Published October 16, 2026
Real-world data can help manufacturers understand how medical devices and diagnostic tests perform in routine care. Turning those data into evidence that supports a regulatory decision requires a clear question, an appropriate study design, and confidence in the underlying information.
On October 9, 2026, the Medicines and Healthcare products Regulatory Agency (MHRA) published its Real-World Evidence: knowledge and guidance hub. The resource brings together guidance, publications, and support routes relevant to the use of real-world data and evidence for medicines and medical devices in the UK. For scientific and regulatory teams, it provides a central starting point for identifying resources that may inform evidence-generation plans.
A new resource with practical planning value
The hub consolidates existing materials, including guidance on data quality, randomized trials using real-world data, and a draft guideline on external control arms. Its publication does not introduce a new set of evidence requirements. Teams should check the scope and status of each linked document, particularly because several resources focus specifically on medicines. The external-control-arm document, for example, remains identified as draft guidance.
For device and IVD manufacturers, the practical opportunity is to use this collection to strengthen evidence planning: identify relevant principles, assess their applicability, and document how they inform a particular study or evaluation.
Start with the decision the evidence must support
The MHRA describes real-world data as information about patient health or healthcare delivery collected outside a clinical study. Sources can include electronic health records, disease registries, administrative records, patient-reported outcomes, and measurements collected through wearables. Analysis of these data may generate real-world evidence.
A practical application for manufacturers is to define the evidence question before selecting a dataset. Is the objective to characterize long-term safety, assess performance in an underrepresented population, understand routine-use outcomes, or investigate an uncertainty identified during an earlier evaluation?
That choice determines which variables, follow-up periods, comparators, and populations are needed. A large dataset may still be unsuitable if it cannot answer the question attached to the intended claim.
Data quality needs an explicit assessment
The MHRA’s underlying guidance emphasizes understanding data provenance, accuracy, validity, reliability, and any processing or transformation. It also addresses population representativeness, completeness of study variables, linkage methods, and documented data-management processes.
Using routine-care data does not remove the need to manage bias or apply sound study-design principles.
For a clinical evaluation report (CER), a useful practical approach is to document how closely the dataset reflects the device, users, clinical setting, and outcomes under evaluation. For an IVD performance evaluation report (PER), teams should also examine whether the records identify the assay version, specimen type, testing threshold, and reference standard needed to interpret performance.
Consider a hypothetical retrospective dataset containing thousands of diagnostic results. If confirmatory testing occurred mainly after positive results, estimating sensitivity and specificity across the full intended-use population could be problematic. The dataset might support some questions while leaving others unresolved. Making that distinction explicit strengthens the evaluation.
Connect real-world evidence with the wider evidence package
For teams preparing PER, CER, and state-of-the-art assessments, a practical recommendation is to assign each evidence source a defined role. Literature may establish clinical context and current practice; analytical studies may characterize measurement performance; routine-care data may help address particular uncertainties about use and outcomes.
An evidence map can make those relationships visible. For each question, record the supporting source, its methodological limitations, its applicability to the claimed use, and the remaining gap. This helps teams avoid treating an additional publication or dataset as sufficient evidence simply because it increases the volume of available information.
Use the appropriate route for regulatory discussion
The hub also points developers toward MHRA support. Its device regulatory-advice service covers medical devices and IVDs where applying existing guidance is not straightforward. The MHRA encourages applicants to review published guidance first and distinguishes this service from clinical-investigation advice and approved-body dialogue.
For teams working across UK and EU markets, the hub can inform planning, while applicability to each jurisdiction still needs separate assessment. It should not be treated as evidence of automatic acceptance under the EU MDR or IVDR.
Before the next evidence-plan update, manufacturers can use the hub to review three connected questions: what uncertainty needs resolving, whether the available data can resolve it, and what additional work is needed. That assessment turns access to real-world data into a purposeful evidence-generation strategy.
Source: https://www.gov.uk/government/publications/mhra-real-world-evidence-knowledge-and-guidance
