Statistical Properties of Large Sample Tests for Dose Content Uniformity

• Activity Number: 37
• Type: Contributed
• Date/Time: Sunday, August 3, 2014 : 2:00 PM to 3:50 PM
• Sponsor: Quality and Productivity Section
• Author(s): Meiyu Shen*+ and Yi Tsong and Xiaoyu Dong
• Companies: FDA and FDA and FDA
• Keywords: dose content uniformity ; large sample sizes ; tolerance interval ; OC curve ; acceptance probability

Abstract

Test for uniformity of dosage units using large sample sizes in European Pharmacopoeia 7.7 consists of 2 tests. EU Option 1 is a parametric two-sided tolerance interval based method modified with an indifference zone and counting units outside of(0.75M,1.25M), here Xbar=sample mean and M=98.5% if Xbar< 98.5%, M=101.5% if Xbar>101.5%,and M=xbar o.w. EU Option 2 is a nonparametric counting method with an indifference zone. We extended the parametric two one-sided tolerance intervals based method proposed by Tsong for dose content uniformity test with 30 tablets and by Shen & Tsong for large sample sizes with the restriction that all operating characteristic (OC)curves of two one-sided tolerance intervals for any given sample size intersect with OC curve of USP harmonized method for 30 tablets at 90% acceptance probability assuming data normally distributed with 100%LC mean. We compared the acceptance probabilities of EU Options 1 and 2 and our proposed method by simulation and found that both EU methods produce larger acceptance probabilities for off target mean. For a given variability, the acceptance probability of EU option 2 at mean of 102%LC is larger than that at mean of 100%LC.