Methods for analyzing matched designs with double controls: excess risk is easily estimated and misinterpreted when evaluating traffic deaths

Available approaches provide methods for analyzing count data in matched designs with double controls and help clarify the distinction between increases in total risk and increases in excess risk.