreading from person to person
Show the answer and explanation

Answer: C. A reporting artefact caused by the new diagnostic test

An apparent excess may be artefactual. Improved diagnostic procedures (a more sensitive test, reporting of every positive) raise notifications without any change in true incidence. Unchanged admissions, case definition, and reporting units, with a rise starting exactly at the test change, point to a diagnostic artefact rather than a real outbreak.

Why the other options are wrong

  • A. A change in the brucellosis case definition: The case definition is stated to be unchanged; the only change is the laboratory method.
  • B. A seasonal peak linked to the lambing and kidding period: A seasonal rise would also appear in the same quarter of previous years; this quarter is compared with those years and is three times higher.
  • D. A common-source outbreak from contaminated dairy products: A true common-source outbreak would raise admissions too; they are unchanged, and the rise began with the test change.
  • E. A propagated outbreak spreading from person to person: Brucellosis is a zoonosis that rarely spreads between people, and nothing shows successive waves of cases.

Key point: Before declaring an outbreak, exclude artefactual rises from new diagnostic tests, changed reporting procedures or case definitions, heightened awareness, seasonal variation, and laboratory error.

References

Question 5 · Biostatistics and epidemiology, Preclinical

In ORACLE, women with preterm prelabour rupture of the membranes were allocated erythromycin, co-amoxiclav, or placebo. What proportion allocated co-amoxiclav delivered within 48 hours?

  1. A. 29.1%
  2. B. 34.0%
  3. C. 39.8%
  4. D. 56.7%
  5. E. 84.3%
Show the answer and explanation

Answer: A. 29.1%

The table gives delivery within 48 hours as 34.0% with erythromycin, 29.1% with co-amoxiclav, and 39.8% with placebo.

Why the other options are wrong

  • B. 34.0%: 34.0% of women allocated erythromycin delivered within 48 hours.
  • C. 39.8%: 39.8% of women allocated placebo delivered within 48 hours.
  • D. 56.7%: 56.7% of women allocated co-amoxiclav delivered within 7 days.
  • E. 84.3%: 84.3% of women allocated co-amoxiclav delivered before 37 weeks.

Key point: Delivery within 48 hours was 29.1% with co-amoxiclav in ORACLE women with preterm prelabour rupture of the membranes.

More preclinical disciplines

Every discipline has its own page with its topics and sample questions.

Start free

7 days of the library free plus 50 credits.