Comparative and Veterinary Pharmacology

veterinary chemical product. In both situations, residues deplete to below the LOQ
of the analytical method. An ESI which is shorter than the withdrawal period is
never assigned as it would jeopardise food safety.
Before an ESI can be estimated for a veterinary chemical product, the risk
assessment policy must determine the level of lot rejection considered acceptable.
This first step in the ESI-setting procedure takes into account trade data for meat
and edible offal and the level of risk acceptable to major stakeholders (producer
organisations, meat processors, the veterinary pharmaceutical industry and govern-
ment). The second step utilises a suite of algorithms to compute the probability of
lot rejection of meat consignments when the treated animals are slaughtered at
various intervals after treatment. The ESI is computed as the time when the upper
tolerance limit about the regression line for the censored data intersects with the
residue concentration associated with the acceptable level of risk. Where no MRL
has been established in the importing country, the endpoint for the ESI determina-
tion is taken to be the LOQ of the method of analysis. Since the relationship
between the probability of lot rejection for meat consignments and ESI values is
displayed graphically (Fig. 5 ), the risk manager’s task of assigning ESIs and the risk
communicator’s task of conveying to stakeholders the decisions taken and the basis
for those decisions are greatly facilitated.

Time after last treatment

Probability of lot rejection MRL

LOQ

Upper

confidence

limit

Probability of lot rejection

ESI

Fig. 5Schematic diagram of the relationship between the probability of lot rejection of meat
consignments and the time after the last treatment of animals with an approved veterinary product.
The risk assessment policy takes into account the level of lot rejection considered acceptable. The
export slaughter interval (ESI) is computed as the time when the upper confidence limit about the
fitted regression line intersects with the residue concentration associated with the acceptable level
of risk

284 P.T. Reeves