Pathology Narrative De-Identification with anonym.plus

Clear identifiers from the report while specimen and diagnosis detail stay.

In simple terms, PII redaction is the on-device process of finding and masking personally identifiable information in a document before it is shared.

Pathology-narrative de-identification removes patient and accession identifiers from the free-text part of a histopathology report. The reporting items themselves — specimen, block, grade, stage — follow the Royal College of Pathologists cancer datasets and are what makes the report worth reusing, so they stay. Where the report describes DNA analysis, the Human Tissue Act 2004 also applies, and s.45 of that Act makes non-consensual DNA analysis a criminal offence. anonym.plus runs locally and keeps the gross, microscopic, and diagnosis text whole.

When this applies

These reports feed cancer registration, which uses confidential patient information without individual consent under NHS Act 2006 s.251 and the Health Service (Control of Patient Information) Regulations 2002. For a teaching set or a local audit you have no such cover, so strip the identifiers first.

How anonym.plus handles it

  1. Load the report into anonym.plus on your device.
  2. It finds the patient name, the accession or case number, and the requester.
  3. The gross, microscopic, and diagnosis text stays in place.
  4. Swap identifiers with steady labels across linked specimens.
  5. Save the clean report on your device.

What you need to provide

Patient data entity types detected

Categoryanonym.plus entity typeExample
NamesPERSONG. Fischer → [PATIENT]
Record IDsMEDICAL_RECORD_NUMBERS26-4471 → [ACCESSION]
NamesPERSONDr Sato → [PROVIDER]
DatesDATE_TIME02 Mar → [DATE]
LocationLOCATIONPath Lab, Newcastle → [LAB]
ContactPHONE_NUMBER0191 496 0123 → [PHONE]

Compliance achieved

Anonymise pathology narratives offline — see plans & start free →

Limitations & cautions

A rare tumour type, a collection date, and a small catchment area can re-identify someone even after the direct identifiers are gone. For uncommon cases, broaden dates and apply the ICO motivated-intruder test rather than relying on field-level de-identification alone. Genetic findings deserve particular care: they also identify relatives.

Frequently asked questions

Does the swap touch the diagnosis text?

No. The gross description, the microscopic findings, and the final diagnosis all stay. Only identifiers such as the patient name and the accession number change, so the report still carries every reporting item the Royal College of Pathologists datasets ask for.

Is a genetic result treated differently?

Yes, it deserves more care. UK GDPR Art. 4(13) defines genetic data as personal data on inherited or acquired genetic characteristics that gives unique information about a person, and Art. 9(1) makes it special category. It also says something about blood relatives, which conventional identifier removal does not address.

Is the LIS export supported as well as the PDF?

Yes. The narrative PDF or DOCX and the structured laboratory information system export both work, and an accession map keeps every part of one case under a single label across both formats.