Validation Data

Your medical device might be beautifully engineered. It might solve a genuine clinical problem. But without rigorous verification and validation of data, it is just a prototype with potential.

Here’s a statistic that should give every innovator pause: nearly 10% of warning letters from the MHRA over the past five years have claimed inadequately defined and executed validation processes. Not faulty devices. Not safety failures. These are documentation failures, with companies that built something that worked, but could not prove it on paper. Verification and validation are the twin pillars of regulatory credibility. Verification confirms you built the device correctly. Validation confirms you built the correct device. The distinction matters enormously, and mistaking the two costs companies time, money, and market access.

Understanding the difference

Think of it this way. Verification asks whether your design outputs match your design inputs. Did you build what you said you would build? Validation asks whether those outputs meet actual user needs in real-world conditions. Does it work for the people who will use it? Both questions demand objective evidence. Both require systematic documentation. And both need to be completed before any regulator will take your device seriously.

The MHRA defines verification as confirmation through examination and provision of objective evidence that specified requirements have been fulfilled. Validation, meanwhile, establishes that device specifications conform to user needs and intended use. The difference is not semantic. It is structural.

New AI medical devices have the potential to increase the accuracy of healthcare decisions, save time and improve efficiency, leading to better outcomes for the NHS and patients across all healthcare settings. But we need to be confident that AI-powered medical devices introduced into the NHS are safe and stay safe, and perform as intended through their lifetime of use.

_{-Dr Laura Squire. MedTech Regulatory Reform Lead. MHRA}

Through their lifetime of use is the heart of validation. You’re not just proving your device works in the laboratory. You’re proving it works in the clinic, in the home, and in the hands of the people who depend on it.

Building traceability

The concept of traceability is fundamental. Every verification and validation activity must link directly back to specific design inputs. Every test result needs to connect to a requirement.

Imagine a mobile hospital ventilator. A user need might be: the device must run on battery power during patient transport. This translates into a design input: the ventilator shall run on battery power at maximum settings for a minimum of 90 minutes. Therefore, the design output becomes: a lithium-ion battery pack rated for at least 100 amp hours.

This traceability matrix is not bureaucratic box-ticking. It’s your proof that every decision was deliberate, every requirement was addressed, and every claim can be substantiated.

Statistical rigour

Sample sizes matter. Analysis methods matter. And both require justification. Regulators will ask why you tested a particular number of units. They will want to know why you chose specific statistical methods. They will scrutinise whether your acceptance criteria were defined before testing began (not retrofitted afterwards to match convenient results).

The principle is straightforward: your validation approach should be designed to demonstrate what you claim about your device. If you claim a 95% success rate, your testing methodology must be capable of detecting whether that rate is achieved. If you claim durability over five years, your accelerated ageing studies must provide valid extrapolations.

A future-fit legal framework must be efficient, predictable, and innovation-friendly. One that ensures patient safety, while enabling timely access to technologies across Europe.

_{Petra Zoellner. Director of Regulatory Affairs. MedTech Europe}

That balance between patient safety and timely access hinges on evidence quality. Strong validation data speeds approval. Weak validation data triggers questions, requests for additional information, and delays that can extend timelines by months or even years.

Presentation matters

Even excellent data can fail if poorly presented. Reviewers are evaluating dozens of submissions. They are looking for clear organisation, logical flow, and easy navigation. Group related activities together. Use consistent terminology. Cross-reference supporting documents clearly. Provide executive summaries for complex test reports. Every predefined acceptance criterion should be stated explicitly, with corresponding results showing whether it was met.

Meeting acceptance criteria

Before any testing begins, define what success looks like. These criteria need to flow directly from your design inputs and, ultimately, from the intended use of your device. Vague criteria like acceptable performance will trigger reviewer questions. Precise criteria, such as burst pressure exceeding 150 psi with zero failures across 30 units, will not. Post-hoc criteria changes are red flags. If you revise criteria during testing, document the rationale thoroughly. Repeated revisions only undermine confidence in your development process.

The link to performance claims

Here is the critical connection many innovators miss: your validation data must directly support the performance claims in your labelling and marketing materials. If you claim your diagnostic achieves 99% sensitivity, your validation testing must demonstrate that sensitivity. If you claim your implant withstands one million loading cycles, your fatigue testing must prove it. Every claim is a promise, and every promise requires evidence. Regulators will compare your claims against your data. Discrepancies between what you say and what you prove are among the most common reasons for deficiency letters.

VP Med Ventures Workshop

Verification and validation are where engineering meets regulation. They transform your device from something that works in your laboratory to something that can be trusted in clinical practice. Together with VP MED Ventures, we can help you prepare for rigorous verification and validation of data. Your data tells the story of your device (let’s make it a story worth telling).

Waypoint checklist

A considered reminder for your validation data:

• Clearly link V&V results back to the specific design inputs they address.
• Ensure all planned verification and validation activities are fully documented.
• Present the data in a clear, organised, and easily reviewable format.
• Include statistical justification for sample sizes and analysis methods used.
• Demonstrate that all predefined acceptance criteria for V&V activities were met.
• Ensure a clear and direct link between your verification and validation data and the specific performance claims and intended use stated for your device.