Theoretical Basis for the EDA Concept - Good Cascade Impactor Practices, AIM and EDA for Orally Inhaled Products - page 145

Chapter 7
Theoretical Basis for the EDA Concept
Terrence P. Tougas and Jolyon P. Mitchell
Abstract Efficient Data Analysis (EDA) was designed specifically to address quality
control (QC) decisions with respect to the CI-measured APSD from an OIP. The
general goal of QC testing is to confirm that the batch in question is of suitable qual-
ity. In the case of EDA, this testing is intended to confirm that the OIP in question
generates an aerosol with expected particle size characteristics to deliver drug to the
human respiratory tract. Note that this process necessarily takes the form of sam-
pling a relatively small number of units, measuring properties of the aerosols gener-
ated by these samples, and making a decision concerning the quality of the sampled
batch. This practice leads to three primary considerations:
1. The properties measured should be relevant to detecting significant abnormali-
ties from the expected APSD.
2. The measurements should possess sufficient precision and accuracy over the
range of interest.
3. The decision process based on the measurements should reliably make correct
inference about the quality of the batch by appropriately minimizing and balanc-
ing the risk of decision errors, i.e., judging a batch suitable when it is not suitable
and conversely judging a batch unsuitable when it is suitable.
This chapter will briefly introduce the latter two considerations, but will primar-
ily focus on the first. A detailed discussion of the evaluation of measurements and
the decision-making process is the topic of Chap. 8 .