Dr Raphael (Raphy) Bar, BR Consulting, Israel
Dr Gerald Kindermann, F. Hoffmann-La Roche, Basel, Switzerland
Philip Lienbacher, Shire, Vienna, Austria
The aim of this course is to discuss the process of the statistical sampling by attributes of units of finished drug products, of packaging materials (primary and secondary) and of medical devices as well as sampling of starting materials (APIs and excipients) and to define the prerequisites for implementing a system for reduced sampling and reduced testing for these products.
This course is also intended to give a practical training on the use of the most common sampling standards: ISO 2859-1:1999 and ANSI/ASQ Z1.4. These Standards are widely employed in various types of industries which are required by Regulatory Authorities to follow a statistically sound sampling plan. Starting with regulatory and compendial requirements around sampling, this course will also address charting and trending nonconformities and non-conformant items and good quality practice around sampling plans.
Sampling of materials is one of the most important processes in pharmaceutical companies. Today there are more and more detailed questions during regulatory GMP Inspections, both in Europe and in the US (FDA) about the amount of samples to be taken.
Sampling by Attributes is a process of inspecting a representative sample of identical product units of product for presence or absence of non-conforming units or nonconformities before accepting or rejecting the whole lot of product. Regulatory agencies require a sampling plan that utilizes basic elements of statistical analysis or provide a scientific rationale for taking a representative sample according to the lot size.
In the light of these regulatory requirements, one may wonder whether the Square Root of N is a statistically valid sampling plan.
According to the revised Chapter 6 of EU GMP Guide, the sampling plan used should be appropriately justified and based on a risk management approach.
Representative samples should be taken and recorded in accordance with approved written procedures.
FDA requires as well in the Code of Federal Regulations (21 CFR Part 211.84), that sampling should be done upon statistical criteria.
In the past the Military Standard 105 D was commonly used in the pharmaceutical industry, but this standard has been withdrawn and is now obsolete. Today. either the ISO Standard 2859:1-1999 or the ANSI Z1.4 are applied and the course participants will thus learn how to read and to use this standard for selecting a sampling plan with an understanding of the associated producer and consumer risks.
This GMP Education Course is directed at all those employees from quality control units and production units in the pharmaceutical industry who are competent or responsible for sampling, testing and release of starting materials (APIs and excipients), packaging materials (primary and secondary) as well as finished pharmaceutical products. This course is also of interest to personnel from quality assurance and to those employees from API, excipient or packaging material suppliers who want to inform themselves about the requirements of the pharmaceutical industry on the testing of these materials.
The course does not require prior knowledge in sampling and statistics. It teaches the participant how to use the multiple tables and plots of the Standard for designing a sampling plan.
Regulatory and compendial requirements around sampling
Acquaintance with basic sampling distributions
- History of sample taking and sampling
- Sampling plans
- Regulations: US GMPs, EU GMPs, WHO, PIC/S
- Articles sampled in pharma and bio-tech (discrete units vs. granular or liquid materials)
Classification of nonconformities and allocating AQL to classes
- What is Acceptance Sampling:
- Sampling Attributes vs. sampling by variables
- Nonconforming items and nonconformities
- Hypergeometric distribution
- Binomial distribution
- Poisson distribution
- Normal distribution
- Concept of probability of acceptance
Acceptance Sampling Plans in ISO-2859-1/ANSI Z1.4
- Classification schemes
- Classification of nonconforming items (Class A, B, C...)
- Classification of nonconformities (Class A, B, C...)
- Examples of nonconformities in pharmaceutical preparations (Optional)
- Allocating AQL to various classes
Risks in sampling with ISO-2859-1/ANSI Z1.4
- Structure of the Standards
- Single sampling Plan
- Double sampling Plan
- Multiple sampling Plan
- Switching rules between Normal-Tightened-Reduced inspections
- Acceptance sampling of an isolated lot using ISO-2859-2
Sampling for visual inspection of particles in parenteral drugs
- Probability of acceptance
- Producer risk
- Consumer risk
- Operating Curve
- Misconceptions of sampling
- Determining product and consumer risks in a sampling plan
How to effectively reduce the amount of samples to be tested?
- Regulations for sampling for visual inspection
- Types of sampling
- AQL for sampling
- Policy of sampling and inspection
Sampling according to the WHO Guide
- Reduced Testing concepts
- Internal testing vs. external testing
- Using the suppliers CoA instead of in-house testing
- Use of devices to reduce amount of samples (Rapid ID testing, Rapid Mibi Testing)
Sampling of Powders (APIs and Excipients)
- Sampling of starting materials
- Full testing vs. testing for identity
- Qualified supplier vs. unreliable supplier
- n, p and r plans
- Criticism of the sampling plans
Tools for Sampling in a pharmaceutical plant
- Regulatory requirements
- Risk assessment for sampling
- Sampling plans / sampling schemes
- Training for sampling
- Retention / Reference samples
- Starting material Identity testing
- Sampling for the purpose of Assay
- Sampling of raw materials (WHO guide, n/r/p-sampling plans)
Sampling and inspection of packaging materials
- Techniques of drawing samples
- Prerequisites / Requirements for correct sampling
- Sampling devices and –containers
Charting and trending nonconformities and non-conforming items
- Regulations and guidance for packaging and labeling control
- Primary packaging: containers and closures: What is inspected?, AQL for sampling
- Defects in PPMs
- Secondary packaging: labels, leaflets and folded boxes: What is inspected?, AQL for sampling, Sampling in printing house, Sampling in manufacturer’s site, Defects in labels
Good quality practice around sampling plans
- Run chart and control chart
- Charting the number of defectives
- Charting the number of non-conformities
- Detecting a trend in your inspection quality
- Determining your process average
- Does your inspection data confirm your AQL?
- Deriving statistically your allowed percent defectives
√N rule: its uses and misuses
- How to document the sampling system within the company?
- How to incorporate it into specifications (FDP/raw materials)?
- How to incorporate it into the LIMS system?
- What you should discuss with the supplier! (tailgate samples, pre-delivery shipment samples, statements of homogeneity)
- Origin of the rule
- Uses and misuses
- How confident is it?
Step-by-step use of ISO-2859-1/ANSI Z1.4
- Procedure for a Single sampling Plan (Exercises)
- Procedure for a Double sampling Plan (Exercises)
- Procedure for a Multiple sampling Plan (Exercises)
- Exercise in using tables of sampling
Implementing Reduced Testing
- Exercise implementation of reduced testing concepts on real life examples
Relevant tables from the ISO Standard ISO 2859-1:1999
Sampling procedures for inspection by attributes
Part 1: Sampling schemes indexed by acceptance quality limit (AQL) for lot-by-lot inspection –
will be made available to the course participants for the purpose of practicing.