Efficient Computing for One and Two Variance Components Parametric Tolerance Interval Testing. (3rd April 2019)

• Record Type: Journal Article
• Authors: Novick, Steven; Hudson-Curtis, Buffy
• Is Part Of: Statistics in biopharmaceutical research. Volume 11:Number 2(2019)
• Page Start: 146
• Page End: 151
• Publication Date: 2019-04-03
• Subjects: Content uniformity -- Generalized pivotal quantity -- Sufficient statistics; Pharmacy -- Statistical methods -- Periodicals; Pharmaceutical biotechnology -- Statistical methods -- Periodicals; Biopharmaceutics -- Periodicals; Biometry -- Periodicals; Pharmacy -- Statistical methods; Periodicals; 615.190727
• DOI: 10.1080/19466315.2018.1447994
• Languages: English
• ISSNs: 1946-6315

Abstract

ABSTRACT: In biopharmaceutical settings, testing of content uniformity measures the amount of active ingredient in solid dosage units of a manufactured batch. Though content uniformity has historically been evaluated via USP⟨905⟩, the two one-sided parametric tolerance interval test (PTI-TOST) has emerged as a viable alternative. In this work, sufficient statistics for the PTI-TOST are developed for two-tiered testing and are shown to dramatically speed up the computer time spent generating operating characteristic curves. While the original PTI-TOST assumes that the data are independent and identically distributed Gaussian random variables, a two-variance components PTI-TOST has been established for the situation in which solid dosage units are sampled in strata (location). Sufficient statistics were also developed for the two-tier, two-variance components PTI-TOST. Computer simulations evaluated the computational time savings from using the efficient algorithms suggested and discussed in this work.