For many years under the Medical Devices Directive, a well-conducted literature review was a defensible foundation for clinical evaluation. Under EU MDR, that approach faces increasing scrutiny — particularly for Class IIb and Class III devices, and for devices that cannot demonstrate equivalence to another device with available device-specific clinical data. Understanding where a literature-only approach remains viable and where it does not is central to planning a defensible clinical evaluation strategy.

What the MDR says about clinical data

EU MDR Article 61 and Annex XIV Part A require clinical evaluation to be based on clinical data that is either generated from clinical investigations of the device itself, derived from equivalent devices, or obtained from published literature. Critically, MDR raises the bar on all three routes. Clinical investigations have more robust requirements. Equivalence is more tightly defined. And literature-based evaluations must demonstrate that the published data is sufficient to evaluate the specific performance and safety profile of the device in question — not just the device category.

The core problem with pure literature CERs

Literature describes the state of the art — what the published evidence says about device categories, clinical outcomes in comparable populations, and the known risk profile of similar technologies. It does not, by itself, describe the performance of your specific device. A CER that draws exclusively on literature data is making an implicit argument that the published evidence about comparable devices is transferable to the device being evaluated. Under MDR, that argument needs to be made explicitly — with a documented justification for why the literature evidence is adequate to support the benefit-risk conclusions for this device.

For well-established technologies with long post-market histories and robust published evidence, this argument may be sustainable. For devices with novel design features, new materials, new intended populations, or limited published clinical data, it typically is not.

The equivalence route under MDR

Article 61(4) allows clinical data from equivalent devices to support a clinical evaluation, but MDR tightens the equivalence requirements significantly compared to MDD. Technical, biological, and clinical equivalence must each be demonstrated, and under Article 61(4), manufacturers must have contractual access to the technical documentation of the equivalent device if it is manufactured by a different company — a requirement that has made equivalence to competitor devices practically very difficult to establish.

For devices within the same manufacturer's portfolio — different sizes of the same design, for example — equivalence may remain viable where the technical and clinical differences can be rigorously documented. The key test is whether those differences could reasonably affect safety or performance, and whether that question can be answered from the existing evidence base.

What device-specific clinical data looks like

Device-specific clinical data is not limited to prospective clinical investigations. Under MDR, post-market clinical data — PMCF findings, post-market study results, real-world performance data from registries, and long-term follow-up data from implant databases — qualifies as device-specific clinical evidence and can significantly strengthen a literature-based CER.

For manufacturers with devices that have been on the market for several years, an organised review of complaint data, post-market literature, and any available clinical use data often surfaces device-specific evidence that was not previously consolidated into the CER. This does not always require new clinical activity — it requires a structured assessment of what data already exists.

Where literature-only approaches may still hold

A literature-based CER remains more defensible in a narrower set of circumstances under MDR:

  • Class I and lower-risk Class IIa devices with well-established mechanisms of action and minimal residual risk profiles
  • Devices in categories with extensive published clinical literature covering the specific intended purpose and patient population
  • Devices where PMS data — complaint rates, post-market literature, PMCF outputs — strongly corroborates the conclusions of the literature review

Even in these cases, the CER needs to justify explicitly why the literature evidence is sufficient for this device — not simply present the literature and assert that it supports the benefit-risk conclusions.

Assessing your exposure

For teams reviewing a literature-based CER ahead of an MDR renewal or Notified Body surveillance audit, the most useful first step is a gap assessment that asks a focused question: is the clinical evidence in this CER specific enough to support the benefit-risk conclusions for this device, and if not, what is the most efficient route to strengthen it? For some devices, that means a PMCF design exercise. For others, it means a targeted literature consolidation. For a few, it means a clinical investigation — but that is rarely the only option.