Clinical evaluation — MDR Article 61 & Annex XIV explained

What it is

Clinical evaluation — MDR Article 61 & Annex XIV is the planned, ongoing analysis of clinical data to show a device meets the General Safety and Performance Requirements (GSPRs) for its intended purpose. Manufacturers must generate, appraise, and analyze evidence, then keep it up to date across the lifecycle; the result is a Clinical Evaluation Report (CER) supported by a Clinical Evaluation Plan (CEP) and, when needed, Post-Market Clinical Follow-up (PMCF) (MDR 2017/745 Art. 61; Annex XIV Parts A–B).

Regulatory framework

• Article 61: Requires a clinical evaluation for every device; sets when new clinical investigations are needed and when justified alternatives may apply (e.g., equivalence or well-established technologies) (Art. 61(1), 61(3)–(6)).
• Annex XIV Part A: Methods for planning, identifying, appraising, and analyzing clinical data; outputs include the CEP and CER.
• Annex XIV Part B: PMCF planning, conduct, and reporting; PMCF is expected unless you provide a robust, device-specific justification.
• Related hooks: GSPRs (Annex I), PMS/PSUR (Arts. 83–86), vigilance (Arts. 87–92), technical documentation (Annex II–III).

What it covers

• Evidence scope: Clinical investigations, literature, real-world data, registries, and—if fully justified—data from an equivalent device with access to the underlying files (Annex XIV Part A §3).
• Claims linkage: Endpoints and outcomes that match the intended purpose, indications, users, and settings.
• Benefit–risk: A reasoned conclusion that the benefits outweigh the risks for each claimed indication and population.
• Updates: Continuous integration of PMS and PMCF findings into the CER, labeling, and risk files.

Process — how it works

• 1) Plan: Write the CEP with scope, endpoints, acceptance criteria, search strategy, appraisal tools, comparators, and PMCF trigger logic (Annex XIV Part A).
• 2) Identify data: Collect internal and external clinical evidence; include legacy data if still relevant and adequate.
• 3) Appraise & weight: Judge relevance, reliability, and robustness; document methods and rationales.
• 4) Analyze: Synthesize results against GSPRs and intended purpose; address gaps and uncertainties.
• 5) Conclude: Draft the CER with clear benefit–risk conclusions, supported claims, and residual-risk rationale.
• 6) PMCF: Plan and execute PMCF when needed; alternatively, justify why PMCF is not necessary and keep the justification current (Annex XIV Part B).

Clinical evaluation vs clinical investigation

• Clinical evaluation: Desk-based analysis of all relevant clinical data; mandatory for every device and updated over time (Art. 61; Annex XIV Part A).
• Clinical investigation: A human study that generates data; required when existing evidence is not enough and, by default, for most implantable and Class III devices unless a justified exception applies (Art. 61(4)–(6)).

Common pitfalls

• Equivalence claimed without full technical, biological, and clinical comparison and access to the data (Annex XIV Part A §3).
• Endpoints that do not align with claims, users, or settings; therefore, evidence fails to support labeling.
• Literature reviews without transparent methods, inclusion/exclusion rules, or quality appraisal.
• Static CERs that ignore PMS/PMCF signals or do not update benefit–risk and labeling.
• Assuming PMCF is optional; if you do not run PMCF, you must justify why not and revisit the rationale.

Quick checks

• Are claims specific, testable, and mapped to endpoints and success criteria in the CEP?
• Do appraisals show why each data source is relevant, reliable, and robust?
• Is the benefit–risk positive for each indication and user group, not just overall?
• If you rely on equivalence, do you have access to detailed data and full matching across all three dimensions?
• Does the CER pull in PMS/PMCF results and update labeling and risk files accordingly?