MDR Classification of Medical Devices: Class I–III

The risk class of a medical device determines which conformity assessment route applies, whether a notified body must be involved, and how extensive the technical documentation and clinical evaluation need to be. Getting the classification wrong either costs unnecessary time and money or, worse, puts the legality of placing the device on the market at risk. This article explains how classification works under the Medical Device Regulation (MDR).

Legal basis: Article 51 and Annex VIII

Regulation (EU) 2017/745 distinguishes four risk classes: I, IIa, IIb and III. The basis is Article 51 together with Annex VIII, which contains

• Chapter I: definitions, for example on duration of use and invasiveness,
• Chapter II: implementing rules on how the classification rules are applied,
• Chapter III: the 22 classification rules.

What counts is always the intended purpose defined by the manufacturer, not every technically possible use. Wording in the instructions for use, on the label and in marketing material should therefore match the intended class from the outset.

The four risk classes at a glance

Class	Risk	Notified body	Examples
I	low	no (exceptions: Is, Im, Ir)	walking aids, wheelchairs, non-sterile dressings for intact skin
IIa	medium	yes	many diagnostic software products, hearing aids
IIb	elevated	yes	ventilators, many long-term implantable devices
III	high	yes	hip, knee and shoulder joint replacements, breast implants, devices in direct contact with the heart or central circulatory system

Within class I there are three special cases where a notified body must be involved for one specific aspect:

• Is: devices placed on the market in sterile condition (sterility)
• Im: devices with a measuring function (metrology)
• Ir: reusable surgical instruments (reprocessing)

The criteria behind the rules

The 22 rules build on a small number of criteria:

Duration of use (Annex VIII, Chapter I):

• transient: normally intended for continuous use for less than 60 minutes
• short-term: normally intended for continuous use for between 60 minutes and 30 days
• long-term: normally intended for continuous use for more than 30 days

Invasiveness: non-invasive, invasive via a body orifice, surgically invasive or implantable.

Active device: does the device depend on a source of energy other than that generated by the human body or gravity?

Anatomical location and effect: for example contact with the central nervous system, the heart or the central circulatory system, or delivery of energy or medicinal products.

The 22 rules in four groups

• Rules 1 to 4: non-invasive devices. Rule 1 sets the baseline: non-invasive devices are class I unless another rule applies.
• Rules 5 to 8: invasive devices. The longer and deeper the intervention, the higher the class. Rule 8 covers implantable and long-term surgically invasive devices, which are generally class IIb and in certain cases class III.
• Rules 9 to 13: active devices, such as therapeutic devices that deliver energy, active diagnostic devices and software (Rule 11).
• Rules 14 to 22: special rules, including devices incorporating medicinal substances, contraception, disinfection, nanomaterials and substance-based devices.

Software under Rule 11

Rule 11 brought some of the biggest changes under the MDR. Many software products that were class I under the old directive (MDD) are now at least class IIa. In simplified terms:

• Software that provides information used to take decisions with diagnostic or therapeutic purposes is class IIa.
• If those decisions may cause a serious deterioration of health or a surgical intervention, it is class IIb.
• If they may cause death or an irreversible deterioration of health, it is class III.
• Software intended to monitor physiological processes is class IIa, or class IIb when it monitors vital parameters where variations could result in immediate danger.
• All other software is class I.

First, it must be clarified whether the software is a medical device at all (qualification). Guidance is provided by MDCG 2019-11 on the qualification and classification of software.

Key implementing rules

Some of the implementing rules in Annex VIII Chapter II are often overlooked in practice:

• The strictest rule applies: if several rules apply, the one resulting in the highest class prevails.
• Accessories are classified in their own right, separately from the device they are used with.
• Software that drives or influences the use of a device falls into the same class as the device. Independent software is classified in its own right.
• Multiple intended purposes: the most critical one determines the class.

What follows from the class

The class determines the route to CE marking:

• Class I: the manufacturer draws up the technical documentation and the EU declaration of conformity under its own responsibility. For Is, Im and Ir, a notified body reviews the respective aspects.
• Class IIa and IIb: a notified body assesses the quality management system and samples the technical documentation. The sampling increases with the class; for implantable class IIb devices, each device is generally assessed individually.
• Class III: the notified body assesses the QMS and the technical documentation of every device. For certain high-risk devices, additional procedures apply, such as consultation of an expert panel on the clinical evaluation.

As the class increases, so do the requirements for clinical data, for post-market surveillance (a PMS report for class I, a periodic safety update report, PSUR, from class IIa) and for the frequency of reviews.

Transition periods for legacy devices

Devices with a valid certificate or declaration of conformity under the old directive may, under certain conditions, still be placed on the market during a transition period. Regulation (EU) 2023/607 extended these deadlines to 31 December 2027 (class III and implantable class IIb devices, with exceptions) and 31 December 2028 (all other devices). Conditions include, among others, that the manufacturer has set up a QMS in line with the MDR and applied to a notified body in time. Manufacturers whose devices are up-classified by Rule 11 or other rules should review their situation early.

Common classification mistakes

• Intended purpose worded too broadly: marketing claims such as "supports diagnosis" can push the class upwards.
• Only one rule checked: all 22 rules must be considered and their application documented.
• Accessories and software forgotten: both are assessed in their own right.
• Old MDD classification carried over: rules have changed, especially for software, substance-based devices and reusable instruments.
• Rationale not documented: the classification belongs in the technical documentation with a traceable justification.

MDCG 2021-24, the guidance on the classification of medical devices, offers many practical examples. All guidance documents are available on the European Commission website.

Conclusion

Classification is one of the first and most consequential regulatory decisions in any MedTech project. A carefully worded intended purpose, a systematic review of all rules and a documented rationale create certainty for timeline, budget and the choice of notified body. From class IIa onwards, a QMS is also required, usually according to ISO 13485, along with solid preparation for the notified body audit.

For questions about classifying a device or your regulatory strategy, an outside perspective helps. Learn more about our MedTech consulting.