A case-control study in rural Maharashtra investigates the association between maternal anaemia and preterm birth. The Odds Ratio is calculated as 3.2 (95% CI: 2.1–4.8). All of the following interpretations are correct EXCEPT:

Explanation

## Case-Control Study: OR Interpretation in Indian Context ### Clinical Scenario Setup **Study Design:** Case-control (retrospective) - **Cases:** Preterm births (n = exposed and unexposed to anaemia) - **Controls:** Term births (n = exposed and unexposed to anaemia) - **Exposure:** Maternal anaemia - **OR = 3.2** (95% CI: 2.1–4.8) ### Why Option 3 is the Incorrect Statement **Key Point:** Odds Ratio does NOT represent "probability ratio." It represents the **odds ratio** — a fundamentally different concept. $$OR = \frac{\text{Odds of exposure in cases}}{\text{Odds of exposure in controls}} = \frac{a/b}{c/d}$$ where: - a = cases with exposure (preterm + anaemia) - b = cases without exposure (preterm + no anaemia) - c = controls with exposure (term + anaemia) - d = controls without exposure (term + no anaemia) **OR tells us:** Among the cases (preterm births), the odds of having been anaemic are 3.2 times the odds among controls (term births). **OR does NOT tell us:** The probability that an anaemic woman will have a preterm birth (that would be RR in a cohort study). **Warning:** A common NEET PG trap is confusing OR with RR or interpreting OR as a direct probability. Option 3 makes exactly this error by calling OR a "probability ratio." ### Why the Other Three Statements Are Correct | Statement | Why Correct | |-----------|-------------| | **Option 0** | Preterm birth prevalence ~5–8% meets the rare disease assumption (<10%). Therefore, OR ≈ RR. This is the foundation of case-control study validity. | | **Option 1** | If this were a cohort study, the same numerical OR (3.2) would actually represent RR. Because preterm birth is relatively common (~7%), the true RR would be lower than 3.2 due to the mathematical relationship between OR and RR at higher disease frequencies. | | **Option 2** | The 95% CI (2.1–4.8) does not include 1.0, confirming statistical significance at α = 0.05. This is a correct interpretation of confidence intervals. | ### Mermaid: OR vs RR Interpretation by Study Design ```mermaid flowchart TD A[Study Design]:::decision --> B{Cohort or<br/>Case-Control?} B -->|Cohort| C[Calculate RR directly]:::action B -->|Case-Control| D[Calculate OR]:::action C --> E[RR = Incidence Exposed<br/>/ Incidence Unexposed]:::outcome D --> F[Check disease<br/>prevalence]:::decision F -->|Rare &lt;10%| G[OR ≈ RR<br/>Valid interpretation]:::outcome F -->|Common &gt;10%| H[OR &gt;&gt; RR<br/>OR overestimates effect]:::outcome E --> I[Direct probability<br/>interpretation valid]:::outcome G --> J[Case-control valid<br/>surrogate for cohort]:::outcome ``` ### Clinical Pearl: Indian Epidemiology Context Maternal anaemia prevalence in rural India is ~50–60% — very common. If a cohort study were conducted with the same true association, the RR would be substantially lower than 3.2. The case-control OR of 3.2 still approximates RR for preterm birth (rare outcome), but would overestimate if we misinterpreted it as the probability of preterm birth given anaemia. **High-Yield:** Remember the distinction: - **RR (from cohort):** "Anaemic women are 3.2 times more likely to have preterm birth" - **OR (from case-control):** "Among preterm births, the odds of prior anaemia are 3.2 times the odds among term births" These sound similar but are mathematically distinct. OR approximates RR only when disease is rare. [cite:Park 26e Ch 3; Epidemiology: Gordis 6e]