Digital Health & Data
What is real-world evidence (RWE)?
In short: Real-world evidence (RWE) is clinical evidence about a product's use and effects, generated from real-world data — electronic health records, registries and routine care — rather than only from randomised trials. It shows how something performs in everyday practice.
RWD vs RWE
Real-world data (RWD) is the raw data from routine care; real-world evidence (RWE) is the analysed insight derived from it. Good RWE depends on data quality, a clear question and sound method.
Why it matters
RWE complements trials by showing performance across real populations and settings. It supports market access, reimbursement and buyer confidence, and can surface safety and effectiveness signals over time.
Generating credible RWE
It starts with a fit-for-purpose registry or data model, an outcomes dictionary, appropriate information governance, and a pathway to analysis and publication. Done well, it becomes reusable evidence rather than a one-off report.
Meds Global Health designs registries and evidence pathways. See Registry, Outcomes & RWE, part of Healthcare Data & Analytics.
Common pitfalls
Where real-world evidence goes wrong
RWE is only as trustworthy as its method. A few failure modes recur, and each can quietly invalidate an otherwise impressive result:
- Confounding. The groups you compare differ in ways — age, severity, comorbidity — that explain the outcome rather than the intervention. Without adjustment, an apparent benefit may be an artefact of who received the treatment.
- Selection bias. If the data only captures certain patients (for example, those who completed a pathway), the conclusion will not generalise to the whole population.
- Data quality gaps. Missing fields, inconsistent coding and free-text that never reaches analysis erode reliability. Routine-care data is collected for care, not research, so it must be assessed before use.
- Analysis without a question. Searching the data for any significant finding invites false positives. A pre-specified question, population and outcome set is the single biggest safeguard.
The practical defence is design discipline: decide what you are measuring and in whom before you open the dataset, and document the method so the result can be defended and reproduced.
In practice
How RWE fits the wider evidence picture
Randomised trials and real-world evidence are complementary, not rival. A trial answers can this work under controlled conditions?; RWE answers does it work, safely and affordably, in everyday NHS practice?. Regulators, payers and NHS buyers increasingly want both — the controlled efficacy signal and the routine-care effectiveness signal.
That makes a well-governed registry a strategic asset. Set up once with a clear outcomes dictionary and information governance, it produces evidence repeatedly: for safety monitoring, for value and reimbursement conversations, and for the business case a trust needs before adoption. Linking RWE to your assurance evidence strengthens the whole proposition — see NHS Buyer Readiness and our guide to how the NHS buys digital health. For products using AI, ongoing real-world monitoring also supports the post-deployment assurance described in AI Validation & Clinical Safety.
Answers
Frequently asked questions
What is real-world evidence?
Real-world evidence (RWE) is clinical evidence about how a product is used and how it performs, generated by analysing real-world data (RWD) — such as electronic health records, registries, and routine care data — rather than only randomised trials.
Why does RWE matter?
It shows how a product performs in everyday practice, supports market access and reimbursement conversations, and can demonstrate value to buyers and payers in ways trials alone cannot.
How do you generate it?
With a sound registry or data model, an outcomes dictionary, appropriate governance and a publication pathway. See Registry, Outcomes & RWE.
What are common pitfalls in RWE?
The biggest risks are confounding (the groups being compared differ in ways that explain the result), missing or inconsistent data, selection bias, and analysing without a pre-specified question. Defining the question, population and outcomes before you look at the data is the main defence.
How does RWE relate to randomised trials?
They answer different questions. Trials establish efficacy under controlled conditions; RWE shows effectiveness, safety and value in routine practice across broader populations. Strong evidence packages usually combine both rather than treating them as alternatives.
Can RWE support NHS adoption?
Yes. Demonstrating real-world performance and value strengthens business cases and reassures NHS buyers, alongside the assurance evidence in NHS Buyer Readiness.
Need real-world evidence?
We design the registry and pathway to generate it.