Medication-history de-identification is the removal of patient IDs from a drug list. A drug history is health data, so it is special category under UK GDPR Art. 9(1), and research reuse leans on Art. 9(2)(j) with the Art. 89(1) safeguards. anonym.plus does this locally. The regimen stays readable; the file no longer names the person.
When this applies
A drug list ties many medicines to one named patient over time. For a study or a teaching set, you strip the IDs but keep the regimen and the gaps between start and stop dates, since those gaps carry the clinical meaning.
How anonym.plus handles it
- Open the list in anonym.plus on your device.
- It scans for the name, dates, NHS number, and contacts.
- Local OCR reads scanned pages, so printed lists are caught.
- Confirm the flags; the medicine names are not IDs and stay.
- Swap each ID for a label, or shift the dates.
- Save the clean file; the source stays on your machine.
What you need to provide
- The list (PDF, DOCX, TXT, or EPR export).
- An operator (Replace keeps the list readable).
- Optional date-shift to keep the start-and-stop gaps.
Patient data entity types detected
| Category | anonym.plus entity type | Example |
|---|---|---|
| Names | PERSON | Callum Fraser → [PATIENT] |
| Dates | DATE_TIME | started 03/2024 → [DATE] |
| Record IDs | MEDICAL_RECORD_NUMBER | MRN 220914 → [MRN] |
| Contact | EMAIL_ADDRESS | c.fraser@example.com → [EMAIL] |
| Location | LOCATION | 12 Birch Lane → [ADDRESS] |
| Identifiers | NHS_NUMBER | 485 777 3456 → [NHS_NUMBER] |
Compliance achieved
- Research reuse rests on UK GDPR Art. 9(2)(j) and the Art. 89(1) safeguards, with the condition in DPA 2018 Schedule 1, Part 1, paragraph 4.
- Where the output is genuinely anonymous, UK GDPR Recital 26 takes it out of scope of the Regulation altogether.
- Anonymity is judged with the motivated-intruder test in the ICO anonymisation guidance, not simply by the absence of a name.
- Date-shift preserves the treatment intervals, so minimisation under UK GDPR Art. 5(1)(c) costs you no analytical value.
Anonymise medication histories offline — see plans & start free →
Limitations & cautions
UK GDPR Recital 26 and the ICO anonymisation guidance require that re-identification is not reasonably likely, judged against a motivated intruder. The tool removes the direct IDs; the medicines stay. A rare regimen plus a small area can still hint at one person, so record your residual-risk assessment before you release the file.
Frequently asked questions
What identifiers are removed?
Names, NHS numbers, addresses, dates tied to the person, phone, email, MRN, and any other code that could identify the patient under the ICO motivated-intruder test.
Are the medicines treated as identifiers?
No. Drug names are clinical content, not personal data in themselves, so they stay. The list is still useful for research after the IDs go.
Can the date gaps survive the swap?
Yes. Date-shift moves all dates by one offset, so the start-and-stop gaps stay intact while the true calendar dates are hidden.