Back to Blog

Patient Requests for Doctors: GDPR-Compliant in the medatixx Workflow (2026)

Henri Jung, Co-founder at Superkind
Henri Jung

Co-founder at Superkind

Superkind logo on an abstract orange grain texture

Monday, 08:05. The phone rings for the fourth time before the practice door opens. A patient wants a repeat prescription for blood pressure medication. Another needs a referral to a specialist. A third asks whether lab results have come back. The MFA juggles the calls, the waiting room fills up, and every answer requires a look into medatixx first. For practices running medatixx, this guide shows how to hand the lookup, the drafting and the scheduling to an AI employee, step by step, while the doctor keeps sign-off on every patient-facing reply - and how Superkind’s ready-made AI team for patient requests does exactly that, GDPR-compliant and hosted in Europe.

Key takeaways

The AI drafts, the doctor approves - Superkind reads the patient request, looks up the record in medatixx via computer use, drafts the reply, and waits for the doctor’s OK before anything reaches the patient.

Get a ready-made AI team for patient requests - Superkind comes with a practice team (Request Handler, Appointment Coordinator, Lab Notifier) that handles prescriptions, scheduling and results automatically. Start for free with unlimited users, or build your own.

Keep control, keep data in Europe - health data under DSGVO Article 9 and professional secrecy under section 203 StGB require clear rules. Superkind is hosted in Europe, GDPR compliant, never trains on your data, and you decide which steps need your approval.

Works in medatixx without an API - no integration needed. Superkind works in medatixx natively via computer use, in the browser or the desktop client, just like an MFA would.

Your AI team for patient requests

Request: repeat prescription enalapril 10 mg
Monika Berger
Good morning, I need a repeat prescription for enalapril 10 mg. My last appointment was in September.
Request HandlerAI employee
New email to praxis@...
Patient record checked in medatixx. Last prescription: enalapril 10 mg, 29 Sep. Next check-up due 15 Jan. Draft reply prepared.
MedicationEnalapril 10 mg, 100 pcs
Last visit29 Sep 2026
Check-up due15 Jan 2027
Send prescription confirmationWaits for doctor's OK before replying.ApproveReject
Example: Request Handler at work in Outlook

What does it mean to answer patient requests with AI?

Answering patient requests with AI means an AI employee receives the incoming request (by phone note, email or online form), looks up the relevant patient data in the practice management system, drafts a reply following the practice’s rules, and routes it to the doctor for approval before anything goes to the patient. The AI handles the administrative lookup and the text, not the medical decision. One example is Superkind: an AI team that reads medatixx via computer use, drafts prescription confirmations, appointment suggestions and result notifications, and waits for the doctor’s sign-off - hosted in Europe.

How it works with Superkind: your AI team for patient requests

You write to Superkind by email, in Microsoft Teams or in the Superkind web app, like you would write to a colleague, and Superkind does the task in your programs - including medatixx via computer use, plus 1,700+ integrations such as Outlook, Calendly and Google Calendar. Hosted in Europe, GDPR compliant.

With Superkind you don't start from zero: it comes with ready-made AI employees for medical practices. They run automatically in the background, and you decide where they ask you first. Use them right away, adapt them, or build your own.

Request Handler

AI employee for patient requests

Patients call or email for repeat prescriptions, referral letters and appointment changes, and every answer needs a look at the patient file first. When a new request arrives, the Request Handler opens medatixx in the browser, checks the patient record, drafts the reply, and waits for the doctor's OK before anything goes out.

Works inOutlookMicrosoft Teams+ 1,700 more
Request: repeat prescription enalapril 10 mg
Monika Berger
Good morning, I need a repeat prescription for enalapril 10 mg. My last appointment was in September.
Request HandlerAI employee
New email to praxis@...
Patient record checked in medatixx. Last prescription: enalapril 10 mg, 29 Sep. Next check-up due 15 Jan. Draft reply prepared.
MedicationEnalapril 10 mg, 100 pcs
Last visit29 Sep 2026
Check-up due15 Jan 2027
Send prescription confirmationWaits for doctor's OK before replying.ApproveReject
Example: Request Handler at work in Outlook

Appointment Coordinator

AI employee for appointments

The phone rings, the waiting room is full, and three patients need a new slot this week. Every morning the Appointment Coordinator checks the medatixx calendar for free slots, matches them to open requests in the inbox, and posts the suggestions in Teams so the MFA confirms with one click.

Works inMicrosoft TeamsOutlook+ 1,700 more
AppointmentsTeam
Appointment CoordinatorAI employee
Routine - every morning at 7:15
Calendar checked in medatixx. 4 appointment requests matched to free slots this week.
Monika Berger, blood drawWed 09:30
Thomas Richter, follow-upThu 11:00
Karla Hoffmann, check-upFri 08:45
Runs automatically, no approval needed
Example: Appointment Coordinator at work in Microsoft Teams

Lab Notifier

AI employee for lab results

Lab results sit in medatixx, the patient calls twice, and the MFA cannot answer without the doctor's review. The Lab Notifier watches for new results in medatixx, prepares a plain-language summary, and sends the notification only after the doctor has reviewed and approved.

Works inOutlookMicrosoft Teams+ 1,700 more
Superkind
New chat
My tasks
AI employees
Lab Notifier
Request Handler
Appointment Coordinator
Lab result notification

When new lab results arrive in medatixx: prepare a summary and notify the patient after doctor review.

Reasoning ⌄
Lab Notifier 14:20
New lab results in medatixx

New blood work for Thomas Richter arrived. Summary prepared: HbA1c 6.1 % (in range), cholesterol 212 mg/dl (slightly elevated). Ready for doctor review.

PatientThomas Richter
Key valuesHbA1c 6.1 %, Chol 212
StatusWaiting for doctor review
Send result to patientDoctor reviews before patient is notified.
ApproveReject
Send a message ...
Example: Lab Notifier at work in Superkind

Every employee is ready to use right away. Or build your own on the platform.

Safe by default

Health data under DSGVO Article 9 and professional secrecy under section 203 StGB: Superkind processes everything in Europe, never trains on your data, and a DPA is available through the Trust Center.

What you need

Checklist before the first request

  • A medatixx installation (medatixx, x.isynet, x.comfort or x.concept) with current patient records
  • An inbox where patient requests land: Outlook, Gmail or the x.patient app
  • An agreed rule for which requests need doctor approval (prescriptions, referrals, results) and which the MFA handles alone (appointment confirmations, general information)
  • A confidentiality agreement with the AI service provider under section 203(4) StGB, plus a DPA/AVV - Superkind offers both through its Trust Center
  • A free Superkind account at app.superkind.ai
  • With 1,700+ integrations, Superkind connects to Outlook, Google Calendar and Calendly alongside medatixx via computer use

Step by step with Superkind

Step 1: Connect your inbox and set up medatixx access

Sign up at app.superkind.ai and connect your practice email (Outlook or Gmail). For medatixx, Superkind uses computer use: it logs into medatixx in the browser or the desktop client on your practice workstation, the same way an MFA would. The session stays inside your practice infrastructure. Tell Superkind which patient fields it may read (name, medication, appointments, lab results) and which are off-limits.

  • Check - Superkind only sees the medatixx screens and inbox folders you allowed, and an audit log records every action.
  • Tip - start with one request type (repeat prescriptions) and expand after the first week.

Step 2: Forward the first patient request

Take the next patient email asking for a repeat prescription. Forward it to Superkind, or write in Teams:

OutlookMFA to Superkind

Patient Monika Berger asks for a repeat prescription enalapril 10 mg. Check her record and draft the reply.

Superkind opens medatixx, finds Monika Berger’s patient file, checks the last prescription date, confirms the medication is still current, and drafts a reply: “Prescription is ready for collection from tomorrow, please bring your insurance card.” The draft appears in Outlook for the doctor to approve.

Step 3: Doctor reviews and approves

The doctor sees the draft in Teams or Outlook with the medatixx data attached: last visit, current medication plan, next check-up date. One click to approve, and the reply goes to the patient. If the doctor wants a change (“schedule a blood pressure check first”), they type the correction and Superkind updates the reply.

Step 4: Handle referrals and appointment requests the same way

The same flow works for referral requests (Superkind checks the indication in medatixx and drafts the referral letter), appointment changes (checks the medatixx calendar for free slots and suggests three options) and lab result enquiries (looks up the result status and drafts a notification once the doctor has reviewed the values).

TeamsDr. Weber to Superkind

Thomas Richter’s blood work is in. Summarise the key values and prepare a notification. I will review before it goes out.

Step 5: Build the practice rules

After a few days, tell Superkind the rules that repeat: “Repeat prescriptions for blood pressure medication: check the last visit was within 6 months, draft the confirmation, doctor approves. If the last visit was more than 6 months ago, suggest an appointment instead.” Superkind stores this in its Company Brain and applies it to every future request of that type.

Set it up as a routine

Once the rules are in place, patient requests run as routines. Example: “Every new email to praxis@drweber.de - check whether it is a prescription, referral or appointment request, look up the patient in medatixx, draft the reply, and put it in the approval queue.” The MFA monitors the queue instead of handling each request from scratch.

  • Prescription routine - check medication, last visit date, draft reply, doctor approves.
  • Appointment routine - check medatixx calendar, suggest three slots, MFA confirms.
  • Lab result routine - watch for new results in medatixx, prepare summary, doctor reviews before patient is notified.

Data protection, professional secrecy and EU hosting

Patient data in a medical practice falls under three overlapping rules: DSGVO Article 9 (health data as a special category), section 203 StGB (professional secrecy for physicians) and section 9 of the Model Professional Code for Physicians (MBO-Ae). Using an AI service provider is legally possible under these rules, but it requires specific steps2,4,5.

  • DSGVO Article 9 - health data processing for treatment purposes is covered by Article 9(2)(h) in conjunction with section 22 BDSG. Handing administrative work to a processor (the AI service) requires a DPA/AVV under Article 28 DSGVO3,8.
  • Section 203 StGB - since the 2017 amendment, section 203(3) allows physicians to involve external service providers (“sonstige mitwirkende Personen”) when their participation is necessary for the physician’s work. The provider must be bound to confidentiality under section 203(4), and the physician must document this obligation2,5,6.
  • MBO-Ae section 9 - the professional code mirrors the criminal law duty and extends it to all information learned in the physician’s capacity4.
  • Practical steps - sign the DPA/AVV (available through Superkind’s Trust Center at trust.superkind.ai), add the section 203(4) confidentiality agreement, limit the AI employee’s access to the fields it needs, and keep the audit log.

Other ways to handle patient requests

MethodEffort per requestError riskBest for
SuperkindUnder 1 min (review only)Low (rule-based)All request types
MFA manually3-8 minMediumComplex medical queries
Online reception1-3 min (triage only)LowAppointment booking
Phone answering service2-5 minMediumAfter-hours calls

Common mistakes

  • Skipping the section 203 agreement - a DPA alone does not cover the criminal-law confidentiality duty. The AI provider needs a separate section 203(4) commitment.
  • Sending results without doctor review - lab values and diagnoses must be reviewed by the physician before the patient sees them. Set the approval step as mandatory for these.
  • Giving the AI full access to all patient data - limit access to the fields the AI actually needs: name, appointment, medication, lab status. Diagnosis text and therapy notes stay with the doctor.
  • No audit trail - every AI action on patient data should be logged. This protects the practice if a data protection authority asks.
  • Treating AI replies as medical advice - the AI handles administrative text (prescription ready, appointment confirmed, results available). It does not diagnose, assess or recommend treatment.

6 tips for answering patient requests with AI

1. Start with repeat prescriptions

They are the most common request, the rules are clear (medication, last visit, next check-up), and the risk is low because the doctor still signs off.

2. Keep approval mandatory for clinical content

Prescriptions, referrals and lab results always go through the doctor. Appointment confirmations and general information can run without approval once the rules are set.

3. Use the medatixx task list

medatixx has a built-in task management for recurring items like lab callbacks. Let Superkind read this list to pick up tasks that are already tracked there.

4. Train the team first

Show MFAs how to forward requests and where to check the approval queue. A 15-minute introduction avoids most confusion in the first week.

5. Check data protection before go-live

DPA/AVV signed, section 203(4) agreement in place, access limited to the right fields, audit log active. Document all of this before the first patient request goes through the AI.

6. Review the routines monthly

Patient request patterns change with seasons (flu season, check-up campaigns). Review the rules and add new ones as the practice learns what works.

Conclusion

Answering patient requests with AI does not mean the practice loses control. It means the lookup in medatixx, the drafting and the scheduling happen automatically, while the doctor approves every reply that matters. The fastest way to see whether it fits: forward the next repeat prescription request to Superkind and compare the draft with how you would have answered it yourself. For a broader look at AI in medatixx, read our guide to AI for medatixx.

Frequently Asked Questions

Yes, for the administrative parts. An AI employee like Superkind reads the patient request, looks up the relevant data in medatixx via computer use, and drafts a reply. The doctor or MFA approves before anything reaches the patient. Medical decisions stay with the doctor.

Health data falls under DSGVO Article 9 and needs an explicit legal basis. Superkind is hosted and processed in Europe, GDPR compliant, aligned with the EU AI Act and does not train on your data. A data processing agreement (DPA/AVV) is available through the Trust Center. The practice stays the data controller and decides which data the AI employee may access.

medatixx has no open API for third-party AI tools. Superkind works in medatixx natively via computer use - in the browser or directly in the desktop client - the same way an MFA would: logging in, reading the patient file, checking appointments and entering notes. The session runs inside your practice infrastructure, not outbound. With 1,700+ integrations, Superkind also connects to Outlook, Teams and Calendly.

Section 203(3) StGB, amended in 2017, allows physicians to share patient information with external service providers when their involvement is necessary, provided they are bound to confidentiality under section 203(4). This obligation must be documented. Combined with a DPA/AVV and the GDPR safeguards, an AI service provider can process patient data for a medical practice within these legal bounds.

You start for free with up to 150 hours of work at app.superkind.ai, no credit card and no sales call. The Business plan starts at EUR 50 per month for 50 hours of work, for the whole practice with unlimited users. No price per seat. As of November 2026. See the pricing page at superkind.ai/pricing for details.

Repeat prescription requests, referral letters, appointment scheduling and changes, lab result notifications, insurance form preparation and general information requests. Medical assessments, diagnoses and treatment decisions stay with the doctor. The AI handles the lookup, drafting and scheduling, not the clinical judgement.

Related Articles

Henri Jung, Co-founder at Superkind
Henri Jung

Co-founder of Superkind, the AI employee teams write to in Outlook, Teams and Slack. Henri helps companies, from small teams to the Mittelstand, hand their routine work to AI - without new tools, consultants or an in-house AI team.

Send Superkind your first patient request.

Start for free in Outlook, Teams or Slack. Or let us show you in 30 minutes how it runs in your practice.