Distinguish confounding from effect modification.

• A. They are the same concept.
• B. Confounding distorts the exposure-outcome relationship; effect modification (interaction) means the effect differs across levels of another variable.
• C. Confounding is the same as bias; effect modification is randomness.
• D. Confounding occurs only in cross-sectional studies.

Answer: (B)

When you’re looking at how the exposure relates to an outcome, a confounder is a third variable that is linked to both the exposure and the outcome and is not on the causal path. This link can distort the true relationship you’re trying to estimate. Because of that, you adjust for the confounder to remove the distortion and get a clearer estimate of the exposure’s effect.

Effect modification, or interaction, is a different story. Here, the impact of the exposure on the outcome actually changes depending on the level of another variable. It’s not a distortion to be removed; it’s a real difference in effect across subgroups. If you find evidence of interaction, you report the effect separately for each level of the modifying variable rather than just giving a single overall effect.

So the best description is that confounding distorts the exposure–outcome relationship, while effect modification means the effect differs across levels of another variable.