Medical Affairs as the Evidence Engine in MedTech: Turning Insights into Outcomes

Medical Affairs and Evidence Generation in MedTech

In medical devices, the product is rarely the whole story. Surgical technique, training, theatre workflow, patient selection and postoperative pathway all shape what a patient experiences. Two hospitals implanting the same device can produce meaningfully different outcomes, and the difference often has little to do with the device itself.

That is why I have stopped treating evidence as something we finish at CE marking or FDA Clearance/Approval. Under the EU Medical Device Regulation, post-market clinical follow-up is explicitly a continuous process that feeds back into the clinical evaluation. The FDA has moved in a similar direction on real-world data for device decision-making. Neither framework says the job ends at market entry. Both assume a device will continue generating questions after approval, and that someone must be accountable for turning those questions into learning. In most organisations I have worked with, that accountability is the part that quietly goes missing.

The Role of Medical Affairs in MedTech

I regularly meet teams who are uncertain what Medical Affairs is for. Commercial leaders may expect it to validate messaging, Regulatory colleagues may treat it as a second-line reviewer, and Quality teams may see it as the place for whatever does not fit elsewhere. Each framing pulls Medical Affairs away from where it adds most value.

The working definition I use is narrower. Medical Affairs in medtech has three jobs: credible scientific exchange with clinicians, structured capture and interpretation of the insights these exchanges produce, and support to the evidence strategy that turns those insights into data.

It should not own regulatory submissions, make quality system decisions, or adjudicate vigilance cases. We can contribute clinical context to all three, but ownership blurs lines the regulations expect kept separate.

Medical Affairs Evidence Engine: From Clinical Insights to Evidence

I think of this as an evidence engine with six connected steps: Capture, Adoption Context, Triage, Translation, Generation and Communication. Capture records the signal. Adoption Context explains where and why it arose. Triage decides what needs escalation. Translation turns it into a question. Generation tests it. Communication returns learning to practice. That sequence is what I mean by Medical Affairs acting as an evidence engine.

Capture Clinical Insights from Healthcare Professionals

Field interactions generate a lot of material but little structured information. What has worked for me is recording what the clinician is worried about, the conditions under which it arose, whether it could be a safety topic, and what would move us from uncertainty to clarity. That last question is often skipped. Without it, insights accumulate and nothing happens.

Understand Clinical Adoption Context Without Bias

In orthopaedics, variability almost never lives in the implant. It lives in preoperative planning routines, whether the right instrument trays are available on the day, how responsibilities are split in theatre, and how rehabilitation is organised. I once spent two days at a high-volume centre convinced that a particular reconstruction technique was underperforming because of a component issue. It turned out the surgeon had inherited a planning template from a previous device and had not updated it. The implant was fine. The template was the problem. That experience changed how I capture context. I now write down the workflow before I write down the complaint.

Triage Clinical Insights and Safety Signals

Some things cannot wait. If a field insight hints at a safety signal, it goes into vigilance and quality processes within hours, not weeks, and Medical Affairs is not the function that decides whether something is reportable. We can help interpret the clinical picture, but the call on reportability sits with the people whose job it is to make that call. Crossing that line damages trust in ways that take years to rebuild.

Translate Clinical Insights into Evidence Questions

If I could change one habit in device evidence generation, it would be how we choose endpoints. Too often we pick familiar or easy endpoints rather than outcomes a patient would recognise as meaningful: function, pain, complications, reinterventions, return to normal life and durability. If repeated clinician feedback suggests slower recovery, for example, that may justify adding a patient-reported functional measure or time to mobilisation to PMCF rather than relying only on radiographic endpoints. That is the evidence engine at work: an observation becomes a testable question. The MDCG 2020-7 template for PMCF plans is useful because it forces us to state what we are trying to learn and why before designing anything.

Generate the Right Clinical Evidence

Real-world evidence is useful for questions such as long-term survivorship, rare complications or signal detection. For other questions it is not enough, and Medical Affairs has to be willing to say so. If clinicians repeatedly ask whether outcomes differ in a specific patient group, a retrospective analysis may answer the first question, while the result may justify prospective follow-up. The point is to choose evidence that resolves uncertainty. When a clinical investigation is needed, ISO 14155 sets out the expectations. Medical Affairs does not run clinical operations, but it should engage with protocol design, endpoint selection and interpretation.

Communicate Clinical Evidence Effectively

Good evidence changes nothing if it is communicated badly. I have sat through presentations where clinical data were summarised without the training conditions, learning curve or patient selection criteria the study relied on. The data may have been accurate, but the story was incomplete. Clinicians noticed. Communication is the final step of the evidence engine because it returns learning to practice. When evidence becomes balanced education, clearer patient selection or realistic expectations around technique and training, it can improve consistency in device use. Getting this right over years is how Medical Affairs earns the trust that makes the rest of the function possible.

Medical Affairs as the Evidence Engine for Continuous Learning

Medical device lifecycles are long and sensitive to conditions of use and implantation in ways a pill is not. Regulation increasingly expects continuous learning rather than one-time approval. I see Medical Affairs as the natural home for the evidence engine, provided it stays within clear functional boundaries. Capture the insight, understand the adoption context, triage what matters, translate it into a question, generate the right evidence, and communicate the learning back into practice. The payoff is not more evidence for its own sake. It is care that becomes safer and more consistent because the organisation is learning from what it sees.

References

• Regulation (EU) 2017/745 (MDR), consolidated version 01 January 2026, Annex XIV, Part B, Post-Market Clinical Follow-up. https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX%3A32017R0745
• US Food and Drug Administration. Use of Real-World Evidence to Support Regulatory Decision-Making for Medical Devices. https://www.fda.gov/regulatory-information/search-fda-guidance-documents/use-real-world-evidence-support-regulatory-decision-making-medical-devices
• ISO 14155:2026, Clinical investigation of medical devices for human subjects, Good clinical practice. https://www.iso.org/standard/83968.html
• MDCG 2020-7 Post-market clinical follow-up (PMCF) plan template. https://health.ec.europa.eu/system/files/2020-09/md_mdcg_2020_7_guidance_pmcf_plan_template_en_0.pdf

Disclaimer. The views and opinions expressed in this article are solely those of the author and do not necessarily reflect the official policy or position of Test Labs Limited. The content provided is for informational purposes only and is not intended to constitute legal or professional advice. Test Labs assumes no responsibility for any errors or omissions in the content of this article, nor for any actions taken in reliance thereon.