pseudonimized

Pseudonimized (or pseudonymized) data refers to information in which direct identifiers have been replaced with pseudonyms or tokens, making the data less immediately identifiable while preserving the ability to re-link to individuals with additional information held separately by a trusted party. The process aims to reduce privacy risk during processing while maintaining data usefulness for analytics and research.

Pseudonymization is reversible and distinct from anonymization, which removes identifiability to the point where re-identification is not reasonably possible. In pseudonymization, re-linking to the original identity typically requires access to a separate key or dataset. Techniques include tokenization, reversible encryption, salted or keyed hashing, and other data masking methods. The security of pseudonymized data depends on protecting the linkage key, separation of duties, and robust access controls.

Applications include medical research, customer analytics, and administrative data where individuals' identities are not needed for analysis but may be re-identified for authorized purposes such as clinical follow-up or regulatory reporting. Benefits include reduced privacy risk, compliance with data protection principles like data minimization, and enabling dataset linking across systems without exposing direct identifiers.

Regulatory context: under data protection laws such as the GDPR, pseudonymization is recognized as a security measure that can reduce risk, but pseudonymized data remains personal data if the linkage to individuals is still possible. Organizations should implement governance, risk assessment, and clear policies on when re-identification is permitted, monitoring, and data subject rights.