FR

By Jean-Pierre Amiel, ASQ Senior, CQA ret., Secretary, Web committee and Audit Chairs.

In this presentation on September 28th, 2022, Kia Kahhali, B. eng. and founder of APS Compliance Consultants Inc (also known as "Access Pharma Solutions"), a private firm of experts assisting the Pharmaceutical and Food industries gave us insight and a plethora of information on quality, compliance, validation, and regulatory are as affecting the quality and integrity of the data in pharmaceutical, biopharmaceutical and allied industries.

Well once again, a look into the ever evolving world of data documentation of processes, but for me, a real update on the thoughts that go into the preparation and the life cycle of DATA. Let's be truthful, it was actually "overwhelming" as our presenter tried to squeeze so much information into his 90 minutes presentation that it was like... data overload 😁, pun intended.

Kia started off ensuring that the 33 participants present understood that there were two basic definitions of what data was and that the Data Integrity Scope is much wider than just the analytical testing results as it includes sampling, inspection and testing activities and any deviation and investigation through the process.

One of the most interesting aspects of his presentation was his comment that he was using standards written over 20 years ago and which have had few recent updates — not because there is nothing new in data principles — but rather because the principles remain "d'actualité", they are still current. He also stated that data integrity controls should be based on the risks of losing that data. But the most challenging, and I have faced that one, is that the system used MUST allow the future reader/regulator/auditor to access and read the ORIGINAL data. Now there is a problem... as he suggested that only the original data stream be stored, not the interpreted or summarized versions, so that it may be reproduced. Is this a requirement for particular domains? This also infers that the 'raw' data must survive whatever technological evolution happens. And of course, this means that the methods used for validation and data transfer are also part of laboratory records and have to be included within the scope of the data integrity system.

• Raw data definition: The original (data) which can be described as the first-capture of information including all contextual metadata from sampling to reporting, whether recorded on paper or electronically, permitting full reconstruction of the activities and evaluating reports. Information that originally captured in a dynamic state should remain available in that state.
• Interpretation: Whether this has been captured in a dynamic state and generated electronically, paper copies cannot be considered as ‘raw data’.

As for the system requirements, there was the usual management support and involvement but also the need for employee training and the recording of that commitment, as the objective is to "ensure an engagement of the whole organization from the board room down to the laboratory bench and production floor". As an aside to this item, ASQ has been requiring for the last four years that all your Section Leadership Team follow and pass an on-line course on the General Data Protection Regulation (GDPR), a European privacy and security system, which imposes obligations on organizations to protect various members' personal information — a commitment.

Kia explained that for the data integrity approach, regulatory agencies have opted to issue Guidance documents rather than modifying the regulations so as to offer clarifications. He presented a Data Integrity Model for a pharmaceutical quality system that has three levels and explained that a Computerized Maintenance Management System, such as his company's Alleye software, can best address the Level 1 - Equipments and systems and is working on another system for Level 2 - Analytical Methods and procedures. In his PDF handout, Kia provided a reference library of the various documents quoted.

Do you remember paper and pen? That was relatively easy to store, but it falls into the category of interpreting information. What about magnetic tape cassettes and reels, 8 inch, 5 1/4 inch and 3 1/2 inch disquettes, the computers and software to operate and drive these? I would have liked to ask these clarifications, but there was so much more about the strategies and structures of the data integrity systems that we just barely made it in the allocated time. This may mean that another session on this subject is to be considered.

You definitely missed a good overview of this popular subject – an 87% vote by the participants. Well done! Interesting again! Wonderful evening.

Thank you all for participating!