An Example of Absence of a Data Culture

"Compliance Tools" discusses the various techniques and methods used by quality and compliance professionals in their daily work responsibilities. We intend this column to be a resource for daily work applications.

Reader comments, questions, and suggestions for discussion topics are needed to help us fulfill the column objective. Please send your comments and suggestions to column coordinator Lynn Torbeck at lynn@torbeck.org or to journal managing editor Susan Haigney at shaigney@advanstar.com.

KEY POINTS

The following key points are discussed:

* This discussion describes an actual event involving rejects in a tablet compressing process.

* Statistical evaluation of all data (passing and failing) provided a much different conclusion compared to the original evaluation.

* Data collection without understanding and integration of all available data will not yield correct decisions. Process variation must be understood.

* Statistical tools including histograms and control charting are invaluable in process understanding and problem-solving.

* Organizations must develop a data culture, understand the theory of variability, be able to do basic data analysis, and draw scientifically based conclusions.

* Compliance professionals must work with site management to accomplish this end for their organization and for the benefit of the patients who use our products.

INTRODUCTION

This article presents a case study of how statistical evaluation and the establishment of a data culture create scientifically based solutions to quality assurance (QA) problems in manufacturing.

A QA CASE STUDY

The following case study is an actual occurrence. It can be told because the company is gone, and the product has been removed from the commercial market for many …