Post-market surveillance (PMS) sounds like something you do after you launch.
But under EU MDR, your post-market system is part of your pre-market credibility. Notified Bodies want to see that you’ve planned how you’ll monitor safety and performance once real users touch your device.
The good news: you don’t need a “big company” PMS machine. You need a startup-sized plan that’s proportionate.
If you’re still getting oriented in MDR, start here:
EU MDR for Startups: Where Do I Start? (Practical Roadmap)
And if you’re building your plan like a roadmap, this is the companion piece:
Map Your MDR Pathway Like a Product Roadmap
PMS in plain English
PMS is your system to:
– collect feedback and data from the field
– detect trends, complaints, and safety signals
– confirm your device continues to perform as intended
– feed improvements into risk management, clinical evaluation, and labeling
Think of it as: “How will we stay in control once we’re live?”
Why PMS matters before you sell a single unit
If you treat PMS as a later problem, you usually end up with:
– a weak (or generic) PMS plan that doesn’t match your device
– PMCF activities that are either unrealistic or missing
– gaps between clinical claims and how you’ll confirm them in the real world
– avoidable findings during conformity assessment
Planning early gives you:
– a cleaner clinical evaluation story
– fewer surprises during audit
– a smoother transition from pilot → launch → scale
The three building blocks: PMS, PMCF, and vigilance
1) PMS plan (your operating system)
Your PMS plan describes:
– what data you’ll collect
– where it comes from
– how often you review it
– who is responsible
– what triggers action (CAPA, labeling updates, risk updates)
2) PMCF plan (your “clinical follow-up” logic)
PMCF is typically a subset of PMS focused on confirming clinical performance and safety in real-world use.
Important: PMCF is not automatically a clinical study. It can include surveys, registries, follow-up questionnaires, RWE collection, or targeted clinical investigations—depending on risk.
3) Vigilance (your “serious events” workflow)
This is how you handle:
– serious incidents
– field safety corrective actions
– reportability decisions
– timelines and documentation
Even pre-market, you should know what you’ll do when something goes wrong.
A startup-sized PMS system you can set up now
Here’s the lean version that works for most early-stage teams.
Step 1: Define your data sources (before you have customers)
Common early sources:
– usability studies and formative/summative testing feedback
– pilot users / clinical partners
– customer support tickets (even if it’s just email)
– installation logs / training records
– software logs (if SaMD)
– complaints + nonconformities
Definition of done:
You can list your data sources and show how they’ll be captured.
Step 2: Create a simple complaint handling workflow
Minimum viable workflow:
– intake form (what happened, device version, user, environment)
– triage (complaint vs feedback)
– investigation steps
– decision: reportable? CAPA needed?
– closure + documentation
Definition of done:
You can demonstrate a repeatable process, not ad-hoc Slack messages.
Step 3: Set your review cadence (and keep it realistic)
Pick a cadence you can actually maintain:
– monthly review (typical for early stage)
– quarterly trend review (once volume grows)
Definition of done:
You have calendar-based reviews with documented outputs.
Step 4: Connect PMS to risk management and clinical evaluation
PMS isn’t a standalone folder.
Your PMS outputs should feed:
– risk management updates
– clinical evaluation updates
– IFU/labeling updates
– CAPA
If you want the clinical side framed clearly, start here:
Step 5: Draft PMCF logic that matches your claims
Ask:
– Which claims are most important to confirm post-market?
– What’s the simplest way to collect that evidence?
– What would trigger a deeper PMCF activity?
Definition of done:
You can explain your PMCF approach in plain language and it feels proportionate.
Common mistakes (that create findings)
– Copy-pasting a generic PMS plan that doesn’t match the device
– Treating PMCF as “optional” without justification
– No clear complaint handling process
– No link between PMS outputs and risk/clinical updates
– Review cadence that’s unrealistic (so it never happens)
Quick checklist: “Are we PMS-ready?”
– We know our PMS data sources
– We have complaint intake + triage
– We have a review cadence and owners
– We know what triggers CAPA and reporting decisions
– Our PMS outputs feed risk and clinical evaluation
– Our PMCF logic matches our claims
Want a PMS/PMCF setup that stays startup-sized?
If you want a lean PMS + PMCF system that satisfies MDR expectations without turning your startup into a bureaucracy, book a free 60-minute MDR Strategy Call.
Book here: https://calendly.com/niko-mangold-consultants/30min