Author: Sarah VanVickle Chaves, Director of IVD and Scientific Writing Services, Ph.D. Molecular and Cellular Biology
Systematic literature reviews (SLRs) have become a critical component of clinical evidence generation for medical devices and in vitro diagnostic devices (IVDs). Under the European Medical Device Regulation (EU MDR) and In Vitro Diagnostic Medical Device Regulation (EU IVDR), manufacturers must identify, evaluate, and analyze relevant scientific literature as part of clinical or performance evaluation.
But identifying the literature is only part of the challenge.
Manufacturers also need to demonstrate that decisions about which publications were included or excluded were made systematically, objectively, and with appropriate controls for bias.
This is where dual-reviewer screening becomes important.
Although the MDR and IVDR do not explicitly state that two reviewers must independently screen every literature search result, dual-reviewer screening aligns closely with established systematic-review methodology and with the level of methodological rigor that Notified Bodies expect when assessing clinical evidence.
For manufacturers preparing Clinical Evaluation Reports (CERs), Performance Evaluation Reports (PERs), Scientific Validity Reports (SVRs), or post-market clinical evidence, understanding this distinction is essential.
What Do the MDR and IVDR Actually Require?
The starting point is the regulatory text.
Under EU MDR 2017/745, Annex XIV, Part A, manufacturers must identify available clinical data relevant to the device and its intended purpose—and identify gaps in clinical evidence—through a systematic scientific literature review. Manufacturers must then appraise relevant clinical data and analyze those data to reach conclusions regarding device safety and clinical performance.
The MDR also places the literature-review methodology squarely within the scope of Notified Body review. Under the MDR’s conformity-assessment requirements, the Notified Body’s assessment of the clinical evaluation covers, among other elements, the planning of the clinical evaluation, the methodology used for the literature search, relevant documentation from the literature search, the clinical evaluation report, and the manufacturer’s resulting conclusions.
The IVDR is similarly explicit about methodological rigor. Article 56 requires performance evaluation to follow a defined and methodologically sound procedure. Annex XIII further requires manufacturers to identify relevant data through a systematic scientific literature review, appraise all relevant data, and
document the literature-search methodology, protocol, and literature-search report within the performance evaluation documentation.
The regulatory message is clear: a literature review supporting European clinical evidence cannot simply be a collection of papers selected by an expert. The process itself must withstand scrutiny.
So, Is Dual Screening an Explicit MDR or IVDR Requirement?
No—and this distinction matters.
There is no sentence in the MDR or IVDR that says: “Two independent reviewers shall screen all literature search results.” Describing dual-reviewer screening as an explicit legal requirement would therefore overstate the regulations.
Instead, the requirement develops from the combination of two factors: (1) MDR and IVDR require systematic, methodologically sound clinical evidence processes; and (2) established systematic-review methodologies recognize independent review as an important safeguard against errors and selection bias.
Dual-reviewer screening is therefore best understood as a methodological control that helps a manufacturer demonstrate the rigor expected of a systematic review. And that becomes particularly important when the SLR will be scrutinized during conformity assessment.
What Notified Bodies Are Looking For
Notified Bodies are not simply reviewing the final list of included publications. They need to understand how the manufacturer arrived at that evidence base.
BSI’s Best Practice Guidance for Clinical Evaluation Submission, written from a Notified Body perspective, states that literature-search protocols are expected to be comprehensive and should make an effort to retrieve all available data so that selection bias is minimized. BSI also emphasizes that the search should be documented sufficiently for an assessor to replicate it.
TÜV SÜD likewise emphasizes the need for an appropriate and transparent literature search and appraisal. Its guidance on clinical evaluation notes that multiple databases should be used and that both favorable and unfavorable data must be included.
These expectations point toward the same underlying principles: completeness, transparency, reproducibility, objectivity, and control of selection bias.
The screening process matters because even a perfectly constructed database search can produce a biased evidence base if relevant studies are incorrectly excluded afterward.
Why One Reviewer Can Become a Methodological Weak Point
Literature screening requires judgment. A reviewer may need to decide whether a publication evaluates the correct patient population, addresses the intended purpose of the device, contains relevant safety or performance outcomes, represents the appropriate device or technology, meets predefined study-design
requirements, contains sufficient clinical information, is relevant to the state of the art, or should be excluded for one of several protocol-defined reasons.
Even with well-written eligibility criteria, some records will fall into gray areas. With a single reviewer, one person’s interpretation becomes the only control between a potentially relevant publication and exclusion from the evidence base.
That creates two related risks: error and bias. A relevant paper can be excluded inadvertently. More subtly, ambiguous publications may be handled inconsistently based on one reviewer’s interpretation of the protocol.
Dual-reviewer screening creates an additional methodological safeguard.
What Established Systematic-Review Methodology Says
The expectation for independent review is not unique to medical-device regulation. It comes from the broader science of systematic reviews.
The Cochrane Handbook for Systematic Reviews of Interventions states that at least two people working independently should determine whether studies meet eligibility criteria. Cochrane identifies duplicate study selection as a means of reducing both errors and the possibility that selection decisions are influenced by one person’s biases. For final full-text eligibility decisions, Cochrane treats independent assessment by two people as essential within its intervention-review methodology.
PRISMA 2020 reinforces the importance of transparency around this process. Its expanded checklist asks systematic-review authors to report how many reviewers screened each record and full-text report, whether reviewers worked independently, and how disagreements were resolved.
PRISMA does not itself create an MDR or IVDR legal obligation. But it provides a widely recognized framework for demonstrating that a systematic review was performed transparently. When an SLR is supporting a regulated clinical evaluation, those methodological controls become highly relevant.
Dual Review Creates an Audit Trail, Not Just a Second Opinion
The value of dual screening extends beyond having “two sets of eyes.” A properly designed dual-review process produces a defensible audit trail.
For each record, the literature-review system can document: Reviewer 1 decision → Reviewer 2 decision → Conflict → Resolution → Final decision.
This makes it possible to demonstrate exactly how the evidence base was constructed. For example, imagine that Reviewer 1 excludes a publication because the population appears outside the device’s intended use. Reviewer 2 believes a relevant subgroup within the study meets the predefined eligibility criteria. Instead of that publication disappearing from the evidence base based on a single judgment, the disagreement is identified and adjudicated according to the review protocol.
The process does not guarantee that every decision will be perfect. It does, however, make the decision-making process more controlled, transparent, and reproducible. Those are precisely the characteristics that strengthen an SLR under regulatory scrutiny.
Dual Screening Is Particularly Important at Full-Text Review
Organizations sometimes use different levels of review depending on the size and purpose of the literature search. Title and abstract screening may involve hundreds or thousands of records, whereas full-text screening generally involves a much smaller subset.
The consequences of an exclusion also increase at full-text review. Once a study has reached full-text assessment, it has already been identified as potentially relevant. Excluding it may directly affect the body of clinical evidence, reported safety-event rates, performance benchmarks, the state-of-the-art analysis, equivalence or similar-device discussions, benefit-risk conclusions, and identified clinical evidence gaps.
This is consistent with Cochrane methodology, which distinguishes between initial title/abstract screening and final eligibility decisions. Although Cochrane allows more flexibility at the initial screening stage in some review types, it calls for at least two independent reviewers for final study-inclusion decisions.
For a high-stakes regulatory SLR, manufacturers should therefore consider whether the efficiency gained through single-reviewer screening is worth the additional methodological risk.
The IVDR Makes Methodological Defensibility Especially Important
For IVD manufacturers, literature review can play a particularly important role in demonstrating scientific validity.
IVDR Annex XIII requires the manufacturer to identify relevant data through a systematic scientific literature review and to appraise the suitability of those data. Scientific validity may be demonstrated using sources including peer-reviewed scientific literature, expert consensus, proof-of-concept studies, and clinical performance studies.
The Performance Evaluation Report must also document the justification for the approach used to gather clinical evidence, as well as the literature-search methodology, protocol, and report.
That means the screening process is not merely an internal literature-management exercise. It is part of the methodological chain connecting the search results to the manufacturer’s clinical evidence conclusions.
If a pivotal publication supporting—or challenging—the scientific validity of an analyte-marker relationship is excluded incorrectly, the consequences can flow directly into the SVR and PER.
“We Had an Expert Do the Screening” Is Not the Same as Independent Review
A common misconception is that dual review is unnecessary when screening is performed by a highly experienced clinician, scientist, or medical writer.
Expertise is unquestionably important. But expertise and methodological control solve different problems.
Subject-matter expertise helps a reviewer correctly interpret the literature. Independent review helps control the risk that one person’s interpretation determines the evidence base.
In fact, highly specialized reviews may contain particularly difficult eligibility decisions. Device nomenclature changes. Procedures evolve. Older publications may describe technologies differently. Similar and
equivalent devices may not be named consistently. Safety events may be reported using terminology that differs from contemporary usage.
An expert reviewer is valuable. Two appropriately qualified reviewers working within a predefined protocol provide both expertise and an additional quality control.
What a Regulatory-Grade Dual-Reviewer Process Should Look Like
- Predefined inclusion and exclusion criteria: Eligibility criteria should be established in the literature-review protocol before screening begins and should align with the clinical or performance evaluation question.
- Reviewer training and calibration: Reviewers should apply the criteria to a sample of records before full screening begins. Differences in interpretation can then be discussed and the criteria clarified where appropriate.
- Independent screening: Each reviewer evaluates the record without being influenced by the other reviewer’s initial decision.
- Documented conflict resolution: Disagreements should be resolved using a predefined process—for example, consensus discussion or adjudication by a third qualified reviewer.
- Standardized exclusion reasons: At full-text screening, exclusion reasons should be predefined and consistently applied.
- Traceability: The review platform or documentation system should preserve individual reviewer decisions, conflicts, adjudications, and final dispositions.
- PRISMA documentation: The number of records identified, screened, excluded, assessed at full text, and ultimately included should reconcile with the review’s PRISMA flow diagram.
Together, these controls make it possible to reconstruct how the final evidence set was generated.
What Could a Notified Body Question?
Consider an SLR supporting a CER in which 1,800 records were identified, 95 proceeded to full-text review, and 24 were ultimately included.
A reviewer examining the methodology may reasonably want to understand who performed screening, whether eligibility criteria were defined before screening, whether decisions were performed independently, how disagreements were handled, why the other 71 full-text publications were excluded, whether those exclusions can be traced to predefined criteria, how the manufacturer ensured unfavorable evidence was not inadvertently excluded, and whether the review process can be reproduced or independently verified.
The MDR expressly requires Notified Bodies to assess literature-search methodology and the relevant documentation supporting the search. BSI’s Notified Body guidance similarly emphasizes comprehensiveness, minimizing selection bias, and providing sufficient documentation for the assessor to replicate literature searches.
Dual-reviewer screening gives manufacturers a stronger answer to many of these questions.
The Real Question Is Not “Where Does MDR Say Two Reviewers?”
This is perhaps the most important point.
Regulatory teams sometimes approach SLR methodology by asking: “Show me the sentence in MDR that requires two reviewers.” There isn’t one.
But that may be the wrong question. A better question is: “Can we demonstrate that our process for selecting the literature was sufficiently systematic, objective, reproducible, and protected against avoidable selection bias?”
MDR requires systematic identification and appraisal of relevant clinical evidence. IVDR requires a defined and methodologically sound performance-evaluation process and systematic scientific literature review. Notified Bodies assess the methodology and documentation supporting those processes.
Meanwhile, established systematic-review methodologies identify independent assessment as an important safeguard against erroneous or biased study-selection decisions. Taken together, these expectations make dual-reviewer screening a highly defensible approach for regulatory-grade SLRs.
Building the SLR for the Reviewer Who Will Eventually Examine It
A strong regulatory SLR should be designed with its eventual reviewer in mind.
That means the manufacturer should be able to demonstrate not only what evidence was found, but also how the search was designed, how eligibility criteria were established, how records were screened, how ambiguous studies were handled, why individual full-text publications were excluded, how favorable and unfavorable evidence was captured, how the included literature was appraised, and how the evidence ultimately informed the clinical or performance evaluation.
Dual-reviewer screening strengthens one of the most judgment-dependent stages of that chain. It converts study selection from an individual judgment into a controlled, documented process.
The Bottom Line
Dual-reviewer screening is not an explicitly stated MDR or IVDR requirement. It is something arguably more important in practice: a well-established methodological control that helps manufacturers demonstrate that their systematic literature review is rigorous enough to withstand regulatory scrutiny.
The MDR requires systematic identification, appraisal, and analysis of clinical evidence and specifically places literature-search methodology within the scope of Notified Body assessment. The IVDR similarly requires systematic literature review within a defined and methodologically sound performance-evaluation process.
BSI emphasizes comprehensive, reproducible searches designed to minimize selection bias, while TÜV SÜD highlights transparent literature searching and the inclusion of both favorable and unfavorable evidence.
Established systematic-review frameworks such as Cochrane and PRISMA further reinforce independent screening and transparent reporting of reviewer decisions.
For medical device and IVD manufacturers, the practical implication is straightforward: if the literature review will support a regulatory conclusion, the study-selection process should be designed to withstand regulatory review. Dual-reviewer screening helps make that possible.
How Criterion Edge Supports Regulatory-Grade Literature Reviews
Criterion Edge provides systematic literature review services for medical devices, IVDs, and pharmaceutical products, including literature search and screening, PRISMA flow diagrams, evidence weighting and appraisal, data extraction, literature-review reports, and post-market literature surveillance.
Our approach is designed around the principle that an SLR supporting a regulatory submission must be more than comprehensive—it must be structured, transparent, reproducible, and defensible.
Whether literature is supporting an EU MDR Clinical Evaluation Report, an IVDR Performance Evaluation Report or Scientific Validity Report, state-of-the-art assessment, PMCF/PMPF activity, or another regulatory deliverable, establishing a rigorous methodology at the beginning of the review can help reduce questions later in the review process.
Need support developing or executing a systematic literature review for your medical device or IVD? Contact Criterion Edge to discuss a regulatory-ready literature review strategy.
References
- Regulation (EU) 2017/745 (EU MDR), Article 61 and Annex XIV, Clinical Evaluation and Post-Market Clinical Follow-Up. https://eur-lex.europa.eu/eli/reg/2017/745/2017-05-05
- Regulation (EU) 2017/746 (EU IVDR), Article 56 and Annex XIII, Performance Evaluation, Performance Studies and Post-Market Performance Follow-Up. https://eur-lex.europa.eu/eli/reg/2017/746/2024-07-09/eng
- European Commission, Medical Device Coordination Group (MDCG), MDCG 2020-13: Clinical Evaluation Assessment Report Template. https://health.ec.europa.eu/medical-devices-sector/new-regulations/guidance-mdcg-endorsed-documents-and-other-guidance_en
- BSI, Best Practice Guidance for Clinical Evaluation Submission. https://www.bsigroup.com/globalassets/meddev/localfiles/en-gb/webinars/bsi-md-clinical-masterclass-best-practice-guidance-uk-en.pdf
- TÜV SÜD, Clinical Evaluation of Medical Devices. https://www.tuvsud.com/en/industries/medical-devices/medical-device-regulation/clinical-evaluation-of-medical-devices
- Cochrane, Cochrane Handbook for Systematic Reviews of Interventions, Chapter 4: Searching for and Selecting Studies. https://www.cochrane.org/authors/handbooks-and-manuals/handbook/current/chapter-04
- PRISMA, PRISMA 2020 Expanded Checklist, Item 8: Selection Process. https://www.prisma-statement.org/s/PRISMA_2020_expanded_checklist-yc78.pdf
