Purpose

Antimicrobial activity of drug products containing antibiotics is often measured using microbiological assays. However, the high values of measurement uncertainty associated with the analytical results obtained from microbiological assays may be an issue to conformity decisions.

Methods

The aim of this work was to estimate the risk of false decisions in conformity assessment due to measurement uncertainty for the potency of apramycin in pharmaceutical drug products. Monte Carlo method (MCM) simulations were performed in order to estimate global consumers’ (R_c) and producers’ (R_p) risks using a Bayesian approach and specific consumers’ (R′_c) and producers’ (R′_p) risks using a frequentist approach.

Results

Despite of the high value of measurement uncertainty, R_c and R_p were found to be 0.0% and 0.3%, respectively. However, R′_c and R′_p were found to be high when the analytical result is close to the specification limits. Risk estimation using Bayesian approach is recommended to be applied by manufacturers, while frequentist approach may be an alternative to regulatory and third-party laboratories.

中文翻译：

---

基于测量不确定度的符合性决策—以琼脂扩散微生物分析为例

目的

含有抗生素的药物产品的抗菌活性通常使用微生物测定法进行测量。但是，与从微生物分析获得的分析结果相关的测量不确定性的高值可能是一致性决定的问题。

方法

这项工作的目的是评估由于阿普霉素在药物产品中效力的测量不确定性而导致的合格评估中错误决定的风险。蒙特卡洛方法（MCM）模拟，以便估计全球消费者（执行ř _Ç）和生产者（[R _p）风险使用贝叶斯方法和具体的消费者（[R ' _ç）和生产者（[R ' _p）的风险使用常客的方法。

结果

尽管测量不确定度值很高，但发现R _c和R _p分别为0.0％和0.3％。然而，[R ' _ç和- [R ' _p被认为是高时的分析结果是接近的规格界限。建议制造商采用使用贝叶斯方法进行风险评估，而频繁性方法可能是监管实验室和第三方实验室的替代方法。