The guidance describes data integrity as the degree to which data are complete, consistent, accurate, trustworthy and reliable throughout the data lifecycle. It does not prescribe technology. Instead it asks organisations to understand their data, assess its criticality and risk, and put proportionate controls in place — organisational as well as technical. The arrangements that make this work are called data governance.
Key expectations
- Data governance — the sum of arrangements, including culture, that ensure data are recorded, processed, retained and used to give a complete, consistent and accurate record throughout the data lifecycle.
- Data criticality and risk — controls proportionate to how much the data matters to product quality and patient safety, and how vulnerable it is to unauthorised change.
- ALCOA+ — data that are attributable, legible, contemporaneous, original and accurate, and also complete, consistent, enduring and available.
- Original records and true copies — the first capture of data, or a verified copy that keeps its content, meaning and metadata.
- Audit trails and their review — audit trails for GxP-relevant changes, reviewed as part of routine data review, with a risk-based approach to frequency.
- Access and system administration — individual user accounts, access appropriate to the role, and system administrator rights restricted and, where possible, independent of the people who generate the data.
- Data review and approval — a documented process for reviewing data, including relevant metadata, before it is used to make decisions.
- Hybrid systems, data transfer and archiving — controls where paper and electronic records combine, where data moves between systems, and where it is retained for the long term.
What an MHRA inspector looks for
In practice, UK GMP inspectors follow the data. They pick a result or a batch, ask to see the raw data and its metadata, check who did what and when, look at audit trails around key decisions, test whether shared logins or uncontrolled spreadsheets exist, and ask how your QA team reviews audit trails. Data integrity observations often arise when a system allows changes without a reason, lets the person who produced a result approve it, or keeps the original data somewhere nobody reviews.

