In a cohort study investigating the association between oral contraceptive use and venous thromboembolism, smoking status is found to be associated with both the exposure and outcome. What type of epidemiological error is smoking status in this scenario?

Explanation

## Identifying Confounders in Epidemiological Studies **Key Point:** A confounder must satisfy three criteria: (1) association with the exposure, (2) association with the outcome, and (3) not in the causal pathway between exposure and outcome. ### Criteria for Confounding 1. **Associated with the exposure** — Smoking is related to oral contraceptive use (smokers may have different patterns of contraceptive use) 2. **Associated with the outcome** — Smoking is an independent risk factor for venous thromboembolism (VTE) 3. **Not in the causal pathway** — Smoking does not cause oral contraceptive use; it is an independent variable ### Why Smoking is a Confounder Here ```mermaid flowchart TD A[Smoking Status]:::outcome --> B[VTE Risk]:::outcome C[OCP Use]:::outcome --> B A --> D[Confounding Effect]:::urgent C --> D D --> B ``` Smoking creates a spurious association: the observed association between OCP use and VTE may be partly due to smoking, not solely due to OCP use. ### Control of Confounding - **Matching** — Select smokers and non-smokers equally across OCP users and non-users - **Stratification** — Analyze the association separately in smokers and non-smokers - **Multivariate analysis** — Adjust for smoking in a regression model **High-Yield:** The classic confounder in epidemiology is one that is **associated with both exposure and outcome but not in the causal chain**. Smoking is a textbook example in drug safety studies. **Mnemonic:** **CONFOUNDER** — **C**aused by neither, **O**ccurs with both, **N**ot in pathway, **F**akes association, **O**bscures true effect, **U**nrelated causally, **N**eeds control, **D**istorts results, **E**xtraneous, **R**eal association.