Patient registration in Tasy: how to prevent and fix duplicates
A patient with two registrations has two records, two histories and wrong KPIs. Where duplicates come from, how to measure them, how to prevent them at the front desk and how to fix them without losing anything.
A duplicate registration happens when the same person has more than one registration in Tasy. It looks like a front-desk housekeeping issue, but it affects everything downstream: the medical record is split in two, the physician does not see the full history, the payer receives accounts with different data for the same member, and the readmission and patients-seen KPIs are wrong.
Where duplicates come from
Almost always from a registration made without a search, or with a search that did not find the existing record:
- search by name only, with a different spelling: "Luiz" and "Luis", a missing middle name, an abbreviation;
- married or maiden name, preferred name;
- date of birth typed wrong in one of the records;
- rush in the emergency department, with a patient who arrives with no ID;
- newborns registered before the birth certificate and again with the final name;
- integrations from other systems (lab, online booking) that create a record instead of searching.
What it costs
| Where it shows | The effect |
|---|---|
| Medical record | Allergies, diagnoses and tests in two places; a clinical risk |
| Account | Accounts open under two records; documents and authorizations on the wrong one |
| Payer | Member with conflicting data; denials for registration data |
| KPIs | Patients counted twice; readmissions underestimated |
| Privacy | One person's data mixed with another's when the merge is done badly |
How to measure
A simple query shows the size of the problem: records with the same name (no accents, no double spaces) and the same date of birth, or the same ID number on different records. Run it by month of creation and by source (front desk, emergency, integration). Duplicates per thousand new registrations is a good indicator of front-desk quality.
How to prevent it at the front desk
- Search before registering, always. By more than one criterion: full name, date of birth and ID.
- Tolerant search: part of the name, no accents, main surname.
- ID in hand to register, with the spelling on the document.
- Mandatory fields filled in on the spot, not "later".
- Emergency patient without ID: use the hospital's unidentified-patient workflow and complete the record as soon as possible, searching for the existing one.
- Integrations set up to search before they create.
- Training with real, anonymized duplicate cases.
The front desk checklist and safe patient identification complete this routine.
How to fix it without losing history
Duplicates are fixed through the hospital's official workflow, by whoever has permission to merge records, and never by deleting one of them. The routine:
- Freeze: nobody records anything new on either record until a decision is made.
- Confirm it is the same person, with an ID. Matching name and date of birth are not enough: real namesakes exist.
- Choose the record that stays (usually the oldest or the most complete).
- Check open accounts on both records with billing before merging.
- Merge using the Tasy function, which moves encounters, record and accounts to the surviving record.
- Check afterwards: full history, accounts on the right record, payer and member card up to date.
- Record the cause (poor search, integration, emergency) to fix the process.
And in BI data
In a BI, the patient is identified by a code. After a merge in Tasy, the old code stops receiving new encounters, but the old ones move to the surviving record. A well-built incremental load reflects that through the update dates; a load that never re-reads the past keeps the duplicate in the BI. See Tasy incremental loads.
Who writes this
Domínio has worked for more than 20 years with hospitals that run Tasy, from registration to accounting. If duplicates are affecting your hospital's accounts and KPIs, talk to us on WhatsApp or at comercial@dominiocontabil.com.br.
Related questions
Who can merge records?
Only people with the permission and training for it, usually someone from registration or medical records. The front desk identifies and escalates.
Can a merge be undone?
It depends on the function and the hospital's setup. That is why confirming it is the same person, with an ID, comes first.
Are namesakes duplicates?
No. Two people with the same name and even the same date of birth exist. Confirmation is by ID and contact details.
How often should we measure?
Monthly, by registration source. A sudden rise usually has a simple cause: a new staff member, a new integration or a change in the search screen.
Written by the Domínio team behind Clic Health: accountants, controllers and developers with more than 20 years of work with hospitals that run Tasy. Have a question about Tasy? Talk to us on WhatsApp.