Since the FDA’s Data Integrity guidance came out in April 2016, there has been a lot of interest on the topic. Indeed, the number of warning letters for drug GMPs increased by 12% in FY2017 as compared to the previous year. 65% of these warning letters cited data integrity violations.

In a world awash with data, where does one start to improve data integrity? From an IT perspective, data is generally classified as:

• Structured data, that normally resides in a relational database such as SQL or Oracle. Enterprise applications such as ERP, MES and LIMS typically store their data in such relational databases and are examples of structured data.
• Unstructured data, that refers to files residing on users’ local machines or network drives. Common examples of unstructured data include spreadsheets, Word/PDF documents, drawings, instrument lab or other raw data files in CSV or TXT format.

Data integrity assurance requires a number of components that include, but are not limited to, security, audit trails and electronic signatures. Applying a risk-based approach to data would lead to the following general conclusions: Continue reading “21 CFR Part 11 Compliance for Spreadsheets”