Discharge Summary De-Identification with anonym.plus

Strip every patient identifier from the file without sending one byte to the cloud.

Discharge summary de-identification removes direct and indirect identifiers so the file is anonymous under UK GDPR Recital 26 and the ICO Anonymisation Code of Practice. anonym.plus does this on your own device. The text stays readable, but it no longer names the patient.

When this applies

You want to share the file, teach from it, or feed it to an analytics tool. Each use needs the data stripped of names, dates, and record numbers first.

How anonym.plus handles it

  1. Open the file (PDF, DOCX, or scan) in anonym.plus on your device.
  2. Local OCR reads scanned pages, so it catches printed text too.
  3. The tool flags names, dates, NHS numbers, addresses, and phone numbers.
  4. Check the flagged items and fix any clinical term caught by mistake.
  5. Swap each identifier for a safe label, or black it out.
  6. Save the clean file. The original never leaves your machine.

What you need to provide

Patient data entity types detected

Categoryanonym.plus entity typeExample
NamesPERSONOliver Bennett → [PATIENT_1]
DatesDATE_TIMEAdmitted 11/03/2026 → [DATE]
Record IDsMEDICAL_RECORD_NUMBERMRN 884213 → [MRN]
NHS NumberUK_NHSNHS 943 476 5919 → [NHS_NO]
LocationLOCATION14 Elm Road, Leeds → [ADDRESS]
ContactPHONE_NUMBER+44 7700 900123 → [PHONE]

Compliance achieved

Anonymise discharge summaries offline — see plans & start free →

Limitations & cautions

UK GDPR Recital 26 requires that no one can reasonably re-identify the person. The tool removes direct and indirect identifiers. You still judge rare free-text clues (a rare illness plus a small town). For those, apply the ICO motivated-intruder test.

Frequently asked questions

Which identifiers must go for UK GDPR anonymisation?

The ICO Anonymisation Code of Practice says the file is anonymous only when no one can reasonably re-identify it. That means removing names, NHS numbers, dates, addresses, phone numbers, and any other direct or indirect identifiers that could single out the person.

Does this need a data-processor contract with a cloud vendor?

No. anonym.plus runs on your own device with no cloud step. No outside party touches the record, so no data-processor agreement is triggered.

Will the file still read well after the swap?

Yes. The Replace operator puts a steady label in place of each identifier, so the text still flows and no longer points to a real person.