When Classification Changes Mid-Development: Clinical Evidence Reset

Written by HATEM RABEH, MD, MSc Ing

Your Clinical Evaluation Expert And Partner

in
S

This scenario appears more often than manufacturers expect. A design modification triggers reclassification. A new intended use crosses into a higher risk category. Or the classification rationale was simply incorrect from the start and the Notified Body flags it during initial review.

The immediate question becomes: what happens to the clinical evidence already gathered?

The answer is rarely simple. And the cost of getting it wrong extends far beyond timeline delays.

Why Classification Changes Occur Mid-Development

Classification boundaries in MDR are more precise than in the previous directives. Rule 11 for software, Rule 8 for implantables, Rule 5 for devices in contact with injured skin—each rule creates clear thresholds.

When a device evolves during development, these thresholds can shift:

A non-invasive monitoring device adds a diagnostic algorithm that influences treatment decisions. Suddenly Rule 11 applies differently.

A Class IIa wound dressing extends its indication to include partial thickness burns. Now it falls under Rule 5(c) as Class IIb.

A reusable surgical instrument design changes to single-use with a drug-eluting coating. The classification jumps from Class I to Class III under Rule 8.

These are not theoretical examples. I see these reclassifications during design control reviews and pre-submission consultations.

The manufacturers often assume the clinical evidence strategy remains valid. It does not.

How Clinical Evidence Requirements Scale With Classification

The MDR does not explicitly state different clinical evidence volumes for each class. Article 61 applies to all devices.

But the underlying expectations change substantially.

For Class I devices, demonstration of safety and performance often relies on bench testing, biocompatibility data, and literature review. Clinical investigations are rarely required unless the device represents a significant innovation.

For Class IIa, clinical data typically includes published literature on equivalent devices plus some level of clinical follow-up or registry data for the specific device. Equivalence pathways are commonly used but must be substantiated.

For Class IIb, expectations rise. Equivalence claims face stricter scrutiny. The clinical investigation dataset must address device-specific performance in representative populations. Literature alone is rarely sufficient unless the technology is well established and the clinical application is not novel.

For Class III, the bar moves higher still. Clinical investigations are almost always required. The evidence must demonstrate not just safety and performance but also that the benefit-risk ratio is favorable compared to existing alternatives. Long-term data becomes critical.

When classification increases mid-development, the clinical evidence that looked adequate yesterday no longer meets the threshold today.

The Equivalence Strategy Becomes Invalid

Equivalence is the most common casualty of mid-development reclassification.

Let’s say you planned to demonstrate equivalence to a predicate device under Class IIa. Your technical and clinical characterization focused on showing comparable materials, design, and mechanism of action. The literature review covered similar devices in the same indication.

Now the device is Class IIb.

The equivalence claim does not automatically carry forward. MDCG 2020-5 is explicit: equivalence must be demonstrated for the actual classification and risk profile of your device. If the predicate was Class IIa and your device is now IIb, the risk profiles differ.

You cannot simply say