Is a symptom checker a medical device?
If your app takes symptoms and produces a diagnosis or triage recommendation, the answer is almost certainly yes.
General information only. This guide does not constitute legal, regulatory, or professional advice. Regulatory classification depends on the specific facts of your product, your intended market, and how your software is described and used. Rules change, and the information here may not reflect the most current guidance. If you are unsure whether your software is a medical device, seek advice from a qualified regulatory consultant or legal professional before making any compliance decisions. QualiHQ makes no representations about the accuracy or completeness of this content and accepts no liability for any decision made in reliance on it.
Symptom checkers occupy a narrow regulatory category with almost no grey zone. If your software takes a user's symptoms as input and produces any output that influences what they do about their health -- see a doctor, go to A&E, wait it out, take this medication -- it is a medical device. Not probably. Not depending on how you frame it. Almost certainly.
This is not a quirk of conservative regulatory interpretation. It is the direct result of how the EU drew the line in MDR and MDCG 2019-11, and understanding why they drew it there explains why disclaimers, reframings, and "general information" positioning do not hold.
Why "we just show information" does not work
The most common mitigation attempt is positioning a symptom checker as a health information tool rather than a diagnostic tool. "We do not diagnose. We just show you relevant information about your symptoms." This argument fails because EU MDR is not written around what the software claims to do. It is written around what the software actually does and how it is used.
MDCG 2019-11 defines the boundary precisely: software that takes patient-specific data as input and provides output "used to take decisions with diagnosis or therapeutic purposes" is a medical device. The word "used" matters. Your software does not need to diagnose. It needs only to provide information that is used in a diagnostic decision. A symptom checker by definition provides exactly that.
When a user enters "chest pain, shortness of breath, left arm tingling" and your app returns a list of possible conditions ranked by likelihood, that output is being used to make a decision about whether to call an ambulance. It does not matter that you called it "information." It does not matter that you added "this is not medical advice." The function of the software is to influence a health decision, and that is the test the regulation applies.
The specific outputs that trigger classification
Not all health software that mentions symptoms is a medical device. What matters is the output and how it relates to a specific individual's care. Here is where the line sits:
What classification means in practice
Most symptom checkers that produce differential diagnoses or triage recommendations would be classified as Class IIa under EU MDR Rule 11. That requires:
The classification pathway exists and is well-trodden. What trips most founders up is starting the process late -- after they have made clinical claims publicly, after they have signed enterprise contracts, after they have raised on the basis of a product that turns out to need significant compliance investment.
The earlier you start, the cheaper it is
The QMS, the technical file, the clinical evidence -- none of this is optional for a regulated symptom checker. But the cost of building it scales with how early you start. A QMS built from the first release captures evidence as you go. A QMS built after the fact has to reconstruct it.
QualiHQ is designed for exactly this situation: founders who know they are building something regulated and want to build the compliance infrastructure in parallel with the product, not after a notified body asks for it.
Check your classification →Common questions
Can a symptom checker avoid regulation by limiting its output?
Potentially, yes. If you redesign the output so it no longer produces a personalised assessment -- for example, returning only general health information without ranking conditions or making a triage call -- you may move outside scope. But this usually means fundamentally changing what the product does. Most founders who try this route end up with a product that is no longer useful.
What class of medical device is a symptom checker?
Most symptom checkers would be classified as Class IIa under EU MDR Rule 11 (software intended to provide information for making decisions with diagnosis or therapeutic purposes where an error could cause serious deterioration). Some may be Class IIb depending on the severity of conditions covered.
Can I avoid being a medical device by adding a disclaimer?
No. A disclaimer does not change the intended purpose of the software. If your app produces a clinical output -- a diagnosis list, a triage decision, a care recommendation -- it is a medical device regardless of any disclaimer text. Regulators look at what the software does and how it is marketed.
Related scenarios
Not sure where your app sits? The tool takes two minutes.
Check my app →