**Core Concept**
Confounding is a type of bias that occurs when a third variable (confounder) affects both the independent variable and the dependent variable, leading to an incorrect association between the two. It's essential to identify and control confounding variables in observational studies to ensure that the results accurately reflect the relationship between the variables being studied.

**Why the Correct Answer is Right**
Confounding cannot be removed by simply ignoring it or not adjusting for it in the analysis. If confounding is present, it can lead to biased estimates of the effect size and potentially incorrect conclusions. The only way to remove confounding is to either match or stratify on the confounder, use a regression analysis that controls for the confounder, or use a study design that minimizes the effect of the confounder, such as a randomized controlled trial.

**Why Each Wrong Option is Incorrect**
**Option A:** This option is incorrect because controlling for confounding is a crucial step in observational studies to ensure that the results accurately reflect the relationship between the variables being studied.
**Option B:** This option is incorrect because ignoring confounding can lead to biased estimates of the effect size and potentially incorrect conclusions.
**Option C:** This option is incorrect because matching or stratifying on the confounder is a way to remove confounding, not an example of something that cannot be done.

**Clinical Pearl / High-Yield Fact**
Remember that confounding is a type of bias that can occur in any observational study, and it's essential to identify and control confounding variables to ensure that the results accurately reflect the relationship between the variables being studied.

**Correct Answer: B. Ignoring confounding can lead to biased estimates of the effect size and potentially incorrect conclusions.**