Which of the following explains a 'moderator' variable?

In biostatistics and clinical research, variables are analyzed to determine the complexity of causal relationships. While independent variables influence outcomes, other factors can significantly impact the nature of that influence. Identifying these factors helps researchers understand for whom or under what conditions an intervention is effective. This distinction is critical for stratified analysis and the development of personalized medicine based on specific patient characteristics. Rationale: A. A variable not included in the analysis that leads to residual confounding is known as a confounder or a lurking variable. Confounders distort the true relationship because they are associated with both the exposure and the outcome. This is different from a moderator, which is typically measured and analyzed to understand interaction effects. Neglecting such variables leads to systematic bias. B. An intermediate variable that links a primary independent variable to a dependent variable is defined as a mediator. Mediators explain "why" or "how" an effect occurs by providing a biological or logical pathway. For example, exercise decreases heart disease through the mediator of weight loss. A moderator, however, does not link the two; it changes the interaction intensity. C. A moderator variable is a factor that influences the strength or direction of the effect between the independent and dependent variables. It answers the question of "when" or "for whom" a relationship exists. For example, a drug might work better in men than in women; here, gender is the moderator. This represents a statistical interaction effect that refines clinical findings.