Measurement Uncertainty for the Pharmaceutical Industry

Recommendation

27-29 February 2024

GMP meets GCP - Live Online Training
Management, Supply and Quality Assurance of Clinical Trials

A Stimuli Article regarding measurement uncertainty (MU) has been published in Pharmacopeial Forum 44(1), January 2018. MU is a component of the lifecycle approach to analytical procedures and the purpose of the Stimuli Article is to introduce MU to the pharmaceutical industry. According to the authors the article "is intended for the people in the pharmaceutical industry who generate reportable values, as well as those who use the reportable values—regulatory affairs, pharmaceutical technology, production, and quality assurance".

The Stimuli Article

• presents concepts, explains their meaning, and compares them to existing terms and concepts in the pharmaceutical industry,
• collects and presents definitions and discussions of terms and concepts in an Appendix,
• describes the process for evaluating measurement uncertainty including how to report MU,
• provides an example for a typical pharmaceutical analysis, and
• contains references that include many worked examples. 

The International Organization for Standardization (ISO) standard ISO 21748:2017 states:

"Knowledge of the uncertainty associated with measurement results is essential to the interpretation of the results. Without quantitative assessments of uncertainty, it is impossible to decide whether observed differences between results reflect more than experimental variability, whether test items comply with specifications, or whether laws based on limits have been broken. Without information on uncertainty, there is a risk of misinterpretation of results. Incorrect decisions taken on such a basis may result in unnecessary expenditure in industry, incorrect prosecution in law, or adverse health or social consequences."

MU, in particular, the target measurement uncertainty (TMU), is a key component of the lifecycle approach to analytical procedures now being adopted in the pharmaceutical industry. TMU is the maximum value for the uncertainty associated with a reportable value in order for the reportable value to be fit for purpose. In the analytical target profile (ATP), TMU helps define the quality required in the reportable value (e.g. content of API in a tablet) produced by the analytical procedure. The decision rule provides the acceptable probability of being wrong when making the decision. MU is required to determine the probability stated in a decision rule. TMU is a key performance criterion for analytical procedure performance throughout its life, from development through qualification and continued performance verification. Additionally, the evaluation of MU is a component of risk analysis according to ICH Q9 because it expresses an event (i.e., the true value of the measurand quantity) probabilistically.

What does it mean for analytical procedures in the pharmaceutical industry? 

For analytical procedures, many of the requirements and much of the data for evaluating MU may already exist:

• Accuracy studies provide information on the magnitude of the systematic error;
• Repeatability, intermediate precision, and reproducibility studies provide data on precision (the random error component of uncertainty);
• The clear definition of measurand reduces the definitional uncertainty;
• Focus on eliminating the risk of mistakes and reduction of definitional uncertainty caused by not knowing the true value;
• Bayesian probability concepts and methodology enable and assist risk analysis and informed decision making.

The article emphasizes that "it is important that measurement results are reported with sufficient knowledge and information to support the decisions made with them. The availability of expanded uncertainty accompanying a reportable value makes it possible to determine the probabilities of decision errors based on that reportable value" (for example: false release or rejection of products). The article shows that the pharmaceutical industry has much of the data and understanding in place and that a reframing of the data to include all uncertainty components might improve the quality of reportable values.

After registration to the Pharmarcopeial Forum you get access to the complete stimuli article "Measurement Uncertainty for the Pharmaceutical Industry".