Comparison of criteria for content uniformity

A two one-sided parametric tolerance interval test for control of delivered dose uniformity. Part 1--characterization of FDA proposed test

The FDA proposed a parametric tolerance interval (PTI) test at the October 2005 Advisory Committee meeting as a replacement of the attribute (counting) test for delivered dose uniformity (DDU), published in the 1998 draft guidance for metered dose inhalers (MDIs) and dry powder inhalers (DPIs) and the 2002 final guidance for inhalation sprays and intranasal products. This article (first in a series of three) focuses on the test named by the FDA "87.5% coverage." Unlike a typical two-sided PTI test, which controls the proportion of the DDU distribution within a target interval (coverage), this test is comprised of two one-sided tests (TOST) designed to control the maximum amount of DDU values in either tail of the distribution above and below the target interval. Through simulations, this article characterizes the properties and performance of the proposed PTI-TOST under different scenarios. The results show that coverages of 99% or greater are needed for a batch to have acceptance probability 98% or greater with the test named by the FDA "87.5% coverage" (95% confidence level), while batches with 87.5% coverage have less than 1% probability of being accepted. The results also illustrate that with this PTI-TOST, the coverage requirement for a given acceptance probability increases as the batch mean deviates from target. The accompanying articles study the effects of changing test parameters and the test robustness to deviations from normality.

Specification of limits for particulate contamination in pharmaceutical dosage forms

Limits for the control of particulate contamination in large volume parenteral solutions and metered-dose aerosols are discussed. It is suggested that is would be desirable to use limits based on measurement of both mean and the standard deviation of the particle counts obtained for each of the containers tested. Use of the statistic ST, assuming a target value of zero, is considered to be an appropriate means of measuring the container to container variation in particulate contamination.

Uniformity of content requirements for tablets and capsules

Present pharmacopoeial requirements for uniformity of content of tablets and capsules are based on tests by attributes. A test for uniformity of content based on the root mean square deviation about target (ST) is more reliable than the pharmacopoeial tests and has the added convenience of using a single statistic to control both the mean content and the variation about the mean. Implications of the use of the statistic ST in uniformity requirements are discussed.