How to Write a Clinical Evaluation Report (CER) Under EU MDR

The Clinical Evaluation Report (CER) is one of the most scrutinized documents in any EU MDR technical file. Over 30% of Notified Body non-conformities relate to clinical evaluation deficiencies. For startups and small companies, the CER is often the document that makes or breaks the conformity assessment process. This guide walks you through what the CER requires and how to get it right.

Legal Basis and Key Guidances

The CER is required under MDR Article 61 (Clinical evaluation and clinical investigations) and must follow the methodology described in Annex XIV Part A (Clinical evaluation). The CER forms part of the technical documentation required by Annex II.

The Medical Device Coordination Group (MDCG) has published several key guidance documents that define expectations:

• MDCG 2020-1: Guidance on clinical evaluation (MDR) — overview of the clinical evaluation process
• MDCG 2020-5: Guidance on clinical evaluation — equivalence — how to claim and justify equivalence to another device
• MDCG 2020-6: Guidance on sufficient clinical evidence for legacy devices — critical for devices transitioning from MDD
• MDCG 2020-13: Guidance on clinical evaluation of legacy devices transitioning under Article 120
• MDCG 2024-3: Guidance on clinical evaluation — literature search and review — updated methodology requirements
• MDCG 2024-10: Guidance on post-market clinical follow-up — PMCF plan and reporting

The Three Data Sources

MDR Annex XIV Part A identifies three sources of clinical data that can support your CER:

1. Clinical Investigations of the Device

Original clinical studies conducted with your specific device. These provide the strongest evidence but are also the most expensive and time-consuming to generate. For Class III devices and implantables, clinical investigations are generally expected unless you can provide strong justification based on existing data.

2. Equivalent Device Data

Clinical data from a device you claim is equivalent to yours. Equivalence must be demonstrated across three dimensions:

• Technical equivalence: Similar design, specifications, materials, manufacturing process, and sterilization methods.
• Biological equivalence: Same materials in contact with the same body tissues for the same duration. For biological equivalence, you need a contract or agreement with the equivalent device manufacturer giving you access to their technical documentation.
• Clinical equivalence: Same intended purpose, same clinical conditions, same patient population, same site in the body, similar severity and stage of disease.

Critical: The biological equivalence requirement effectively means you cannot claim equivalence to a competitor’s device unless you have a contractual data access agreement with them. This is a major change from MDD and makes equivalence claims nearly impossible for most startups.

3. Published Literature

A systematic literature review following MDCG 2024-3 methodology. This includes peer-reviewed studies, registry data, and other published clinical evidence relating to your device or equivalent devices. The review must be systematic, reproducible, and clearly documented with search protocols, inclusion/exclusion criteria, and appraisal of evidence quality.

PMCF Plan Integration

The CER does not exist in isolation — it must be integrated with your Post-Market Clinical Follow-up (PMCF) plan as required by Annex XIV Part B. The PMCF plan should address any residual uncertainties identified in the CER and define ongoing clinical data collection activities.

A typical PMCF plan includes:

• Objectives and scope of ongoing clinical data collection
• Methods: clinical investigations, registries, surveys, literature monitoring
• Timelines and milestones for data collection and CER updates
• Statistical considerations and sample size justification
• Reference to general PMS plan and PSUR/PMSR reporting cycles

Common Notified Body Findings

Based on published NB audit trends, here are the most common CER-related non-conformities and how to avoid them:

1. Inadequate equivalence justification: Claiming equivalence without sufficient technical, biological, or clinical evidence, or without a data access agreement for biological equivalence. Always document each equivalence criterion thoroughly.
2. Non-systematic literature review: Using ad-hoc literature searches instead of a reproducible, systematic methodology. Follow MDCG 2024-3 strictly — document your search strategy, databases used, date ranges, and selection criteria.
3. Failure to identify or address risks: The CER must address all risks identified in the risk management file. Every hazard in your ISO 14971 risk analysis should have corresponding clinical evidence addressing its acceptability.
4. Outdated CER: The CER must be a living document, updated at least annually for higher-risk devices. Notified Bodies check whether the latest clinical data and literature have been incorporated.
5. Weak PMCF plan: A generic or boilerplate PMCF plan that doesn’t address the specific clinical questions and data gaps identified in the CER. The PMCF plan should be tailored to your device.
6. Missing benefit-risk analysis: The CER must include an explicit benefit-risk determination, weighing the clinical benefits against residual risks for each intended purpose.

Startup-Specific Advice

For startups and small companies with limited resources, here are practical recommendations:

• Budget realistically: Expect to spend $50K–$150K on CER preparation, including regulatory consultant or medical writer fees. A CER for a Class III device can cost more.
• Plan clinical investigations early: If equivalence is not viable (and for most startups, it won’t be), you need original clinical data. Build clinical investigation costs and timelines into your product development plan from the beginning.
• Hire or contract a medical writer with MDR experience: The CER requires specific regulatory expertise. General medical writers without MDR/MEDDEV experience will produce documents that fail NB review.
• Start your literature review during development: Don’t wait until the end. Monitoring the literature continuously will identify evidence gaps early and inform your clinical strategy.
• Align CER with risk management: Your CER and ISO 14971 risk management file should reference each other. Reviewers cross-check these documents routinely.