The European Commission has submitted a proposal to amend the MDR and IVDR with the aim of simplifying regulatory requirements for medical devices. The focus is also on the application of MDR Rule 11 on the classification of software. The background to this is ongoing concerns that the existing regulations are often too complex, cost-intensive and difficult to plan for manufacturers.

Classification as Class I would have tangible benefits for manufacturers in particular: The involvement of notified bodies would not be necessary, which would reduce costs and additional regulatory work. At the same time, market entries, updates and further developments could be implemented much more quickly. Against this background, the European Commission’s proposal to revise Rule 11 is of particular importance. A clearer and more practical formulation could increase legal certainty and strengthen innovation and planning security in the field of medical software.

Since the MDR came into force, there have been clear challenges in practice:

The European Commission is responding to these developments with the new draft. The explicit aim is not to lower safety standards, but to create a more practical and manageable legal framework that enables innovation while ensuring security of supply.

Previous application

Until now, MDR Rule 11 was interpreted in such a way that software was first checked to see whether it supported diagnostic or therapeutic decisions. If this was the case, it was generally classified as Class IIa, IIb or III. Only software without a corresponding reference was assigned to class I.

The main problem with this system was the considerable scope for interpretation. In practice, there are different opinions as to whether software can be classified as Class I at all. While some stakeholders assume that practically all medical software is at least Class IIa, others take the view that diagnosis and treatment are carried out exclusively by doctors and that purely patient-focused applications can therefore continue to be Class I.

The MDCG document MDCG 2019-11 Rev. 1 in particular should help to clarify these questions. However, the examples of stand-alone Class I software mentioned there, such as an application for determining fertile days or a communication aid for people with speech disorders, are very narrowly defined and offer only limited guidance for other use cases. For many software products, particularly in the field of digital health applications, the classification issue therefore remains unclear.

New approach in the Commission draft

The new draft reverses the previous logic of MDR Rule 11. In future, medical software will generally be classified as Class I unless clearly defined escalation criteria are met. Upgrading will only take place if the risk to patient safety associated with the software justifies this.

Excerpt from the draft (Rule 11):

“Software which is intended to generate an output that confers a clinical benefit and is used for diagnosis, treatment, prevention, monitoring, prediction, prognosis, compensation or alleviation of a disease or condition is classified as class I, unless the output is intended for a disease or condition:
– in a critical situation with a risk of causing death or an irreversible deterioration of a person’s state of health (class III);
– in a serious situation with a risk of causing a serious deterioration of a person’s state of health or a surgical intervention, or to drive clinical management in a critical situation (class IIb);
– in a non-serious situation, or to drive clinical management in a serious situation or to inform clinical management in a critical or serious situation (class IIa).”

Source

This means that the focus is no longer solely on the abstract purpose of the software, but on the specific clinical risk resulting from its use.

Classification as Class I would have considerable practical advantages for manufacturers:

Against this backdrop, the European Commission’s proposal to revise MDR Rule 11 is of particular importance. A clearer and more practical formulation could lead to greater legal certainty and at the same time promote innovation and planning security in the field of medical software.

However, a critical examination shows that the hoped-for simplification does not apply equally to all software classes. Although Class I is explicitly mentioned as the starting point for the first time in the new wording, the subsequent differentiation is strongly based on existing IMDRF definitions, in particular the concept of the “non-serious situation”.

This is already described in detail today and includes diseases or situations where correct diagnosis and treatment is important but not critical to avoid irreversible damage. Against this background, it can be argued that the majority of medical software is used at least in a “non-serious situation” and therefore continues to fall into Class IIa.

The new regulatory text may provide more clarity for high-class software (IIb or III). However, there is still room for interpretation for software in the border area to Class IIa. Whether the draft actually leads to a noticeable simplification here will largely depend on the future interpretation by authorities and market surveillance.

The new approach could reduce national differences and scope for interpretation, but it does not release manufacturers from the obligation to make a clear, risk-based argument. Manufacturers must continue to explain clearly:

The new version of MDR Rule 11 changes the perspective on classification: Class I is mentioned more frequently as the starting point for the assessment. However, this does not mean that medical software automatically qualifies as Class I. Rather, manufacturers must continue to provide sound evidence that the software does not pose a higher clinical risk. Class I is therefore not automatic, but the starting point for a comprehensible and robust regulatory justification.

The draft revision of MDR Rule 11 is a step towards a more risk-oriented and practical regulation of medical software. For manufacturers, it generally opens up opportunities for faster market entry, less certification effort and more flexibility for further developments.

However, a comprehensive simplification is not to be expected. Although Class I will be named as the starting point for classification in future, the distinction will continue to be based on existing IMDRF definitions, in particular the concept of the “non-serious situation”. This means that many software products will continue to be assigned to at least Class IIa in the future.

Regulatory uncertainties therefore remain. Classification as Class I is not a free pass, but still requires a clear risk-based justification as well as complete documentation, risk management and clinical evaluation.

Tip: Also take a look at our other DiGA services such as the Health ID service or the ePA service. Together they form the complete technical basis for your DiGA.