Theoretical Basis for the EDA Concept - Good Cascade Impactor Practices, AIM and EDA for Orally Inhaled Products - page 149

Fig. 7.1 Potential distributional properties of a quality attribute in relationship to quality limits:
case A reflects a nominal distribution of a batch with acceptable quality, case B illustrates a quality
failure due to an abnormal batch mean, case C reflects a quality failure due to abnormal variability,
and case D illustrates a situation where conventional (representation) end-product testing is
unlikely to detect the quality failure

More sophisticated schemas can include tiered testing or simultaneous evalua-
tion of mean and variance. In all cases, the decision-making capability of any par-
ticular schema can be evaluated through an operating characteristic curve (OCC [ 16 ]).
In essence, an OCC is a transfer function that relates the probability of a particular
decision (accept or reject) to true values of the quality attribute being evaluated.
“True” in this context refers to the population parameter that is estimated by mea-
surement of samples. OCCs are also discussed in more detail in Chap. 8 .

7.4 Fundamental Properties of the APSD

Back in Chap. 2 , the idea was presented that data from a CI are not to be linked
directly to specific regions within the HRT. However, CIs by virtue of having a
number of stages in series, each of which acts as a size fractionator to the incoming
aerosol particles, are capable of providing API-linked, mass-weighted APSDs when
combined with appropriate analytical assay method(s) for the API(s) emitted by the
OIP whose in vitro performance is being investigated [ 17 ].
It is important to note several limitations and trade-offs with using the CI
approach to characterizing particle size. First, the resolution of the histogram