Deviation log de-identification is the removal of subject and staff identifiers from the issue list. It supports ICH E6(R2) good clinical practice. anonym.plus runs offline and keeps each entry description readable.
When this applies
A quality lead shares the issue list with an external reviewer. Each row names the person affected and the staff member who logged it.
How anonym.plus handles it
- Open the file (XLSX, CSV, or DOCX) in anonym.plus on your device.
- The tool scans rows for subject codes, staff names, and dates.
- Local OCR reads a scanned paper page if you add one.
- Confirm the flagged people and the affected codes.
- Swap each for a neutral label, or black it out.
- Save the clean list locally with no upload.
What you need to provide
- The file (XLSX, CSV, DOCX, or scan).
- An operator: Replace for labels, Redact for whole columns.
- Optional: a label map for repeated staff names.
PHI entity types detected
| Category | anonym.plus entity type | Example |
|---|---|---|
| Subject code | ID | subject 09-221 → [SUBJECT] |
| Staff | PERSON | logged by P. Andersson → [STAFF] |
| Dates | DATE_TIME | occurred 22/05/2026 → [DATE] |
| Site | LOCATION | Site 09, Malmö → [SITE] |
| Investigator | PERSON | PI Dr. Lind → [INVESTIGATOR] |
| Contact | EMAIL_ADDRESS | qa@example.se → [EMAIL] |
Compliance achieved
- Helps meet ICH E6(R2) GCP quality duties.
- Stays offline, so the tool itself needs no BAA.
- On-device AES-256-GCM guards the working file.
- Supports GDPR Art. 9 for EU subject and staff data.
Anonymize protocol deviation logs offline — see plans & start free →
Limitations & cautions
A description can name a subject inside the free-text cell, not just the code column. The tool scans both. Read the narrative cells closely, since a detailed account can still point to one person at a small site.
Frequently asked questions
What is a protocol deviation log?
It is the running list of times the trial strayed from its plan, with the subject affected and the staff involved. Reviewers need to see the pattern, not the names, so a clean copy is shared.
Where do names hide?
In the free-text description cells, not only the code column. The tool scans the prose in each row and flags any people it finds.
Do the entries stay readable?
Yes. The nature of each issue, its cause, and its fix stay. Only subject codes, staff names, and the site are removed.