Time–risk relationship in pharmacoepidemiology MCQs With Answer

Introduction: This set of MCQs focuses on the time–risk relationship in pharmacoepidemiology, a core topic for M.Pharm students studying how timing of drug exposure influences adverse events and therapeutic outcomes. The questions cover induction and latency periods, time-at-risk definitions, time-varying exposures and covariates, survival analysis principles, and common biases such as immortal time and time-dependent confounding. Emphasis is on study design choices, appropriate analytical methods (e.g., Cox models, self-controlled designs), and interpretation of temporal risk windows. These questions aim to deepen understanding of how temporal considerations shape pharmacoepidemiologic inference and affect regulatory, clinical, and safety evaluations.

Q1. What does the ‘induction period’ refer to in pharmacoepidemiology studies?

• The time between drug exposure and study start date
• The time between onset of drug effect and clinical detection of outcome
• The time between first exposure and the biological initiation of disease process
• The time between last exposure and resolution of adverse event

Correct Answer: The time between first exposure and the biological initiation of disease process

Q2. In pharmacoepidemiologic time–risk analyses, a ‘latency period’ is best described as:

• The period when a medication is being manufactured
• The time from biological initiation of disease to clinical diagnosis or observable event
• The time during which adherence is measured
• The washout period used to exclude prior exposures

Correct Answer: The time from biological initiation of disease to clinical diagnosis or observable event

Q3. Which bias arises when exposed person-time before initiation of exposure is misclassified as exposed, often creating an artificial protective effect?

• Confounding by indication
• Selection bias
• Immortal time bias
• Information bias

Correct Answer: Immortal time bias

Q4. In time-to-event analyses, what does the hazard function represent?

• The cumulative probability of surviving to a fixed time point
• The instantaneous event rate at time t among individuals event-free just before t
• The average follow-up time across a cohort
• The total number of events divided by number of subjects

Correct Answer: The instantaneous event rate at time t among individuals event-free just before t

Q5. When investigators define a ‘risk window’ after drug exposure, what is their primary aim?

• To restrict analysis only to adherent patients
• To identify the period during which an exposure plausibly influences outcome occurrence
• To increase sample size by lengthening follow-up arbitrarily
• To remove covariate imbalance between groups

Correct Answer: To identify the period during which an exposure plausibly influences outcome occurrence

Q6. Which analytic method explicitly handles time-varying exposures and covariates in survival analysis?

• Standard logistic regression without time terms
• Cox proportional hazards model with time-dependent covariates
• Simple t-test comparing means
• Chi-square test for independence

Correct Answer: Cox proportional hazards model with time-dependent covariates

Q7. In a self-controlled case series (SCCS), how is the time–risk relationship typically handled?

• By comparing exposed individuals to an external unexposed cohort
• By using within-person comparisons of predefined risk and control periods
• By randomizing exposure times across subjects
• By excluding all person-time after the first event

Correct Answer: By using within-person comparisons of predefined risk and control periods

Q8. Which approach is commonly used to address time-dependent confounding when treatment changes over time?

• Propensity score matching using baseline covariates only
• Marginal structural models with inverse probability of treatment weighting
• Cross-sectional regression at baseline
• Complete-case analysis ignoring time-varying factors

Correct Answer: Marginal structural models with inverse probability of treatment weighting

Q9. What is the primary purpose of introducing a lag period (exposure lagging) in pharmacoepidemiology?

• To extend person-time to include remote exposures
• To exclude early events unlikely caused by exposure and reduce reverse causation
• To randomize exposure assignment
• To simplify analysis by ignoring time at risk

Correct Answer: To exclude early events unlikely caused by exposure and reduce reverse causation

Q10. ‘Person-time’ is best defined as:

• The total number of persons included in the eligibility criteria
• The sum of individual follow-up times during which participants are at risk
• The time between baseline and study publication
• The average duration of exposure among exposed individuals

Correct Answer: The sum of individual follow-up times during which participants are at risk

Q11. In studies of cumulative dose-response over time, what exposure metric is most informative?

• Binary ever/never exposure indicator only
• Cumulative dose or duration of exposure accounting for time
• Baseline blood pressure only
• Number of study visits attended

Correct Answer: Cumulative dose or duration of exposure accounting for time

Q12. Which design is particularly useful for assessing short-term risks after transient exposures and inherently accounts for time-invariant confounders?

• Case-control design with population controls
• Self-controlled case series (SCCS) or case-crossover design
• Cohort study with baseline matching only
• Cross-sectional survey

Correct Answer: Self-controlled case series (SCCS) or case-crossover design

Q13. How can ‘immortal time’ be prevented in cohort analyses of drug safety?

• By assigning exposure status based on future events
• By using time-fixed exposure measured at baseline only
• By using time-dependent exposure classification that starts person-time at true exposure onset
• By excluding all exposed subjects

Correct Answer: By using time-dependent exposure classification that starts person-time at true exposure onset

Q14. What is a ‘washout period’ used for in pharmacoepidemiologic time–risk studies?

• To ensure complete capture of outcomes after study ends
• To exclude subjects recently exposed to similar drugs before cohort entry
• To randomize treatments
• To increase the number of events during follow-up

Correct Answer: To exclude subjects recently exposed to similar drugs before cohort entry

Q15. When might a spline function be employed in time–risk modeling?

• To model non-linear relationships between continuous time or dose and risk
• To eliminate the need for censoring
• To create categorical exposure groups only
• To impute missing baseline covariates

Correct Answer: To model non-linear relationships between continuous time or dose and risk

Q16. In competing risks settings, why must cause-specific hazards and cumulative incidence be interpreted differently?

• Because cause-specific hazards always equal cumulative incidence
• Because a competing event can preclude the occurrence of the event of interest, altering absolute risk over time
• Because competing risks only affect cross-sectional analyses
• Because they only apply to randomized trials

Correct Answer: Because a competing event can preclude the occurrence of the event of interest, altering absolute risk over time

Q17. What is the main challenge when using administrative claims data to define precise time-at-risk windows?

• High cost of data access
• Poorly recorded or imprecise timing of medication use and outcome onset
• Lack of demographic variables
• Too many laboratory values

Correct Answer: Poorly recorded or imprecise timing of medication use and outcome onset

Q18. Which of the following best describes time-dependent confounding?

• Confounding that is constant over time and fixed at baseline
• Confounding where prior treatment affects subsequent confounders that influence future treatment and outcome
• Confounding that only occurs in cross-sectional studies
• Confounding that can be ignored when sample size is large

Correct Answer: Confounding where prior treatment affects subsequent confounders that influence future treatment and outcome

Q19. In a pharmacoepidemiology study assessing immediate allergic reactions after vaccination, which risk-window definition is most appropriate?

• A one-year risk window after vaccination
• A short, prespecified risk window (e.g., 0–2 days) matched to biological plausibility
• Risk window based on calendar year only
• Risk window starting one month after vaccination to allow for latency

Correct Answer: A short, prespecified risk window (e.g., 0–2 days) matched to biological plausibility

Q20. Why is sensitivity analysis of different time-at-risk definitions important in time–risk studies?

• Because it guarantees statistically significant results
• Because varying the risk-window can test robustness of associations and identify misclassification or reverse causation
• Because it reduces the need for covariate adjustment
• Because it allows removing all censored observations

Correct Answer: Because varying the risk-window can test robustness of associations and identify misclassification or reverse causation

Download