A clinical search feels like a small interaction: type a question, read an answer, close the tab. Behind that interaction sit several different kinds of information. There is the question itself, the account used to ask it, the technical information needed to run the service and any optional analytics or advertising activity. Understanding those categories is more useful than judging privacy from a company's nationality or a reassuring sentence on a landing page.
Standard Medicine's published information provides a useful example. Its European positioning makes the service relevant to clinicians considering alternatives in this market, while its privacy policy explains more precisely how particular parts of the service operate.
Begin with the data the service describes
The privacy policy updated on 8 October 2026 identifies Standard Medicine AB as the controller and describes account information, submitted questions, generated results and usage data. It instructs users not to enter patient-identifiable information.
These categories should be considered separately. A professional account can identify the clinician without a question identifying a patient. A technical log can record activity without being the clinical content of that activity. Conversely, a short free-text question can contain unnecessary detail even when it does not look like a formal medical record.
The first useful habit is therefore to identify the actual information need before typing. If the task is to understand a general recommendation, a general question may be sufficient. The clinical question should determine what information is included, rather than the convenience of copying an existing note.
Removing a name is not the whole task
The Information Commissioner's Office explains that pseudonymisation and anonymisation are different. Information that can be linked back to an individual using additional material can remain personal data. This is relevant whenever a clinician assumes that deleting a name makes a detailed narrative anonymous.
For a literature question, unnecessary details often have no educational value. A precise date, workplace, unusual family circumstance or detailed chronology may contribute little to the concept being explored. The clinician can ask about the relevant population, finding or comparison without reproducing the individual story that prompted the question.
There is a further practical benefit. A concise information question is easier to verify than an entire narrative mixed with assumptions. It encourages the user to distinguish what they know from what they are trying to establish. That improves the search task while also keeping irrelevant material out of the interaction.
European ownership and processing location are separate questions
Standard Medicine's policy says its search index remains in the EU, while some language-model and ranking processing occurs outside the EEA under Standard Contractual Clauses. The published architecture therefore cannot be summarised as exclusively European processing.
The distinction is about precision. A company's headquarters, the location of a search index and the location of a model processing a query describe different parts of a service. One cannot be used as a substitute for the others. A clinician choosing software should be able to ask about each without turning the discussion into a broad judgement about a country or vendor.
For an organisation considering deployment, the practical next step is to map the intended information flow. What would staff submit? Which service receives it? What is retained, and where? Those questions produce a more useful decision than relying on a broad label such as European, local or cloud-based.
What the mobile app labels add
The UK App Store listing includes developer-declared privacy categories for tracking, information linked to users and advertising-related uses. Apple states that these declarations are supplied by the developer and are not verified by Apple. They are useful prompts to read the detailed policy and inspect the permissions offered in the app.
An app label describes categories rather than the complete circumstances of every interaction. The clinician should therefore look for the explanation of purpose and choice. Information used to measure whether an advertisement led to an installation is a different question from how a clinical search is processed. Combining them into a single claim that the platform sells clinical queries would be unjustified.
Equally, a doctor should not assume that every optional permission is necessary to obtain an answer. Understanding the setting is part of choosing the experience they want. This is especially relevant when moving between a web browser and an app, where the controls and operating-system permissions may appear in different places.
Understand consent, retention and deletion separately
The policy describes consent-based marketing cookies in the UK and EEA and consent-based mobile advertising attribution. It also describes queries retained in pseudonymised form for the period needed to operate and improve the service, rather than giving a single fixed query-retention period.
These are separate decisions for the user to understand. Declining an optional marketing permission does not answer every question about the information needed to operate the service. Deleting an account is also a different action from changing a cookie preference. A clear explanation should make those differences visible.
Where the published description is not specific enough for an organisational requirement, the useful response is a precise question. Ask which records the requirement concerns, how long they remain available and what deletion covers. A targeted enquiry about a particular data category is easier to answer than asking whether an entire product is completely private.
Educational work does not need to begin with a patient record
Many reasons for opening a clinical platform are educational. A doctor wants to understand a distinction, prepare for an examination or practise explaining a decision. In those situations, purpose-built questions and cases can provide the context required for learning without making a real consultation the starting material.
This is a practical strength of iatroX's broader learning proposition. Its Socratic Tutor works around an answered practice question, keeping the discussion attached to that question and its examination context. The learner can explain their reasoning, explore the distinction and retain useful learning points.
For a clinician whose main goal is to improve understanding, that is often a better-shaped task than pasting a real narrative into a search box. It gives the conversation a defined educational purpose and a manageable scope. It also makes it easier to judge progress: can the learner explain why their original answer was tempting, and why another answer fits the question better?
This is an argument for choosing the right workflow. It does not require a claim that one company's privacy arrangements are universally superior. A reference query, a structured learning session and patient-specific documentation each need to be considered on their own terms.
A useful reading of the policy
Standard Medicine's privacy information is most useful when read as an account of particular activities: operating search, managing accounts, measuring use and handling optional advertising permissions. The doctor's job is to decide which activities fit their intended use and what information that use requires.
For general clinical learning, a well-defined question or a purpose-built educational case is usually a clearer starting point than a copied record. iatroX's dedicated learning tools make that choice practical. The result is a more deliberate routine in which the clinician knows what they are asking, why they are asking it and what information they have chosen to share.
Frequently asked questions
Does Standard Medicine process everything in Europe?
Its policy describes an EU search index but also some AI and ranking processing outside the EEA under Standard Contractual Clauses. Different parts of the service therefore have different processing arrangements.
Can doctors enter patient-identifiable information?
Standard Medicine instructs users not to do so. A general evidence question should contain only the information needed to explain the topic, without reproducing an identifiable patient's record or story.
How can iatroX support learning without copying clinical notes?
The question bank and Tutor provide their own educational context. A clinician can work through a practice question and discuss the reasoning behind an answer, rather than bringing a real patient narrative into the learning session.
Practise clinical reasoning in a defined educational context with iatroX Tutor.