Golden record · PAT

Patient

The person receiving care, across encounters, departments, sites and time.

Also called Service user, member, resident, beneficiary

01 Why it matters

What depends on getting this right

In one sentence

A duplicate patient record is a clinical safety issue before it is a data one, and every measure of outcome, cost per episode and readmission depends on the same person being recognised across a pathway that crosses several systems.

02 Where it lives

Every system holds a different version

None of these is wrong. Each was built for a purpose and records the part of the entity that purpose needed, which is exactly why resolution is required rather than optional.

SystemWhat it holds of this entity
HIS / EMRencounters, diagnoses, procedures and clinical notes
Master patient indexthe identity layer itself, and its duplicate queue
Revenue cycle / billingclaims and the payer view of the same person
LIS / RIS / pharmacydiagnostics and medicines ordered against the encounter
Schedulingappointments, attendance and did-not-attend history

03 Match keys

What actually matches, and what only looks like it does

KeyHow well it works
National health identifierThe strongest key where a national scheme exists.
Local medical record numberReliable within a site, duplicated across a group.
Name, date of birth and sexThe classic triple, insufficient on its own at scale.
Address and next of kinUseful supporting evidence, volatile on its own.

04 Survivorship

When two records disagree, which value wins

Survivorship is a business decision, not a technical default. These rules should be agreed with the people who own the data and then applied consistently, because changing them later restates history.

Rule 01

Never auto-merge on weak evidence

A wrong merge in healthcare joins two people's clinical histories, which is worse than a duplicate.

Rule 02

Demographics from the most recently verified registration

Confirmed at the point of care rather than inferred.

Rule 03

Clinical data is never survived, only linked

Records are associated under one identity; observations themselves are not overwritten.

Rule 04

Every merge and unmerge is auditable

Reversibility is a clinical safety requirement, not a nice-to-have.

05 The cost of not doing it

What stays broken while it is unresolved

Breaks

Fragmented clinical history

A clinician sees part of the picture, which is a direct safety risk.

Breaks

Cost per episode impossible

Consumption across departments cannot be assembled into one pathway.

Breaks

Readmission rates understated

A readmission under a second record looks like a new patient.

Breaks

Duplicate testing

Tests repeated because the prior result is attached to the other identity.

07 Questions

Frequently asked

Why is auto-merging so dangerous here?

Because the cost of a false positive is not a reporting error. Joining two people's records creates a clinical record that describes nobody, so healthcare resolution runs with high match thresholds and a human review queue by design.

What makes the biggest difference to patient matching?

Capturing and verifying a national identifier at registration, plus standardised address and name handling. Most duplicate creation happens at the front desk, not in the data layer.

See the duplicates in your own data

We will resolve one entity on your systems, live, and show what the duplicates are costing.

Book a live demo