“Data as an asset opens the way for a data-driven future,” proclaims Ulrich Köllisch.

In that future, life sciences manufacturers regularly rely on data-intensive, validated tools: natural language processing (NLP) for categorizing and parsing long lists of deviations, for example, real-time multivariate process control guided toward a “golden batch”; or even more exotic data applications.

Manager and subject matter expert on data integrity at the regulatory compliance consultancy GxP-CC, based in Germany and the United States, Köllisch says developing a “Data Culture” allows life sciences manufacturers to take advantage of these data-driven approaches to reduce process risks, increase control, and serve the goals of business and quality units at the same time.

Köllisch shared his thoughts on reaching the data-driven future in his November webinar for Redica Systems, “Data Integrity and Quality Culture — Enabling Far More Than ‘Just’ Compliance.”

Data Integrity is the Bedrock

Solid data integrity is the foundation of the data-driven future. Redica Systems allows you to dig into the data to find where regulators have already pointed out data-integrity shortcomings in the industry, so that you can be ahead of the curve.

For example, in 483 forms, the FDA issued to manufacturers from 2018–2022. Of the 268 observations and deficiencies involving data integrity, data accuracy is not among the top data-integrity issues.

Instead, data attributability (28%), the existence of original data (20.9%), and issues with system controls (17.5%) lead the pack.

Ulrich webinar blog Data Integrity Issues — Figure 1 | Types of Data Integrity Issues in FDA 483 Forms, 2018–2022, with 268 data-integrity observations and deficiencies

Request a tour of Redica Systems and learn how to strengthen your company’s Inspection Preparation, Vendor Quality, and Regulatory Surveillance.

From Quality Culture to “Data Culture”

As Köllisch says in his presentation, the data-driven future is open to organizations that develop a “Data Culture,” which requires proper data governance applied within a strong quality culture, and where the quality and business units share objectives.

Briefly, a strong quality culture stems from its data governance system and the active engagement of management, according to the PIC/S Guidance on data integrity. This is the environment that protects data integrity.

“I think that’s one key message from the latest publications — in general, quality culture is the responsibility of management,” says Köllisch. As a guide to quality culture, he recommends the 2017 Cultural Excellence Report from the International Society of Pharmaceutical Engineers.

Köllisch’s Data Culture arises in organizations with highly mature quality management that link their business and quality objectives together. They bring together subject-matter experts from quality, data, and other units to understand what data really means, even training staff from quality units in the data technologies they employ.

This is Data Culture, “which really facilitates data as a knowledge deliverer and as an asset,” he says.

Meeting the Data-Driven Future

Using data that meets integrity requirements, organizations can derive information. From there, they create knowledge. And when organizations are constantly learning as a continual process, they build on it to arrive at the right insights for the wisdom of evidence-based decision-making.

Figure 2 data integirty — Figure 2 (Source: researchgate.net) | In the familiar DIKW pyramid, manufacturers with a good Data Culture occupy the top slot. Solid data integrity allows them to take advantage of evidence-based decision-making, even in real time.

It’s this continuous accumulation of knowledge that helps a manufacturer understand the boundaries of its process, reducing risk and leading to a more secure situation.

An organization that succeeds here can implement advanced data-driven tools like the examples above, golden batch-driven multivariate control, NLP, and more.

It doesn’t work with paper systems, outdated systems, and data silos.

“This, of course, does not work if we are doing our three validation runs and never look at the data again,” adds Köllisch. “And the second point where it doesn’t work — and this is also a key message for the talk today — it doesn’t work with paper systems, outdated systems, and data silos.”

Again, there’s much more detail in the webinar, which you can find here.

And be sure to request a tour of Redica Systems if your organization wants to excel in our data-driven future.