Power of Analysis of Variance for Assessing Batch-Variation of Stability Data of Pharmaceuticals

概要

論文の詳細を見る
The present study investigates the effect of batch-variation on the evaluation of stability data obtained from three batches of pharmaceutical products. Using stability data generated by the Monte Carlo method, the power of analysis of variance for assessing batch-variation was determined as functions of significance level and assay error.The results obtained from degradation data simulated for a product exhibiting zero-order degradation of 0.2%/month indicate that a significance level of approximately 0.25 is necessary in order to obtain a minimum detectable difference of 25% if the beta error is fixed at 20%. Assay methods with smaller margins of error are recommended, as is repeating the assay in order to reduce assay error down to approximately 0.5% standard deviation. In addition, the validity of the 0.25 significanece level, which is recommended in the ICH (International Conference on Harmonisation of Technical Requirements for the Registration of Pharmaceuticals for Human Use) Tripartite guideline, is discussed.
公益社団法人日本薬学会の論文
1996-10-15