There exists the possibility that non-target livestock may receive trace exposure to medications in feed due to residue carryover from previous production runs of medicated feeds at feed mills. We have developed a method by which ADI-Derived Drug Carryover Levels (ADCLs) can be established. It is a practical approach compared to the "zero" levels of residue carryover that may be expected or required by regulatory authorities. The methodology involves application of various safety/uncertainty factors to concentrations of active ingredients (a.i.) already approved for use in medicated feeds for target species. The starting point for each a.i., to be consistent, and to represent the highest possible carryover, is the highest approved concentration for any target animal species, recognizing that this is an approved level based on established ADI and agency review of supporting safety data specific to each a.i. (Hence, these guidance values are characterized to be 'ADI-derived'.) Safety factors are then applied to account for: (a) interspecies extrapolation, (b) differences in the body weights of target and non-target species (i.e., smaller animals receive higher exposures on a body weight basis for a given dietary concentration), (c) a.i. with clear contraindications for use in certain non-target species (i.e., a priori knowledge of non-target species sensitivity), and (d) withdrawal times (i.e., for a.i. that require a washout period prior to slaughter there is potential exposure to non-target species through other feeds not requiring a washout period). The values of the safety/uncertainty factors range from 1 to 3, 1 to 3.17, 1 to 10, and 1 to 10, for each of conditions (a), (b), (c), and (d), respectively. The "proposed safety factor" to apply to the approved concentration in medicated feed is calculated as the product of the values for each of (a) through (d). The final safety factor is the greater of the proposed safety factor or a default minimum safety factor of 30. ADCLs were calculated for several a.i. and compared to limits of quantitation available for detection of carryover residues in animal feeds. This methodology may be used in its present or modified form in any jurisdiction in which mediated feeds are approved. As a start, this approach has been applied to several example products approved and in use in Canada.