Statistical considerations for using tolerance interval to set product specification for normally distributed...

Area of Science:

• Pharmaceutical Sciences
• Statistical Quality Control

Background:

• Traditional quality specifications rely on normal distribution assumptions and 3-sigma limits, which may not be precise.
• The k-content tolerance interval is commonly used but can be inaccurate with small sample sizes.
• Existing methods for determining tolerance intervals may lead to overestimation, especially with large coverage percentages.

Purpose of the Study:

• To develop a precise method for determining drug product quality specifications using tolerance intervals.
• To address the imprecision of k-content tolerance intervals with small sample sizes.
• To propose a new precision requirement to avoid overestimation and ensure accurate quality limits.

Main Methods:

• Defined a precision requirement for tolerance intervals based on a pre-specified significance level.
• Utilized the Faulkenberry and Daly
• goodness
• criterion for sample size determination.
• Determined proper coverage percentage (p) and minimum sample sizes for precise k-content tolerance intervals.

Main Results:

• The proposed method ensures precise k-content tolerance intervals, even with small sample sizes.
• Accurate quality specifications can be set by controlling the probability of overestimation.
• Minimum sample sizes were determined to meet the "goodness" criterion for desired coverage and confidence levels.

Conclusions:

• The new approach allows for the proper setting of product specifications by avoiding over-specification of quality limits.
• This method enhances the reliability of quality control in pharmaceutical manufacturing.
• It provides a statistically sound framework for establishing precise drug product quality standards.