If you're new to EU MDR, classification is the first real decision point in your regulatory journey — not a formality to get through before the "real" work starts. Get it wrong, and everything downstream (your conformity assessment route, your Notified Body involvement, the depth of your technical documentation) gets built on a shaky foundation.
Here's how it actually works, in plain terms.
Under Regulation (EU) 2017/745 (MDR), every medical device sold in the EU market must be assigned a risk class before you can determine anything else about your regulatory pathway. Classification decides:
Which conformity assessment route applies to your device
Whether a Notified Body needs to be involved at all
How deep your technical documentation and clinical evidence requirements go
What your post-market surveillance obligations look like
Misclassifying a device — even unintentionally — can lead to submission delays, rejected applications, or enforcement action later. This is why classification is treated as a serious analytical step, not paperwork.
Most people learn MDR classification as four risk classes:
Class I — lowest risk
Class IIa — low-to-medium risk
Class IIb — medium-to-high risk
Class III — highest risk
In practice, Class I has three sub-categories that carry additional requirements: Class Is (sterile devices), Class Im (devices with a measuring function), and Class Ir (reusable surgical instruments). So while people commonly say "four classes," the practical picture under MDR is closer to seven categories once these sub-classes are accounted for.
The higher the class, the more rigorous the conformity assessment — and the more likely a Notified Body needs to review your technical file before you can affix a CE mark.
Classification isn't a lookup table — it's a rule-based system set out in Annex VIII of the MDR, which contains 22 rules organized into four groups:
Rules 1–4: Non-invasive devices
Rules 5–8: Invasive devices
Rules 9–13: Active devices
Rules 14–22: Special rules (covering categories like devices incorporating medicinal substances, devices made of nanomaterials, and software)
Each rule applies criteria such as duration of contact with the body (transient, short-term, or long-term), degree of invasiveness, and whether the device is active (requires an energy source) or non-active. The rule that matches your device's characteristics determines its class.
One detail that trips up a lot of beginners: if more than one rule could apply to your device, the rule resulting in the higher classification takes precedence. You don't get to choose the more favorable rule — the regulation defaults to the more conservative (higher-risk) classification whenever there's ambiguity.
Misjudging duration-of-use categories. MDR defines duration bands precisely — transient (under 60 minutes), short-term (up to 30 days), and long-term (beyond 30 days) contact with the body. It's easy to underestimate which band a device actually falls into, especially for devices used repeatedly or over extended treatment courses, which can push a device into a higher class than expected.
Assuming accessories inherit a lower risk than the parent device. Accessories are classified in their own right under MDR — not automatically bundled at the same class as the device they're used with. An accessory to a high-risk device isn't necessarily high-risk itself, and the reverse is also true.
Underestimating software classification. Software as a Medical Device (SaMD) is explicitly addressed under Rule 11, and most diagnostic or decision-support software ends up in Class IIa or higher — not Class I, as many newcomers initially assume. If your device includes software with any diagnostic or monitoring function, this rule deserves specific attention rather than an afterthought.
Once your device's class is established, it dictates:
Conformity assessment route — Class I (non-sterile, non-measuring) devices can often be self-certified by the manufacturer; anything above that generally requires Notified Body involvement.
Notified Body engagement — the higher the class, the earlier and more extensively a Notified Body needs to be involved in reviewing your technical documentation.
Technical documentation depth — clinical evidence expectations, risk management file scope, and post-market surveillance requirements all scale with device class.
Getting classification wrong doesn't just delay your submission — it can mean redoing technical documentation built around the wrong regulatory pathway entirely.
Before locking in a classification, ask:
Does my device contact the body at all — and if so, how (surface contact, invasive, surgically invasive)?
How long is that contact — transient, short-term, or long-term?
Does the device require an external energy source to function?
Could more than one Annex VIII rule apply — and if so, have I applied the higher-risk rule?
If my device includes software, have I specifically checked it against Rule 11?
Have I classified accessories separately, rather than assuming they inherit the parent device's class?
If you're unsure after working through these, MDCG 2021-24 (the European Commission's guidance on classification) includes decision trees for common device types, and it's worth consulting directly rather than guessing.
Classification is the foundation, but it's just the first step in an MDR technical file. Once you know your device's class, the real work — technical documentation, clinical evaluation, risk management, and post-market surveillance planning — begins.
Want to go deeper into building an MDR technical file once you've confirmed your device's classification?
Learn with our experts with 1:1 training sessions.
Reach out to us at aptskill6@gmail.com or +91 8262083746