.jpg){kind=link}
- Published
- Updated
The same person may appear under different codes in safety and clinical data sources. Incorrect matching can make an event appear missing or attach it to someone else. Reconciliation should first explain which identity fields support reliable matching and how uncertain cases are resolved.
Scope of the decision
Consider participant, site and study identifiers together without circulating direct identifiers unnecessarily. A formatting correction is different from a genuine identity change. Where matching remains uncertain, use a controlled review route instead of automatically selecting the most similar candidate.
Worked review
For example, a safety record uses a site code with a leading zero while a clinical file omits it. The system flags what appears to be a separate person. The team checks the authorised mapping, confirms the identity and retains the reason for correction. A similar event name alone would not establish identity.
Practical checks
- Define matching fields and permitted transformations of their format.
- Refer uncertain candidates for authorised review rather than merging them automatically.
- Check how an identity correction affects links to other events.
Acceptance and limits
Acceptance requires adequate evidence that an event belongs to the correct participant and study. Resolving identity does not demonstrate that event content also agrees. Clinical assessment, follow-up information and reporting status still need consideration within their authorised processes.
Frequently asked questions
- Is the participant’s name necessary?
- Not always. An appropriate authorised code mapping may be sufficient; avoid unnecessary personal-data sharing.
- Does standardising formats resolve every difference?
- No. Similar codes can belong to different people, so verify both context and source.