Copy of Pharmaceutical Analysis I Lecture_3_Quality Control of Analytical Methods

Quality Control of Analytical Methods Lecture 3

Table of Contents

    1. Sampling

    1. Material Control

    1. Accuracy and Precision

    1. Statistical Quality Control

1 - Sampling

Sampling – Introduction

  • A batch can be accepted by a series of steps determining conformity to specifications.

  • Securing a representative sample is crucial yet often ignored by analysts.

  • Fixed rules in sampling are not possible; they depend on material nature and quantity.

Sampling – Advantages of Statistical Sampling (vs 100% testing)

  • Lower Cost: Economically feasible compared to full testing.

  • Faster Data Collection: Quicker results through sampling.

  • Homogeneity: Smaller data sets enhance accuracy and data quality.

  • Representation: Samples accurately represent the entire population.

Sampling – Definition

  • Sampling: The process of removing items from a population to analyze.

  • A sample is a finite selection from the population.

  • Important types: RMQC (Raw Material Quality Control), PMQC (Packaging Material Quality Control), IPQC (In-Process Quality Control), FPQC (Finished Product Quality Control).

Sampling – Sample Types

  • Random Sample: Equal probability for different material fractions.

  • Representative Sample: Sampling ensuring proportional representation of batch parts or properties.

Sampling – Lot/Batch Numbering Systems (Examples)

  • Year Code:

    • T = 2005

    • U = 2006

    • V = 2007

    • W = 2008

    • X = 2009

    • Y = 2010

  • Month Code:

    • A = January

    • B = February

    • C = March

    • D = April

    • E = May

    • F = June

    • G = July

    • H = August

    • I = September

    • J = October

    • K = November

    • L = December

  • Day, Location Codes: Specific numbers indicate date/locations respectively.

Sampling – Inspection Methods

  • Single Sampling: Decision after one sample.

  • Double Sampling: Results obtained after second sampling,

  • Triple Sampling: Involves multiple samplings.

Sampling – Sampling Plan

  • Definition: Rules for sample size, frequency, and acceptance basis.

  • Requires three specifications:

    • N: Batch size.

    • n: Sample size drawn.

    • c: Maximum defectives allowed.

Sampling – Sampling Plan Example

  • Example: Random sample of 5 from a lot of 50:

    • If sample exceeds 0 defects, reject the lot. Else, accept it.

Sampling – Risk Consideration

  • Understand the laws of probability:

    • Producer’s Risk (α): Risk of rejecting a good batch.

    • Consumer’s Risk (β): Risk of accepting a bad batch.

Sampling – Types of Sampling Plans

  1. Square Root System: Takes the square root of batch size for sampling.

  2. Government Sampling Plan: Developed by the US Department of Defense and uses AQL (Acceptable Quality Level).

Sampling – Example Scenarios

  • Example 1: 28 drums of muriatic acid, amount to sample using Square Root System.

  • Example 2: Different batch numbers in drums and corresponding sample sizes.

2 - Material Control

Material Control Overview

  • Each incoming batch of material receives a receiving number and its control report.

Material Control – Raw Materials Handling

  • Reception: Visual damage inspection; samples taken per sampling plan.

  • Quarantine:

    • Materials placed on hold with a “HOLD” sticker until QC review.

    • Samples sent for lab testing (identity, potency).

    • Decision labels placed according to QC results; only one sticker disposition allowed.

Material Control – Color-Coded Stickers

  • Stickers: Different colors indicate:

    • Yellow: Quarantine

    • Green: Approved

    • Red: Rejected

Material Control – Printed and Packaging Materials

  • Primary Packaging Components: Direct contact with products.

  • Secondary Packaging: Indirect contact; accessories to primary.

Material Control – Reassay Dates

  • Importance of quality monitoring during material storage.

  • Reassay Date Schedule:

    • Monthly or prior for unstable materials.

    • 6 months for vitamins, flavors.

    • 12 months for dyes, active ingredients.

    • 24 months for excipients.

Material Control – Retesting Requirements

  • Parameters include appearance, pH, identity, moisture, microbial range, etc.

3 - Accuracy and Precision

Definitions

  • Accuracy: Agreement of results with true value expressed in error terms.

  • Precision: Reproducibility within a series of results.

Accuracy Measurements

  • Mean (X̄): Average of results.

  • Absolute Error: Difference between mean and true value.

  • Relative Error: Absolute error as a %.

Precision Measurements

  • Indicates closeness of results, expressed via average deviation, standard deviation, or range.

  • Results can be precise yet inaccurate due to determined error.

Statistical Measures

  • Average Deviation (d): Average of differences from the mean.

  • Standard Deviation (σ): Shows data variation from mean; small values suggest tight clustering around mean.

  • Relative Standard Deviation: σ as a % of the mean.

Range and Outliers

  • Range: Difference between the highest and lowest measurement.

  • Outlier: A data point significantly distant from others.

Example Problems

  • Calculations for various scenarios involving accuracy and precision.

4 - Statistical Quality Control

Overview

  • SQC: Statistical methods for monitoring quality during production stages.

Statistical Process Control (SPC)

  • Establish standards, measure, take corrective actions during production.

  • Periodic sampling and control chart plotting are critical.

Control Chart Types

  • Attribute Chart: Discrete data classification.

  • Variable Chart: Continuous measurement records.

Control Chart Components

  • UCL (Upper Control Limit) and LCL (Lower Control Limit) determined for control.

  • Key statistical formulas provided.

General Control Steps

  1. Select a random sample of size n.

  2. Calculate averages and variances.

  3. Plot statistics on a control chart.

  4. Identify control status based on plotted points.

Note on Control Charts

  • If values fall outside established limits, the process is considered out of control, requiring investigation.