Attributable Risk vs Absolute Risk Reduction


I am hoping to understand the difference between attributable risk and absolute risk reduction. Mathematically, the two entities seem equivalent, but the formulas for calculating their confidence interval are quite different ( interval/CI for relative risk.htm, interval/CI for ratio.htm). I presumed that attributable risk is used in prospective studies in the context of discussing risk ratio, while ARR is used in retrospective studies in the context of discussing odds ratio.

I appreciate your feedback and comment.