Are you confusing Part 11 with data integrity?
In recent years, pharmaceutical companies have been making a great deal of effort to comply with data integrity requirements. However, many companies seem to be confusing data integrity compliance with Part 11 compliance in their case studies. And the same is true of books and seminars on data integrity. In most cases, people think that data integrity compliance is just a matter of adding security and audit trail functions to computerized systems such as analytical instruments.
This is a huge mistake. Up to 80% of events that threaten data integrity (i.e., reliability of records) are human error.
Here is a question for the reader.
Which is more critical to patient safety, knowingly falsified data or data falsified by human error?
The answer is the same for both.
The first and foremost step in addressing data integrity is to reduce the occurrence of human error. Human errors include the following
・typographical error
・miscalculation
・typographical error
・Reagent errors (type, expiration date, adjustment, storage conditions, etc.)
・Procedural errors (e.g., misunderstandings, assumptions)
Even if security and audit trail functions are added to analytical instruments and other computerized systems, human error cannot be prevented. Human error always occurs at a certain rate. There is no such thing as a person who never makes a mistake. To prevent human error, a system (QMS) such as double checking is necessary.
Data integrity applies equally to paper and electronic records
In addition, data integrity compliance should not be limited to electronic records. Here is the question again for the reader. Which is more critical to patient safety, tampering with electronic records or paper records?
The answer is the same for both.
In many manufacturing plants, manufacturing records and other records are still created using paper media. Efforts must also be made to address the data integrity of paper-based records.
Human errors prevented by IT
Of course, there are some events where human error can be prevented by digitizing records. For example, stand-alone analytical instruments such as electronic balances can be connected to the LIMS through a network. This will ensure that analytical data such as weighing values are automatically transferred, preventing transcription and input errors. A validated system would also prevent calculation errors. In addition, data could be protected against overwriting, modification, or deletion. But this assumes that the analysis method, analysis program, parameters, etc. are appropriate and that the raw data collected are correct. In the first place, if there are human errors in the process of generating the raw data, digitization cannot prevent them.
Part 11 compliance is limited data integrity compliance.
The purpose of Part 11 is whether accidental or intentional,
- Ensure that electronic records cannot be easily tampered with (security)
- Know if it has been tampered with (audit trail)
The first is.
In Part 11, the human error measures described above are not in scope.Also, Part 11 is limited to electronic records. In other words, Part 11 compliance and data integrity compliance are different.
Comment