ADR anonymization is the removal of patient and reporter IDs from a reaction write-up. It supports ICH E2B(R3). anonym.plus runs on your device and keeps the suspected medicine and the reaction so the signal stays whole.
When this applies
A reaction write-up holds the person, the one who reported it, and contact lines. To pool it for signal work, you clear the IDs but keep the medicine and event.
How anonym.plus handles it
- Load the write-up into anonym.plus on your device.
- It finds the named people in the free text.
- Dates, ages, and contacts get flagged across the entry.
- Confirm the flags; the suspected medicine stays as a non-ID.
- Swap the IDs with steady labels for pooled work.
- Save the clean file on your machine with no upload.
What you need to provide
- The write-up (DOCX, PDF, TXT, or export).
- Replace with a steady label map for pooled signals.
- Optional age-banding for the patient age field.
PHI entity types detected
| Category | anonym.plus entity type | Example |
|---|---|---|
| Names | PERSON | Mark Vella → [PATIENT] |
| Names | PERSON | nurse reporter → [REPORTER] |
| Age | AGE | age 73 → [AGE_BAND] |
| Dates | DATE_TIME | reaction 11 Apr → [DATE] |
| Contact | PHONE_NUMBER | 030 555 7781 → [PHONE] |
| Identifiers | ID | ICSR 4471 → [ID] |
Compliance achieved
- Supports ICH E2B(R3) safety case reporting.
- Keeps the suspected medicine, since it is not patient PHI.
- On-device AES-256-GCM guards the working files.
Anonymize ADR reports offline — see plans & start free →
Limitations & cautions
Exact age can identify in older people, so band it where your study allows. The suspected medicine stays, as it holds the signal. A rare event plus a small site can still narrow identity even after direct IDs go.
Frequently asked questions
Why keep the suspected medicine name?
The product is not personal PHI and is the heart of the signal. Removing it would gut the report. Only the IDs of the named people are taken out.
How is the subject's age handled?
Age can be a clue, above all over 89. The tool flags it as AGE so you can band it, for example into a decade, where your analysis allows.
Can reports be pooled afterwards?
Yes. A steady label map lets one person map to one alias across many files.