Dental Medical Clearance Request Letters: Use case and AI Confidentiality

A medical clearance request is the letter a dentist writes to a patient’s GP before extractions or surgery when the patient takes a blood thinner or a bone medicine, or has a heart condition. It is a fiddly letter to write between patients, so staff often paste the record into ChatGPT or Claude and ask for a draft. The result carries the patient’s identity, his diagnoses, every medicine, the planned procedure and often something the patient said in confidence. This page marks every item in a routine clearance letter that the Privacy Act, NSW health records law and the Dental Board’s Code of conduct bear on, and shows how the same letter is drafted on Ermos inside the practice.

The medical clearance request letter

Written before extractions or surgery for a patient whose heart condition or medicines change how the procedure is planned. The Code of conduct (section 5.4, item 3) asks practitioners to share sufficient, timely information about the patient and the treatment needed so that care can continue safely, and section 3.3 asks them to seek informed consent before disclosing information and to document it where possible.

Tallowwood Dental · Shop 4, 17 Wombat Lane, Lane Cove NSW 2066
Tuesday 17 November 2026

Dr Meera Banerjee
Banerjee Family Medicine, Suite 2, 40 Brolga Street, Lane Cove NSW 2066

Re: . Request for medical advice before dental extractions.

Dear Dr Banerjee,

He needs two extractions, and I would like your advice on his heart and blood-thinning medicines before I go ahead.

History as he gave it to us.

Planned treatment.

What I would like to know.

Something else.

An invented clearance request. The practice, dentist, GP and patient are fictional and the registration and record numbers and phone number are placeholders. The clinical questions are an example of what a dentist might ask and are not clinical guidance. Hover or tap any highlighted item for the analysis; the full list is below.

Medical clearance request drafter

An Agent+ that prepares a letter asking a patient's doctor about their fitness for dental surgery. It works from the treatment plan and the medical alerts on file, uses the practice's own letter template and includes only what the doctor needs to respond.

What it does

  1. Reads the surgery plan and the medical alerts
  2. Drafts the letter from the practice's template
  3. Flags details that go beyond what the doctor needs

What stays with your people. The dentist reads and signs the letter and confirms the patient has agreed to it being sent.

An Agent+ is built once and anyone in the practice can run it. It works from your own documents and runs on Ermos Dominion, the AI unit in your office, or on Ermos Edge, your own instance in an Australian data centre.

What leaves the practice

  1. 1
    The patient

    A full name, date of birth, home address and record number. Personal information, and health information as well because it was collected in providing a health service (Privacy Act section 6FA). A dental practice that provides a health service and holds health information is covered by the Privacy Act whatever its turnover (section 6D(4)(b)). Patient data guide.

  2. 2
    A symptom he had not reported

    A new symptom the patient described to the dentist and has not told his doctor. Health information, and sensitive information under the Privacy Act, which includes health information (section 6(1)). Telling his GP is a disclosure to provide a further health service, which APP 6.2(a) and Health Privacy Principle 11(1)(b) contemplate where it is directly related to the primary purpose and the patient would reasonably expect it. Sending it to an AI provider is a different recipient for a different purpose. Patient data guide.

  3. 3
    Diagnoses and medicines

    A heart rhythm condition, blood pressure, four named medicines and a bone condition. Health information that bears on the dental risk. The letter should be built from the record rather than from memory, because section 8.3 of the Code asks for records that are accurate, up to date, factual and objective, and an AI draft can drop or invent a medicine.

  4. 4
    The planned procedure

    Teeth, reasons, date and time. A health service to be provided to a named person is health information (section 6FA(a)(iii)). The date also tells a reader when the patient will be in the chair.

  5. 5
    The questions to the GP

    The dentist’s questions about INR, warfarin, antibiotic cover and bone medicine. These are clinical judgements for the dentist and the GP. Ahpra’s guidance on AI says the practitioner remains responsible and must apply human judgment to any output of AI, so an AI-drafted list of questions has to be checked against what this patient needs.

  6. 6
    A confidence about his wife

    A named third party and a request that something be kept from her. Personal information about Beverley and a confidence the patient gave to the practice. Section 3.3 of the Code asks practitioners to respect patients’ confidentiality and privacy and to make sure staff are aware of the need to. A draft pasted elsewhere loses the instruction with it.

  7. 7
    The record of his agreement

    A line saying the patient agreed to the letter and to a reply. Consent is one of the grounds for disclosure in APP 6.1(a) and Health Privacy Principle 11(1)(a), and section 3.3 of the Code asks practitioners to document consent where possible. Ahpra’s AI guidance asks practitioners to involve patients in the decision to use an AI tool that needs their personal data, and agreement to a letter to the GP does not by itself reach an AI provider. Patient data guide.

  8. 8
    The practice and the dentist

    The practice, the dentist, her registration number and the reply contact. Largely public, since registration is on Ahpra’s public register, but it ties a leaked letter to a practice and a named practitioner.

  9. 9
    The closing lines

    Thanks, an offer to phone and a reply date. Generic wording that says nothing about any patient.

What happens in Frontier Models (ChatGPT, Claude)

  • The whole document is transmitted to the provider’s servers. On personal plans the user cannot choose where it is stored or processed, and for the major providers that has generally meant outside Australia, under that country’s legal process. Some business plans offer storage in Australia, but the provider still receives the content.
  • On consumer tiers, prompts can be retained and used to train future models.
  • The provider’s staff and systems can reach the content under the provider’s own policies.
  • Once sent, the practice no longer controls it. Retention and deletion happen on the provider’s terms.

The same document on Ermos

Ermos Dominion

The document is drafted on a unit in your office, from your own templates and the patient record. Every marked item stays on the unit, which keeps working with the internet switched off. Nothing in it is transmitted to an AI provider, so there is no disclosure to analyse: the confidentiality work is the same as for your file server.

Ermos Edge

Your documents, the index and your access boundaries stay on your own instance in an Australian data centre. When a staff member asks for this draft, only their question and the excerpts they are permitted to see go to a named Australian provider, who is contractually barred from keeping the content or training any model on it. Nothing is processed outside Australia.

Frontier Models (ChatGPT, Claude)Ermos EdgeErmos Dominion
Where the document is processedProvider servers, generally offshore on personal plansYour instance in an Australian data centre; answers generated in AustraliaThe unit in your office
Who can read itProvider staff and systems, under policyYour staff, within the boundaries you set. The named Australian provider processes only permitted excerpts, under contract, and keeps nothingYour staff only
Is it keptOn the provider’s termsKept on your instance; the provider is barred from retaining itStays on the unit
Can it train a modelOn consumer tiers, yesNo, by contractOnly your own unit learns from it, on your hardware

Part of Use cases in dental practices.

Common questions

Do ChatGPT and Claude use the information in a medical clearance request letter?
Everything pasted or uploaded is sent to the AI company's servers and processed there to produce a reply. As at October 2026, whether a conversation on a personal plan (ChatGPT Free, Plus or Pro; Claude Free, Pro or Max) can be used to train the models depends on a setting each user controls. Business plans (ChatGPT Business and Enterprise, Claude Team and Enterprise) and the developer APIs sit under commercial terms, and OpenAI and Anthropic each say they do not train on that content by default (OpenAI, Anthropic). On either product, the thumbs up and thumbs down buttons can send a whole conversation back for training. This letter carries the patient's name, date of birth and address, his heart condition and four named medicines, the planned extractions, and a chest symptom he has not mentioned to his GP.
Is it safe to paste a medical clearance request letter into ChatGPT or Claude if we remove the patient's name?
No, because the letter still identifies him. Removing names rarely makes a document anonymous. Information stays personal information while the person is reasonably identifiable, and the OAIC warns that information can be at risk of re-identification even when it has been de-identified or anonymised. A heart rhythm condition, warfarin with INR checks at a named GP practice, a bone medicine taken since 2020 and a dated surgery together point to one man. His wife's first name and his request that she not be told about the chest tightness add a confidence about a named third party. The Office of the Australian Information Commissioner (OAIC) recommends in its guidance on commercially available AI products that organisations do not enter personal information, and particularly sensitive information, into publicly available generative AI tools, because of the significant and complex privacy risks.
Can ChatGPT or Claude be trusted to draft a medical clearance request letter?
Not without the dentist checking every line against the patient's record. Language models can produce fluent text with details that are not in the source, wrong figures or dates, or references that look real but are not. In this letter the most harmful errors would be a dropped or invented medicine, a wrong tooth number, a wrong surgery date or a missing question about INR, antibiotic cover or bone medicine. Those questions are clinical judgements for Dr Rao and the GP, so a person must decide what is asked.
How can a dental practice use AI for a medical clearance request letter without sending it to ChatGPT or Claude?
Ermos is built so the document stays under your control. On the Dominion AI unit it is drafted inside your office and none of it is sent to an AI provider. On Ermos Edge your documents and index stay on your own instance in an Australian data centre, and only the question and the excerpts the user is permitted to see go to a named Australian provider that is contractually barred from keeping them or training any model on them. The letter is drafted from the patient's record and the practice's own template. Dr Rao checks it against the record, and a person asks for and records the patient's agreement before it goes to the GP.
Can a dentist send a patient's health history to the GP?
The Privacy Act allows a practice to disclose health information for a secondary purpose that is directly related to the primary purpose where the patient would reasonably expect it (APP 6.2(a)), and NSW Health Privacy Principle 11 has a similar rule that includes disclosure to provide a further health service. The Dental Board's Code of conduct also asks practitioners to seek informed consent before disclosing information and to document it where possible.
What does the Code of conduct say about sharing patient information with other practitioners?
Section 5.4 of the shared Code of conduct asks practitioners who delegate, refer or hand over care to share sufficient, timely information about the patient and the treatment needed so that care can continue safely. Section 3.3 asks them to respect confidentiality and privacy and to seek informed consent before disclosing information. A clearance request sits between those two duties.
Is it safe to paste a clearance letter into ChatGPT or Claude?
Not with the patient's identity and history in it. Ahpra's AI guidance says the practitioner remains responsible, must apply human judgment to AI output, should check whether patient data is used to train the model and where it is stored, and should involve patients where a tool needs their personal data. A blank template with no patient details raises none of those questions.

See Dominion answering from your own documents.

A demonstration takes 30 minutes, uses no client data, and comes with no obligation.

Book a demo