Invest in Tech to Avoid Data Integrity Issues

Companies that do not invest in and continually improve their laboratory instruments risk regulatory enforcement actions involving data integrity, according to two industry experts. 

Pharma consultant Bob McDowall and Paul Smith, Global Compliance Marketing Specialist at Agilent Technologies, presented their findings during the May 24 webinar, “Are Laboratories Perpetuating Data Integrity Problems?” Part I examines their analysis of regulatory trends involving laboratory equipment, including recent data integrity guidances.

[Related: To view this webinar in its entirety, including the slide presentation, click here.]

McDowall reviewed three FDA warning letters and showed how they featured similar citations involving laboratory and computer controls—a data integrity issue. First, he showed how two warning letters issued in July 2020, one to Stason Pharma and the other to Tender Corporation, cited similar 21 CFR 211 regulations.

“Both U.S companies, both cited under different parts of the CFR (Figure 1). And you think, well, so what? But the remediation required was almost word for word, identical in the two warning letters. One of the questions I would ask, is FDA losing patience with companies failing to get current?”

He then showcased the mandated remediation for both companies (Figure 2).

“One of the things I want to focus on is down in the bottom left-hand corner, interfacing standalone instruments to the LIMS network.” he said. “The underlying message is ‘get rid of paper.’”

To further emphasize this message, McDowall used the example of a third warning letter: this one issued to BBC Group. This one included the finding that “focus on paper and electronic records are not available.” Specifically, the warning letter stated, “You didn’t retain all original, dynamic records, obtained during the course of testing.” In particular, two specific instruments had the capability to save data but the company did not do it.

“This is one of the things that now you start to see a tightening of the letter, of the way the companies are being inspected,” he said. 

The key to preventing negative enforcement actions involving laboratory equipment data requires continual improvement.

[Related: In the past five years, there were 1,822 FDA 483 observations with primary issues involving the laboratory for global human drug GMP firms. Contact us to see how you can access this data.]

Continually Improving for Data Integrity Success

McDowall emphasized the U.S. and EU regulations (including U.S. regulations going back to 1978) require manufacturers to stay current (the “C” in “CGMP.”).

“What you now have to do is balance the cost of compliance versus the cost of noncompliance (Figure 3),” he explained.

“The cost of compliance is the cost of doing it right first time,” he said.  “That is shown in green. If you manage risk correctly, you can decide where you want to be on that line, and you can defend it. On the right-hand side is the cost of non-compliance. This is the cost of getting caught if you have done it wrong or done a shoddy job. The cost of compliance is usually a linear scale, the cost of noncompliance is usually a logarithmic scale. And if you doubt that, Schering-Plough was fined $500,000,000 2002 through consent decree.”

He cautioned that regulatory interpretations evolve over time, pointing out that 21 CFR 211 does not specifically mention audit trails. But since 2005, global regulatory inspectors expect to be able to review audit trail entries. 

“What you thought was defendable a few years ago, is now slipping down the green line to the left and the increased cost of noncompliance is rising,” McDowall said. “What industry normally does, is it takes into account, what was good enough last inspection is good enough this inspection. That no longer works.”

What Happens When You Do Not Keep Current

McDowall then provided a series of examples of practices to avoid when it comes to data integrity in the lab.

Heavy reliance on spreadsheets. In May of this year, FDA issued a warning letter that cited violations involving a non-validated Excel spreadsheet used for lot assay calculations. The company was unable to provide a copy of the original data. “Please, from an auditing perspective, keep using spreadsheets, please. They keep me in business,” he said in response. 

Using an automated system manually. McDowall used an example from one of his auditing clients. This firm used a fully automated dissolution system but only used the fully automatic functionality for one of 48 procedures. The remaining 47 procedures were performed manually on the system, along with unvalidated spreadsheets for calculating dissolution percentage.   

The world’s most expensive electronic ruler. This is how he characterizes one chromatography data system he saw at a company. This system included, free of charge, capabilities for doing calculations of suitability test samples. Instead of using this capability, personnel printed out paper records and typed the results into a spreadsheet. “This is where you find that the review takes longer than the actual analysis.” 

How static are static data? He explained that while both the PIC/S and FDA guides allow for “static data” from a printout produced with data from a balance or simple piece of equipment, once that data is entered into a LIMS or chromatography data system it becomes dynamic data.

McDowall then recommended the following for avoiding negative enforcement actions due to laboratory instrument data:

• Do not buy rubbish software. “Look at both the instrument and the software, but do not just look at it from the ease of use in the lab and how shiny the instrument is. Spend time evaluating the technical controls to ensure compliance.”
• Plan the equipment purchase properly. “Otherwise, you are going to perpetuate your data integrity problems.”
• Get rid of paper and move to electronic. “You have got to take risk-based approaches to validation.”

Smith then provided concluding thoughts about laboratory equipment and data integrity.

“You need to work with your suppliers. And if you find a problem with your suppliers, software or instrument, you need to tell them about it and they need to fix it. And that interaction tells you a lot about the suppliers you choose.”

Additional Data Integrity Resources

Data Integrity and Your Analytical Instruments

Data Integrity and Your Clinical Investigator: What the Data Shows

Can Quality Culture Prevent GMP Issues Such as Data Integrity?

FDA Regulators Address Data Integrity and Lab Audit Trails

What Can Regulatory Data Tell Us About Data Integrity Trends?

Synergy at the Intersection of Data Integrity and Quality Culture

[Related: To view the full recording of this webinar, including the slide presentation and following Q&A, click here.]