Literature Reviews for Regulatory Submissions: A Guide to Clinical Evaluation Reports and Safety Reviews
A clinical evaluation report (CER) literature review is a protocol-driven, fully documented search and appraisal of clinical and safety evidence built to survive regulatory inspection. It differs from an academic review by prioritizing traceability, reproducibility, and an audit trail. This guide explains how these reviews work under EU MDR, MEDDEV 2.7/1 Revision 4, and Good Pharmacovigilance Practices.
Dr. Elena Vasquez
July 8, 2026
Key Takeaways
A clinical evaluation report literature review is protocol-driven, reproducible, and built to survive inspection, not a narrative summary.
Under EU MDR, MEDDEV 2.7/1 Revision 4 remains the working reference for how to conduct and document the review.
A strong CER runs two distinct searches: a device-specific search and a separate state-of-the-art review.
Systematic literature reviews underpin the clinical and economic sections of regulatory and reimbursement dossiers.
Pharmacovigilance and PSUR searches are continuous, time-bounded, and judged on whether the documented process was followed.
A pre-specified protocol, verbatim searches, transparent appraisal, and a complete audit trail are the four load-bearing practices.
PRISMA aids transparency where it applies, but a review guarantees no regulatory outcome and makes no regulatory determination.
What a regulatory literature review actually is
A Clinical Evaluation Report (CER) literature review is a pre-specified, fully documented search and appraisal of clinical and safety evidence, structured so that an assessor or auditor can reproduce every step and verify every conclusion. In the regulatory context, the literature review is not a narrative summary of what the authors happen to know; it is a controlled process governed by a written protocol, executed against defined databases, and reported with enough detail that an inspector can rerun the searches and reach the same result. This audit-ready, protocol-driven, traceable character is the single feature that separates a literature review for regulatory submission from the reviews most researchers learned to write in academia.
The distinction matters because the reader is different. An academic review persuades peers; a regulatory review must withstand scrutiny from a Notified Body, a competent authority, or an inspector who assumes nothing and checks everything. That shift in audience changes the standard of evidence, the documentation burden, and the tolerance for undocumented judgment calls. The rest of this guide explains how that plays out across device and drug submissions, and where a PhD methodologist adds rigor that a generalist writer cannot.
How regulatory reviews differ from academic reviews
The gap is procedural, not intellectual. Both types of review synthesize evidence, but they answer to different masters. A regulatory review is engineered around four properties that academic reviews treat as optional.
First, it is protocol-driven. Before any search runs, a written protocol fixes the research question, the eligibility criteria, the databases, the search strings, the appraisal method, and the analysis plan. Deviations are logged and justified rather than quietly absorbed. Second, it is reproducible: search strings are recorded verbatim with database, interface, date, and hit counts, so a third party can rerun them. Third, appraisal is : every included and excluded record has a documented reason, and the weight given to each source follows a stated method rather than the reviewer's preference. Fourth, the whole exercise leaves a complete that ties the final conclusions back to the raw search output.
Frequently Asked Questions
8
It is a pre-specified, fully documented search and appraisal of clinical and safety evidence that supports a device's clinical evaluation under EU MDR. It is protocol-driven, reproducible, and built so an assessor can rerun the searches and verify every conclusion. It typically includes both a device-specific search and a state-of-the-art review of the current standard of care.
The difference is procedural. Both synthesize evidence, but a regulatory review is engineered around a written protocol, verbatim documented searches, transparent appraisal with recorded reasons, and a complete audit trail. Academic reviews persuade peers; regulatory reviews must withstand inspection from a Notified Body, competent authority, or auditor who checks every step.
MEDDEV 2.7/1 Revision 4 expects a clear literature search protocol, objective inclusion and exclusion criteria, a documented appraisal of each item for relevance and quality, and a written justification of how the evidence supports the device. It also asks that data on the subject device be distinguished from data on equivalent or similar devices, with equivalence justified technically, biologically, and clinically.
The state-of-the-art review characterizes the current standard of care, the available alternatives, and the accepted safety and performance benchmarks for the device's intended purpose. It lets an assessor judge whether the device's benefit-risk profile is acceptable relative to what else exists. It needs its own protocol, search strings, and appraisal, separate from the device-specific search.
A Periodic Safety Update Report search is continuous and time-bounded to the reporting interval rather than a single event. Under Good Pharmacovigilance Practices, it must be systematic and documented so terms, databases, dates, and results are inspectable, and it must capture case reports and safety signals, not just formal studies. Inspections examine whether the defined process was followed.
PRISMA is a reporting standard, not a regulatory requirement. It is well suited to the systematic literature review components of a dossier and increasingly to the search sections of clinical evaluation reports, where a flow diagram documents records identified, screened, excluded with reasons, and included. It should be applied where it adds transparency and adapted where a regulatory template already dictates the structure.
No. We do not make regulatory determinations, and no methodologist can promise that a Notified Body or competent authority will reach a particular decision. The goal of a rigorous review is to produce evidence that is complete, honest, traceable, and defensible under inspection. Approval depends on the full submission and the reviewing body, which sits outside any review team's control.
A full systematic literature review is appropriate when the clinical or economic sections of a marketing authorization or health technology assessment submission need a complete, unbiased summary of the evidence, especially when the search will feed an indirect treatment comparison or network meta-analysis. The pre-specified method reduces selection bias and produces a search clean enough to support quantitative synthesis.
Share
Found this useful? Share it with your colleagues.
Need help with your evidence synthesis?
Our methodologists handle narrative synthesis, thematic analysis, umbrella reviews, and all forms of evidence synthesis, following established frameworks.
Complex Synthesis? Our PhD Team Specializes in Every Type.
Systematic reviews, narrative synthesis, umbrella reviews, and rapid reviews. We match the right methodology to your research question and deliver a publication-ready manuscript.
Our promise: Free rework on synthesis or GRADE assessment if reviewers push back.
4.9 / 5Quote in minutesPRISMA 2020 + GRADE certaintyPhD methodologistNDA available on request
PhD in Clinical Epidemiology. Twelve years in evidence synthesis across cardiology and infectious disease. Personally signs off on every protocol before screening begins.
Complex Synthesis? Our PhD Team Specializes in Every Type.
Systematic reviews, narrative synthesis, umbrella reviews, and rapid reviews. We match the right methodology to your research question and deliver a publication-ready manuscript.
Quote in minutes. Pay only after you approve your quote. Unlimited revisions until your reviewers are satisfied. NDA available on request.
transparent
audit trail
These are the same instincts that drive a rigorous systematic literature review (SLR) in health research, which is why teams that build defensible reviews for publication tend to build defensible reviews for regulators. If you want the underlying method in depth, our overview of how a rigorous systematic review is planned and executed walks through protocol registration, screening, and reporting in the academic frame that regulatory work borrows from.
Clinical Evaluation Reports under EU MDR
Under the EU Medical Device Regulation (EU MDR), formally Regulation 2017/745, every device requires a clinical evaluation that demonstrates conformity with the relevant general safety and performance requirements. The clinical evaluation report is the document that presents this evaluation, and for most devices the literature route is a central pillar of the evidence, either alongside or in place of a dedicated clinical investigation. The literature review inside a CER is therefore not a background section; it is a primary source of the clinical data on which the conformity argument rests.
The methodological expectations trace back to MEDDEV 2.7/1 Revision 4, the guidance that predates EU MDR but remains the working reference for how to conduct and document a clinical evaluation. MEDDEV 2.7/1 Revision 4 asks for a clear literature search protocol, objective inclusion and exclusion criteria, a documented appraisal of each item for relevance and quality, and a written justification of how the body of evidence supports the device. It also expects the review to distinguish data on the device in question from data on equivalent or similar devices, and to handle claims of equivalence with a documented technical, biological, and clinical rationale rather than assertion.
The two searches inside a CER
A well-built CER literature review usually runs two distinct searches with two distinct purposes, and conflating them is a common finding in Notified Body deficiency letters.
The first is the device-specific search, which gathers clinical data on the subject device and any claimed equivalent device to establish safety and performance. The second is the state-of-the-art (SOTA) review, which characterizes the current standard of care, the available alternatives, and the accepted safety and performance benchmarks for the device's intended purpose and clinical condition. The SOTA review is what lets an assessor judge whether the device's benefit-risk profile is acceptable relative to what else exists, and it is frequently the weakest part of submissions written without a formal search method. A defensible state-of-the-art literature review needs its own protocol, its own search strings, and its own appraisal, not a few citations bolted onto the device search.
Both searches feed the appraisal of clinical data, in which each retrieved record is scored for its contribution to the safety and performance questions. The appraisal method should be pre-specified and applied consistently, and the weighting of higher-quality evidence over weaker sources should follow a stated rule. Because search design is where most reviews succeed or fail, our note on building a reproducible search strategy covers database selection, controlled vocabulary, and hit-count documentation in more operational detail.
Systematic literature reviews for regulatory dossiers
Beyond the device world, a formal systematic literature review (SLR) often underpins the clinical and economic sections of a regulatory or reimbursement dossier. Marketing authorization applications, health technology assessment submissions, and value dossiers all lean on SLRs to summarize efficacy, safety, and comparative evidence in a way that assessors accept as complete and unbiased.
The value of the SLR method here is the same as in health research: a systematic, pre-specified search reduces the risk that convenient studies are cherry-picked and inconvenient ones quietly dropped. For dossiers, the SLR frequently sits alongside an indirect treatment comparison or a network meta-analysis, so the search and screening have to be clean enough to feed a quantitative synthesis without introducing selection bias. Teams preparing this kind of evidence base often combine the review with formal evidence synthesis and meta-analysis so that the qualitative and quantitative layers share a single protocol.
When the endpoint is reimbursement rather than approval, the same SLR usually has to satisfy payer bodies whose evidence requirements differ from a regulator's. Our summary of health technology assessment and market access evidence explains how the same underlying search can be shaped to answer both the regulatory and the payer question without running two separate reviews.
Safety and pharmacovigilance literature reviews
On the drug side, literature review is a continuous obligation rather than a one-time submission task. Under Good Pharmacovigilance Practices (GVP), marketing authorization holders are expected to monitor the scientific literature for information relevant to the safety of their products, and to feed relevant findings into signal detection, risk management, and periodic reporting. This is pharmacovigilance literature searching, and it runs on a defined periodicity with its own documentation standards.
The most visible output is the Periodic Safety Update Report (PSUR), which presents an integrated benefit-risk evaluation over a defined interval. A PSUR literature search has to be systematic, time-bounded to the reporting period, and documented so that the search terms, databases, dates, and results are all inspectable. The searching is not limited to formal studies; it must capture case reports, safety signals, and relevant secondary literature that could bear on the product's evolving safety profile. Because GVP inspections examine whether the process was followed, not just whether the conclusion was reasonable, the search methodology and its records carry as much weight as the safety findings themselves.
A further practical point is periodicity. Because pharmacovigilance searching repeats on a schedule, the method has to be stable enough to run identically each interval, so that a later inspector can see a consistent process rather than an ad hoc search that changed shape every cycle. Version-controlling the search protocol, and logging any deliberate change to databases or terms, is what keeps that consistency inspectable over the life of a product.
The recurring theme across CERs, SLRs, and PSURs is that the search is a regulated artifact. It has to be planned before it runs, executed against a fixed method, and preserved in a form an inspector can audit years later. That preservation obligation is easy to underestimate, since the records may need to be produced long after the people who built the search have moved on.
Why the protocol, the search, and the audit trail matter
Every expectation above collapses into one principle: an inspector must be able to reconstruct what you did and confirm you did not steer the result. That principle drives four practices worth stating plainly.
A pre-specified protocol is the anchor. Fixing the question, eligibility criteria, and analysis plan before seeing the results is what stops the review from drifting toward a desired conclusion, and it is the first document an assessor asks for. Reproducible, documented searches are the evidence that the protocol was followed: verbatim strings, named databases and interfaces, exact dates, and hit counts, so the search can be rerun and matched. Transparent appraisal ensures that inclusion, exclusion, and weighting decisions are visible and defensible rather than resting on undisclosed judgment. And a complete audit trail links the final benefit-risk conclusion back through the appraisal to the raw retrieval, so nothing in the report is unsupported.
Where the writing itself is scrutinized, precise regulatory language matters as much as method, which is why the drafting of these documents belongs with specialists; our overview of regulatory and scientific medical writing describes how the narrative is assembled so that it reflects the evidence without overreaching. Throughout, a credible review states its limitations openly, and it does not guarantee a regulatory outcome. We do not make regulatory determinations, and no methodologist can promise that a Notified Body or competent authority will reach a particular decision; the goal is a review that is complete, honest, and defensible, not one that is dressed to persuade.
Reporting to PRISMA where it applies
For the SLR components of a dossier, and increasingly for the literature-search sections of CERs, the PRISMA reporting framework is the natural way to document flow and transparency. A PRISMA flow diagram captures the number of records identified, screened, excluded with reasons, and included, giving an assessor an immediate, standardized view of how the final evidence set was reached. PRISMA is a reporting standard rather than a regulatory requirement, so it should be applied where it clarifies the search and adapted where a regulatory template already dictates structure. Used well, it turns the search into a picture an inspector can read in seconds, which is exactly the transparency that regulatory review is built to deliver.
Bringing these threads together, a regulatory literature review is a discipline of documentation as much as of synthesis. The intellectual work of appraising evidence is shared with academic review, but the surrounding apparatus, the protocol, the recorded searches, the transparent appraisal, and the audit trail, is what makes the review fit for a submission. If you are scoping a CER, an SLR for a dossier, or a pharmacovigilance search and want to discuss how the method would apply to your product, you can request a scoped quote for your review, and you can read more about our pharmaceutical and regulatory review service for how this work is planned and delivered.
A rigorous, doctoral-level guide to conducting a meta-analysis: defining the question, extracting effect sizes and their variances, choosing a between-study variance estimator, pooling, and diagnosing heterogeneity and bias.
To hire a statistician, you brief a qualified expert on your research question and data, and they choose the right analysis, run it in your required software, and explain every result in writing. Research Gold pairs you with a PhD statistician who handles study design, data cleaning, the analysis itself, and a defensible written interpretation, fully online, by the project or the hour, across the health, social, and physical sciences.
Meta-analysis in psychology pools the effect sizes from many studies into one reliable result. Learn the definition, real examples, and how researchers run one.
