MCQ Quiz: Bias & Confounding

Bias and confounding are the primary threats to the validity of any research study. For pharmacists to practice evidence-based medicine, they must be skilled at critically appraising literature to identify these potential flaws, a core competency taught in the Principles of Evidence-Based Practice course. This quiz will test your ability to define, identify, and differentiate various types of bias and confounding, which is essential for determining the trustworthiness of clinical evidence.

1. A systematic error in the design, conduct, or analysis of a study that results in a mistaken estimate of an exposure’s effect on the risk of disease is known as:

• a. Random error
• b. Confounding
• c. Bias
• d. A p-value

Answer: c. Bias

2. A third variable that is associated with both the exposure and the outcome, and which distorts the true relationship between them, is called a(n):

• a. Effect modifier
• b. Confounding variable
• c. Odds ratio
• d. Selection bias

Answer: b. Confounding variable

3. In a case-control study, patients with a disease (cases) may remember their past exposures differently than those without the disease (controls). This is a classic example of:

• a. Selection bias
• b. Recall bias
• c. Attrition bias
• d. Confounding

Answer: b. Recall bias

4. The best study design feature to control for both known and unknown confounding variables is:

• a. Blinding
• b. Matching
• c. Randomization
• d. Using a large sample size

Answer: c. Randomization

5. A study finds that coffee drinking is associated with lung cancer. However, the researchers did not account for smoking. Because smoking is associated with both coffee drinking and lung cancer, it is a(n):

• a. Bias
• b. Outcome
• c. Exposure
• d. Confounder

Answer: d. Confounder

6. The “Observational Studies” module, where bias and confounding are major topics, is part of which course?

• a. PHA5244 Principles of Evidence-Based Practice
• b. PHA5104 Sterile Compounding
• c. PHA5703 Pharmacy Law and Ethics
• d. PHA5787C Patient Care 5

Answer: a. PHA5244 Principles of Evidence-Based Practice

7. “Selection bias” occurs when:

• a. The measurement of information is systematically different between study groups.
• b. The way participants are selected for the study creates systematic differences between the groups being compared.
• c. Participants drop out of the study.
• d. A third variable distorts the observed effect.

Answer: b. The way participants are selected for the study creates systematic differences between the groups being compared.

8. The “healthy worker effect” is a type of selection bias where the working population is healthier than the general population. This can lead to:

• a. An overestimation of the risk of an occupational exposure.
• b. An underestimation of the risk of an occupational exposure.
• c. No change in the estimated risk.
• d. An issue with confounding only.

Answer: b. An underestimation of the risk of an occupational exposure.

9. In a prospective cohort study, if the research assistants who are assessing for the outcome know which participants are exposed and which are not, what type of bias could occur?

• a. Recall bias
• b. Selection bias
• c. Information (or interviewer) bias
• d. Confounding by indication

Answer: c. Information (or interviewer) bias

10. A study’s “internal validity” refers to:

• a. The degree to which its results can be generalized to other populations.
• b. The degree to which its results are correct and free from bias for the population being studied.
• c. The statistical power of the study.
• d. The reputation of the journal it is published in.

Answer: b. The degree to which its results are correct and free from bias for the population being studied.

11. The appraisal of cohort studies, which includes assessing for bias, is a “Transcending Concept” in the Patient Care 5 curriculum.

• a. True
• b. False

Answer: a. True

12. How can researchers control for confounding in the analysis phase of a study?

• a. By using randomization.
• b. By using blinding.
• c. By using statistical techniques like stratification or multivariable regression.
• d. It is not possible to control for confounding during the analysis phase.

Answer: c. By using statistical techniques like stratification or multivariable regression.

13. A study’s “external validity” is another term for its:

• a. Internal consistency
• b. Statistical significance
• c. Generalizability
• d. Freedom from bias

Answer: c. Generalizability

14. A study on a new antihypertensive drug enrolls only healthy young men. The results show the drug is effective. The study may have high internal validity, but it lacks external validity for:

• a. All other populations.
• b. Young men.
• c. Elderly women with multiple comorbidities.
• d. Anyone taking an antihypertensive.

Answer: c. Elderly women with multiple comorbidities.

15. “Attrition bias” occurs when:

• a. Participants are not selected for the study properly.
• b. Participants who drop out of the study are systematically different from those who remain.
• c. The outcome is measured incorrectly.
• d. The exposure is measured incorrectly.

Answer: b. Participants who drop out of the study are systematically different from those who remain.

16. Increasing the sample size of a study will reduce:

• a. Selection bias
• b. Recall bias
• c. Confounding
• d. Random error

Answer: d. Random error

17. “Journal Club” is an activity where students practice critically appraising studies for issues like bias and confounding.

• a. True
• b. False

Answer: a. True

18. In a randomized controlled trial, “blinding” is used primarily to prevent:

• a. Confounding
• b. Selection bias
• c. Performance and detection bias
• d. Attrition bias

Answer: c. Performance and detection bias

19. Which of the following is NOT a type of bias?

• a. Selection bias
• b. Information bias
• c. Publication bias
• d. Randomization

Answer: d. Randomization

20. An active learning session on appraising cohort studies is part of the Patient Care 5 course.

• a. True
• b. False

Answer: a. True

21. “Confounding by indication” is a major challenge in pharmacoepidemiology. It means that:

• a. The drug was prescribed for the wrong indication.
• b. Patients who are prescribed a drug are often inherently different (e.g., sicker) from those who are not, and this underlying difference is the true cause of the outcome.
• c. The indication for a drug changes over time.
• d. The drug has many indications.

Answer: b. Patients who are prescribed a drug are often inherently different (e.g., sicker) from those who are not, and this underlying difference is the true cause of the outcome.

22. A study that only publishes positive, statistically significant results contributes to:

• a. Publication bias
• b. Recall bias
• c. Selection bias
• d. Attrition bias

Answer: a. Publication bias

23. The “Experimental Studies” module is part of the EBP course.

• a. True
• b. False

Answer: a. True

24. An active learning session on EBP is part of which course?

• a. PHA5244 Principles of Evidence-Based Practice
• b. PHA5163L Professional Skills Lab 3
• c. PHA5781 Patient Care I
• d. PHA5787C Patient Care 5

Answer: a. PHA5244 Principles of Evidence-Based Practice

25. A pharmacist appraising a study must first assess its internal validity before considering its results.

• a. True
• b. False

Answer: a. True

26. Which of the following is a way to control for confounding in the design phase of a study?

• a. Stratification
• b. Multivariable analysis
• c. Blinding
• d. Restriction (e.g., only enrolling non-smokers to eliminate smoking as a confounder).

Answer: d. Restriction (e.g., only enrolling non-smokers to eliminate smoking as a confounder).

27. Unlike bias and confounding, random error:

• a. Leads to a systematic distortion of the results.
• b. Is not affected by sample size.
• c. Is due to chance and can be reduced by increasing sample size.
• d. Only occurs in case-control studies.

Answer: c. Is due to chance and can be reduced by increasing sample size.

28. An active learning session on cohort study appraisal is part of which course module?

• a. Module 5: Urological Disorders
• b. Module 1: Diabetes Mellitus
• c. Module 4: Medication Safety
• d. Module 8: Men’s Health

Answer: a. Module 5: Urological Disorders

29. The main difference between confounding and bias is that:

• a. Confounding is a systematic error, while bias is not.
• b. Confounding reflects a true underlying relationship between variables, while bias is a result of flaws in the study’s methods.
• c. Bias can be controlled for in the analysis phase, while confounding cannot.
• d. There is no difference.

Answer: b. Confounding reflects a true underlying relationship between variables, while bias is a result of flaws in the study’s methods.

30. The “Pharmacoepidemiology Study Designs” module is part of the EBP course.

• a. True
• b. False

Answer: a. True

31. A study compares the rate of falls in a nursing home before and after a pharmacist-led medication review. The results may be biased due to:

• a. Lack of a concurrent control group.
• b. The Hawthorne effect (participants behaving differently because they know they are being observed).
• c. Other changes in the nursing home that occurred at the same time.
• d. All of the above.

Answer: d. All of the above.

32. A case-control study is particularly susceptible to which type of bias?

• a. Attrition bias
• b. Selection bias and recall bias
• c. Performance bias
• d. It is not susceptible to bias.

Answer: b. Selection bias and recall bias

33. If a cohort study has a very high rate of loss to follow-up, its internal validity is threatened.

• a. True
• b. False

Answer: a. True

34. The best way for a pharmacist to assess a study for bias and confounding is to:

• a. Read only the abstract and conclusion.
• b. Critically read the “Methods” section in detail.
• c. Look at the reputation of the authors.
• d. See how many times the article has been cited.

Answer: b. Critically read the “Methods” section in detail.

35. A “confounding by severity” means that:

• a. The disease is very severe.
• b. The drug is more likely to be given to patients with more severe disease, and this severity is what truly drives the outcome.
• c. The drug causes the disease to become more severe.
• d. The severity of the disease is not important.

Answer: b. The drug is more likely to be given to patients with more severe disease, and this severity is what truly drives the outcome.

36. A well-conducted cohort study can prove causation.

• a. True
• b. False

Answer: b. False

37. Which of the following is NOT a criterion for establishing a factor as a confounder?

• a. It must be associated with the exposure.
• b. It must be associated with the outcome.
• c. It must be on the causal pathway between the exposure and the outcome.
• d. It must not be on the causal pathway between the exposure and the outcome.

Answer: c. It must be on the causal pathway between the exposure and the outcome.

38. The “Appraisal of cohort studies” is a lecture in the Principles of Evidence-Based Practice course.

• a. True
• b. False

Answer: a. True

39. A researcher uses a miscalibrated blood pressure cuff that consistently reads 5 mmHg too high for all patients. This is an example of:

• a. Random error
• b. Systematic measurement bias
• c. Selection bias
• d. Confounding

Answer: b. Systematic measurement bias

40. An active learning session covering EBP is part of which course?

• a. PHA5244 Principles of Evidence-Based Practice
• b. PHA5163L Professional Skills Lab 3
• c. PHA5781 Patient Care I
• d. PHA5787C Patient Care 5

Answer: a. PHA5244 Principles of Evidence-Based Practice

41. The primary purpose of matching cases and controls in a case-control study is to:

• a. Increase the sample size.
• b. Control for confounding variables.
• c. Make the study easier to analyze.
• d. Introduce bias.

Answer: b. Control for confounding variables.

42. A study with low internal validity cannot have high external validity.

• a. True
• b. False

Answer: a. True

43. A pharmacist’s ability to critically appraise literature is essential for:

• a. Answering drug information questions accurately.
• b. Making evidence-based recommendations.
• c. Understanding the limitations of clinical evidence.
• d. All of the above.

Answer: d. All of the above.

44. “Blinding” helps prevent bias, while “randomization” helps prevent ________.

• a. bias
• b. confounding
• c. random error
• d. attrition

Answer: b. confounding

45. A study that reports a Relative Risk but does not report a 95% Confidence Interval is difficult to interpret because:

• a. You cannot assess the statistical significance of the result.
• b. You cannot calculate the p-value.
• c. You cannot determine the precision of the estimate.
• d. Both a and c.

Answer: d. Both a and c.

46. A “confounder” must be a risk factor for the disease among the unexposed.

• a. True
• b. False

Answer: a. True

47. “Introduction to Observational Studies” is a lecture in the EBP course.

• a. True
• b. False

Answer: a. True

48. An active learning session on EBP is part of which course module?

• a. Module 5: Urological Disorders
• b. Module 2: Pharmacoepidemiology Study Designs
• c. Module 4: Medication Safety
• d. Module 6: Geriatrics

Answer: b. Module 2: Pharmacoepidemiology Study Designs

49. The overall management of bias and confounding requires:

• a. Careful consideration in the design phase and control in the analysis phase of a study.
• b. Ignoring them as they are unavoidable.
• c. A large sample size only.
• d. A statistically significant p-value.

Answer: a. Careful consideration in the design phase and control in the analysis phase of a study.

50. The ultimate reason to learn about bias and confounding is to:

• a. Be a critical consumer of scientific literature to provide the best care for patients.
• b. Be able to criticize every research article.
• c. Pass the EBP exam.
• d. Make journal club presentations more difficult.

Answer: a. Be a critical consumer of scientific literature to provide the best care for patients.