Observation-chart de-identification removes patient identifiers from a vital signs record while every measurement survives. Those measurements are what the National Early Warning Score (NEWS2), published by the Royal College of Physicians in 2017, aggregates into a deterioration score, and what NICE clinical guideline CG50 expects staff to act on. anonym.plus works locally and keeps the temperature, pulse, blood pressure, and oxygen series whole.
When this applies
Observation charts feed early-warning-score audit and teaching. The numbers are the value; the name, the NHS number, and the timestamps are the risk. Escalating from a rising score is part of safe care and treatment under regulation 12 of the HSCA 2008 (Regulated Activities) Regulations 2014, so the audit trail matters.
How anonym.plus handles it
- Load the chart into anonym.plus on your device.
- It finds header identifiers: name, NHS number, and dates.
- The reading grid and trend lines stay in place.
- Swap identifiers and shift the times to keep the intervals.
- Save the clean copy on your device.
What you need to provide
- The observation chart (PDF, image, or EPR export).
- A date-shift option to keep reading intervals.
- An operator (Replace works well).
Patient data entity types detected
| Category | anonym.plus entity type | Example |
|---|---|---|
| Names | PERSON | header name → [PATIENT] |
| Record IDs | MEDICAL_RECORD_NUMBER | MRN → [MRN] |
| Dates | DATE_TIME | observation times → shifted [TIME] |
| Location | LOCATION | Ward/Bed → [LOCATION] |
| Names | PERSON | recorded-by → [STAFF] |
| NHS Number | UK_NHS | NHS 943 476 5919 → [NHS_NO] |
Compliance achieved
- Keeps the full observation series that NEWS2 (Royal College of Physicians, 2017) scores — every reading survives.
- Deterioration audits against NICE clinical guideline CG50 still run on the clean chart.
- Escalation on a rising score is part of safe care and treatment under reg. 12 of the HSCA 2008 (Regulated Activities) Regulations 2014.
- Header identifiers are UK GDPR Art. 9(1) health data and come out; the readings are not identifiers and stay.
- The source chart is held under DPA 2018 Schedule 1, Part 1, paragraph 2 (health or social care purposes).
- Local OCR reads scanned paper charts; nothing is uploaded.
Anonymise vital signs charts offline — see plans & start free →
Limitations & cautions
Scanned paper charts depend on OCR, and faint handwriting causes errors, so check the header the tool found before you export. The readings themselves stay untouched and are not treated as identifiers — though an extreme, unusual observation set on a small ward can still narrow identity when combined with a date.
Frequently asked questions
Are the readings ever removed?
No. Every temperature, pulse, blood pressure, respiratory rate, and oxygen saturation stays exactly as recorded, so the NEWS2 score can be recomputed from the clean chart. Only the header identifiers and the clock times change.
How are observation times handled?
Times shift by one steady offset per chart, so the gaps between observations are preserved while the real times are hidden. That keeps the sheet usable for escalation-interval audit, which is the question NICE CG50 and NEWS2 work is usually asking.
Can scanned paper charts be cleaned?
Yes. Local OCR reads them, so identifiers printed or handwritten on paper are caught along with the digital text. Review the low-confidence flags on faint charts, since handwriting is where OCR errors concentrate.