Insights from Peter Baker at the 46th International GMP Conference

An investigation into data integrity (DI) issues is “almost inevitable” for every company in the industry within about two years, says Peter Baker, former FDA investigator and president of Live Oak Quality Assurance.

Speaking at the 46th International GMP Conference in March, Baker painstakingly detailed how he has seen DI issues get mishandled—especially, systemic, organizational snafus—and how companies should actually handle them.

[Related: For more on DI issues specific to analytical instruments, click here to access the recording of the Redica Systems webinar, “Are Laboratories Perpetuating Data Integrity Problems?”]

This article is part II of a two-part series covering insights from Peter Baker’s presentation, “Handling the Inevitable: Investigations Where Data Integrity Has Been Implicated.” This second half presents Baker’s insights on managing DI issues with quality risk management (QRM). Please see Part I of this series for Baker’s thoughts on how DI issues can snowball—especially the systemic dynamics that contribute.

Those systemic dynamics can interfere with DI-related investigations in different ways. But, in general, companies get in trouble when they determine the scope, the root cause, and the prime impact “without being transparent about the uncertainties and unknowns that come with data integrity issues,” Baker says. “That is really the key.”

Handling Data Integrity Issues Properly with QRM

First, what is Quality Risk Management? FDA defines QRM as “a systemic process for the assessment, control, communication, and review of risks to the quality of the drug product across the product lifecycle”. FDA has a guidance document on the topic here.

With QRM defined, how should companies show they are using an effective investigation strategy? Baker recommends reviewing the FDA guidance, Data Integrity and Compliance With Drug CGMP. That is where you will find the keywords “risk assessment,” says Baker.

“That is what the regulators are really looking for,” he says. “In addition to these other things—the traditional root cause, scope, etc.—what about the scientifically sound risk assessment?”

With data integrity, the game has changed

When they are using QRM, manufacturers can make certain assumptions about the scope and root cause. But only if they are transparent about the uncertainties and unknowns that DI issues involve, Baker says.

Routine deviations do not have the uncertainty and assumptions of DI issues, so they are easier to investigate. “But with data integrity, the game has changed,” Baker says.

Critical Thinking in Quality Risk Management

For example, consider a case where DI problems land manufacturing outside the validated state. In Baker’s example, when calibration records are completed but certain calibration activities were not performed according to procedure.

Typically, says Baker, the company will say that final QC testing shows that historical results have met specifications.

Using critical thinking, “the first question the investigator is going to ask is, ‘is QC testing designed to catch issues with the calibration program?’” he says. “Would you see testing or even catch it?”

Sites do not have to test every tablet, he says. And not every bit of uncertainty requires them to throw out a batch. But to justify any product release or distribution, it takes critical thinking in quality risk assessment management.

If the issue turns out to have critical major implications for direct or indirect patient product safety, the next questions are clear. “Has the firm addressed the lack of quality assurance? Have they addressed that we may have an issue here as far as acceptability of that product?” says Baker.

And, finally, at the end of an investigation, the scientific justification must not include any opinion or other bias. “It can include a risk evaluation where we are transparent with unknowns and uncertainties and assumptions, that is okay,” he says. Bias undermines the conclusion.

DI issues are inevitable, Baker repeats throughout his talk. Accepting this truth is a big step toward handling them the way regulators want it done.

The Latest Data Integrity Trends

Are you prepared for your next FDA inspection? Below are the top data integrity-related 483 observation trends so far for 2022 for human drug GMP sites worldwide.