Case–control studies and odds ratio calculation MCQs With Answer

Introduction

This quiz collection focuses on case–control studies and odds ratio (OR) calculation, tailored for M. Pharm students studying Pharmacoepidemiology & Pharmacoeconomics. It explains study design principles, selection of cases and controls, bias and confounding issues, matched versus unmatched analyses, and practical OR computations from 2×2 tables. You will practice calculating crude and matched ORs, interpreting OR magnitude and confidence, understanding when OR approximates relative risk, and recognizing analytic methods like conditional logistic regression. These MCQs combine conceptual depth with numerical problems to reinforce critical thinking required for designing, analyzing, and interpreting case–control investigations in pharmacovigilance and drug safety research.

Q1. What best describes a case–control study?

• An observational study comparing past exposures in subjects with a disease (cases) and without the disease (controls)
• A randomized trial assigning exposure to participants prospectively
• A longitudinal cohort study measuring incidence of disease over time
• An ecological study comparing population-level exposure and outcome rates

Correct Answer: An observational study comparing past exposures in subjects with a disease (cases) and without the disease (controls)

Q2. In a standard 2×2 table for an unmatched case–control study, how are rows and columns conventionally arranged?

• Rows for exposure status, columns for disease status
• Rows for disease status (cases, controls), columns for exposure status (exposed, unexposed)
• Rows for time periods, columns for drug doses
• Rows for controls only, columns for cases only

Correct Answer: Rows for disease status (cases, controls), columns for exposure status (exposed, unexposed)

Q3. Which formula gives the crude odds ratio (OR) from a 2×2 table with cells a (cases exposed), b (cases unexposed), c (controls exposed), d (controls unexposed)?

• OR = (a+d)/(b+c)
• OR = (a/c) / (b/d)
• OR = (a × d) / (b × c)
• OR = (a+b)/(c+d)

Correct Answer: OR = (a × d) / (b × c)

Q4. Given: cases exposed = 40, cases unexposed = 60; controls exposed = 20, controls unexposed = 80. What is the odds ratio?

• 1.33
• 2.67
• 0.75
• 4.00

Correct Answer: 2.67

Q5. In a matched case–control study, which statistic estimates the matched OR using pairwise discordant counts?

• Matched OR = (concordant pairs) / (total pairs)
• Matched OR = (number of pairs where case exposed & control unexposed) / (number of pairs where case unexposed & control exposed)
• Matched OR = (a × d) / (b × c) same as unmatched
• Matched OR = prevalence in cases / prevalence in controls

Correct Answer: Matched OR = (number of pairs where case exposed & control unexposed) / (number of pairs where case unexposed & control exposed)

Q6. When does the odds ratio approximate the relative risk (RR) closely?

• When the outcome is common (>20%)
• When the outcome is rare in the study population
• When the exposure prevalence is 50%
• Only in matched designs

Correct Answer: When the outcome is rare in the study population

Q7. An OR of 3.5 from a case–control study indicates which of the following?

• Exposure causes the disease in 3.5 people out of 10
• Those exposed have 3.5 times the odds of disease compared to unexposed
• The risk (incidence) is 3.5 times higher among exposed
• There is no association between exposure and disease

Correct Answer: Those exposed have 3.5 times the odds of disease compared to unexposed

Q8. What is a nested case–control study?

• A case–control study nested within a defined cohort where cases and matched controls are sampled from cohort members
• A case–control study that exclusively uses historical medical records without a cohort base
• A cross-sectional design measuring exposure and outcome at a single time point
• A randomized selection of cases from multiple cohorts

Correct Answer: A case–control study nested within a defined cohort where cases and matched controls are sampled from cohort members

Q9. Which sampling scheme yields an OR that estimates the incidence rate ratio (hazard ratio) when controls are selected using incidence density sampling?

• Choosing controls who never develop the disease during follow-up
• Selecting controls from survivors at the end of follow-up only
• Incidence density (risk-set) sampling where controls are selected at the time each case occurs
• Random sampling of the entire population at baseline

Correct Answer: Incidence density (risk-set) sampling where controls are selected at the time each case occurs

Q10. Which bias is especially a concern in retrospective case–control studies of drug exposure and outcomes?

• Performance bias
• Recall bias
• Detection bias unrelated to exposure
• Allocation bias

Correct Answer: Recall bias

Q11. For valid inference in an unmatched case–control study, controls should ideally represent:

• The exposure distribution in the population that produced the cases
• Only healthy individuals with no exposure history
• A higher exposure prevalence than cases to increase power
• Individuals matched on outcome severity

Correct Answer: The exposure distribution in the population that produced the cases

Q12. In case–control studies, the attributable fraction among the exposed (AFE) can be estimated from the OR using which formula?

• AFE = (OR) / (OR + 1)
• AFE = (OR − 1) / OR
• AFE = (OR − 1) / (OR × prevalence)
• AFE = OR − 1

Correct Answer: AFE = (OR − 1) / OR

Q13. If the 95% confidence interval for an odds ratio includes 1.0, what is the appropriate conclusion at alpha = 0.05?

• The association is statistically significant
• The null hypothesis of no association cannot be rejected
• The OR must be exactly 1.0
• The sample size is sufficient regardless of CI width

Correct Answer: The null hypothesis of no association cannot be rejected

Q14. Which statistical method is most appropriate for estimating associations in individually matched case–control studies with multiple covariates?

• Ordinary least squares regression
• Conditional logistic regression
• Unconditional Poisson regression without adjustments
• Kaplan–Meier survival analysis

Correct Answer: Conditional logistic regression

Q15. Increasing the number of controls per case beyond which ratio typically yields diminishing returns in statistical power?

• 1:1
• 2:1
• 4:1
• 10:1

Correct Answer: 4:1

Q16. Given: cases exposed = 12, cases unexposed = 8; controls exposed = 6, controls unexposed = 14. What is the crude OR?

• 1.75
• 3.50
• 0.57
• 2.00

Correct Answer: 3.50

Q17. In a matched study there are 25 discordant pairs: in 20 pairs the case is exposed and control unexposed (b=20); in 5 pairs the case is unexposed and control exposed (c=5). What is the matched OR?

• 0.25
• 1.00
• 4.00
• 5.00

Correct Answer: 4.00

Q18. When is the odds ratio likely to substantially overestimate the relative risk?

• When the outcome is rare (<1%)
• When the outcome is common (for example >10–20%)
• When exposure is non-differentially misclassified
• Only in matched designs

Correct Answer: When the outcome is common (for example >10–20%)

Q19. Which measure cannot be directly calculated from a conventional (unmatched) case–control study?

• Odds ratio
• Exposure odds among controls
• Absolute incidence (risk) of disease in the population
• Attributable fraction among the exposed using OR

Correct Answer: Absolute incidence (risk) of disease in the population

Q20. Given: cases exposed = 50, cases unexposed = 50; controls exposed = 30, controls unexposed = 70. What is the odds ratio and its qualitative interpretation?

• OR = 1.00; no association between exposure and disease
• OR = 2.33; exposure associated with increased odds of disease
• OR = 0.43; exposure protective against disease
• OR = 3.50; exposure associated with much higher odds of disease

Correct Answer: OR = 2.33; exposure associated with increased odds of disease

Download