to the physicochemical properties of the pure active ingredient, the purity of the test
substance, and any impurities present.
2.1.1 Hazard Identification
The purpose of this step is, firstly, to establish the identity of the drug molecules that
are present as residues in animal-derived foods and which are capable of causing
adverse effects on consumer health. Secondly, this step provides information on the
effects that are considered to be adverse. Although the steps involving hazard
identification and hazard characterisation often source their information from the
same studies, the two steps assess different information. The preferred sources of
information for the hazard identification step are human epidemiological studies,
animal toxicological studies, in vitro assays, and quantitative structure–activity
relationships (WHO 1995 ). Most commonly, a battery of tests in laboratory animals
is conducted. It includes single dose toxicity studies, repeated dose (90-day)
toxicity studies, reproductive toxicity studies, developmental toxicity studies, and
long-term toxicity and carcinogenicity studies. A battery of genotoxicity assays is
also conducted. The endpoints of these studies are the basis for identifying a No
Observed Effect Level (NOEL), or a No Observed Adverse Effect Level (NOAEL),
or a benchmark dose (BMD).
2.1.2 Hazard Characterisation
The hazard characterisation step in the risk assessment of veterinary drugs primarily
focuses on the dose–response relationship for critical adverse effects, identification
of the most sensitive animal species or strain for a given effect, and extrapolation
from animals to humans. The traditional approach to toxicology data analysis
assumes that a threshold dose can be identified in the dose–response relationship
for a substance (situations where a threshold cannot be identified in the dose–
response relationship are discussed below). At sub-threshold doses, there are no
significant increases in the frequency or severity of the effect between the treated
groups and the control groups. The NOEL is the highest dose that does not cause
any detectable effect in the most sensitive species of test animals. Alternative
approaches for characterising hazards have been reported (Crump 1984 ; Kroes
and Kozianowski 2002 ); the BMD is one such approach. The BMD approach
models all the available dose–response data and provides increased weighting to
data near the observed dose–response range by using lower effective doses at, for
example, levels causing 10% and 5% of a specified response (ED 10 and ED 05 ). Safety
factors are applied when deriving the BMD-based acceptable daily intake (ADI).
The ADI is the amount of chemical that may be consumed daily for an entire
lifetime without causing an appreciable risk to human health. It is calculated by
dividing the appropriate NOEL by a safety factor to account for the uncertainties
encountered when extrapolating animal toxicity data to potential effects in humans
Drug Residues 269