An epidemiologist in Mumbai wants to study the relationship between oral contraceptive use and venous thromboembolism (VTE) in women aged 20–45 years. She has two options: (1) enroll 2,000 women without VTE, classify them by OCP use, and follow them for 5 years; (2) enroll 400 women with VTE and 400 women without VTE, and ask about past OCP use. Which design is most appropriate for this rare outcome, and why?

Explanation

## Study Design Selection for Rare Outcomes **Key Point:** For **rare outcomes** (VTE incidence ~1–3 per 10,000 person-years in young women), a **case-control study is far more efficient** than a cohort study. ## Why Case-Control is Optimal Here ### Efficiency Calculation **Cohort Study (Option 1):** - To detect a doubling of risk (RR = 2) with adequate power, you would need: - ~50,000–100,000 women followed for 5 years - Cost: very high; duration: 5+ years - Many participants contribute no events **Case-Control Study (Option 2):** - Enroll 400 cases (women with VTE) and 400 controls (women without VTE) - Ask about past OCP use - Cost: low; duration: months - Every participant contributes information about the outcome **High-Yield:** The **case-control design is the gold standard for rare outcomes** because it selects on disease status, ensuring adequate numbers of cases. ## Comparison: Cohort vs. Case-Control for Rare Outcomes | Feature | Cohort Study | Case-Control Study | |---------|--------------|--------------------| | **Sample size needed** | Very large (50,000+) | Moderate (400–800) | | **Follow-up duration** | Long (5+ years) | None (retrospective) | | **Can measure incidence?** | Yes | No (can calculate odds ratio) | | **Efficiency for rare outcomes** | Poor | **Excellent** | | **Cost** | High | Low | | **Time to completion** | Years | Months | **Clinical Pearl:** VTE is a rare but serious outcome in young women on OCPs. Case-control studies are routinely used in pharmacoepidemiology to detect rare adverse events efficiently. **Mnemonic: RARE** — **Rare outcomes require Appropriate study design; Case-control Excels.** ## Address the Recall Bias Concern While Option 1 claims "unaffected by recall bias," this advantage is outweighed by the impracticality of the cohort design. Recall bias in case-control studies can be minimized through: - Objective medical records review (hospital discharge summaries for VTE diagnosis and OCP use) - Standardized questionnaires - Validation studies These methods are far more feasible than enrolling and following 50,000+ women.