MDR Annex I §5 is the regulation's ergonomic requirement: devices must be designed to reduce as far as possible the risks linked to the ergonomic features of the device and the environment of intended use, taking into account the technical knowledge, experience, education, training, use environment, and medical and physical condition of intended users. Annex I §22 adds stricter obligations for devices intended for lay persons. Together they are the legal foundation every usability engineering file must satisfy.

By Tibor Zechmeister and Felix Lenhard.

TL;DR

  • Annex I §5 is the core ergonomic general safety and performance requirement and applies to every MDR device regardless of class.
  • Annex I §22 layers on additional obligations for devices intended to be used by lay persons, including patients at home.
  • The regulation demands that use-related risks be reduced "as far as possible" considering the intended users and environment, not merely to an ad-hoc threshold.
  • A manufacturer meets these requirements through a documented usability engineering process, typically built around EN 62366-1:2015+A1:2020.
  • Labelling and instructions for use are part of the usability picture, not a shortcut around it. Information is the last line of defence, not the first.

Why the regulation talks about usability at all

Tibor often tells a story from his early career that explains why MDR takes usability so seriously. Medical devices used to be assessed primarily on whether they worked when operated perfectly by a trained user in ideal conditions. In the real world, that is not how devices are used. Nurses work twelve-hour shifts. Home-care patients are sometimes eighty years old, sometimes anxious, sometimes colour-blind, sometimes trying to operate the device one-handed because the other hand is in a sling. The difference between a safe device and a dangerous one is often not whether the device can be used correctly, but whether the device can be used incorrectly without harming anyone.

That is the gap the MDR closes. Annex I §5 makes ergonomic consideration of the real user and the real environment a legal requirement. Annex I §22 goes further for devices intended to reach lay users, because the variability in skill, concentration, and environment is higher and therefore the design burden on the manufacturer is higher.

Felix sees the same story play out in Subtract to Ship coaching. The founders who treat Annex I §5 and §22 as compliance burdens end up resenting them. The founders who treat them as design briefs end up building better devices and sailing more cleanly through notified body review. The regulation is asking, in legal language, the same question a product manager would ask: who actually uses this, under what conditions, and what happens when it goes wrong?

What Annex I §5 actually says

MDR Annex I §5 is the section of the general safety and performance requirements dealing with risks related to the use of the device. It requires that, in eliminating or reducing risks related to use error, the manufacturer reduce as far as possible the risks linked to the ergonomic features of the device and the environment in which the device is intended to be used, considering the device for patient safety. It also requires consideration of the technical knowledge, experience, education, training, use environment where applicable, and medical and physical conditions of intended users. In plain language, the regulation tells the manufacturer: know who will use this, know where they will use it, know what they will bring to the use, and design accordingly.

Three phrases matter in particular. As far as possible is the risk-reduction standard. It is not as low as reasonably practicable and it is not to an acceptable level. It is stricter. It requires the manufacturer to push ergonomic risk down as hard as the state of the art allows, and only then to accept the residual risk against the benefit.

Ergonomic features covers everything from physical dimensions to colour, lighting, glove-compatibility, one-handed operation, auditory alarms, display contrast, and font size. It covers the hardware and the software and the packaging. Tibor has seen auditors call out issues with every one of these.

Environment of intended use reminds the manufacturer that a device designed for a brightly lit hospital ward may behave differently in a dim home bedroom, and that a device tested indoors may behave differently outdoors. This is where the bee story lands: a tongue-controlled wheelchair mouthpiece whose colour was chosen indoors turned out to be exactly what attracts bees, creating an outdoor-use hazard that only surfaced because an auditor asked the right question.

What Annex I §22 adds for lay users

Annex I §22 applies to devices intended for use by lay persons. It requires that such devices be designed and manufactured in such a way that they perform appropriately for their intended purpose, taking into account the skills and means available to lay users and the influence resulting from variation that can be reasonably anticipated in the user's technique and environment. It also requires that the information and instructions provided by the manufacturer be easy for lay users to understand and apply.

The regulation expects more, not less, of devices designed to reach the home. The logic is that a trained clinician can compensate for a poorly designed interface. A lay user at home usually cannot. Therefore the design burden on the manufacturer is higher when the intended user population includes lay persons.

In Tibor's audit experience, Annex I §22 is where summative evaluation choices become decisive. If the intended user group is lay persons, then summative evaluation must recruit lay persons. Substituting clinicians as test participants because clinicians are easier to book is exactly the kind of shortcut that costs a startup months when the notified body pushes back.

A worked example: a home-use blood-pressure monitor

Consider a startup launching a connected home-use blood-pressure monitor intended for patients over 60 managing hypertension. The device is a cuff plus a base unit with a small display and a companion app.

Annex I §5 requires the manufacturer to consider the technical knowledge and physical conditions of the intended users. The design lead notes that many intended users will have reduced dexterity, possibly reduced vision, possibly arthritic fingers, and possibly limited smartphone familiarity. The font size on the base unit is increased. The cuff is designed so that it can be put on one-handed, because some users will be doing exactly that. Auditory feedback is added because some users will not see the screen clearly in low light.

Annex I §22 applies because the intended users are lay persons. The manufacturer writes the instructions for use in plain language at a readability level appropriate for a general adult population, not at a clinical level. A summative evaluation is planned with 15 recruited lay users between 60 and 80, in a simulated home environment, with structured tasks covering unboxing, first-time setup, Bluetooth pairing, routine measurement, and response to an error state.

Two participants struggle to complete Bluetooth pairing. The interface is redesigned and a formative retest confirms the improvement. The summative evaluation is then completed cleanly. The usability engineering file ties every hazard-related use scenario to the risk management file under EN ISO 14971:2019+A11:2021, and the residual risks are reviewed against the benefit the device provides.

When the notified body audits the technical documentation, Annex I §5 is satisfied because the design considered the real user characteristics. Annex I §22 is satisfied because the summative evaluation recruited the correct user population and the information was written for them.

Where labelling fits inside the usability picture

Annex I §23 covers the information supplied with the device, including the label and the instructions for use. Startups sometimes read this as an escape hatch: if the device is confusing, add a warning. Tibor sees this move regularly and it almost always fails at audit. Information is a legitimate risk control, but it is the last resort, not the first. The risk control hierarchy that auditors expect is inherent safety by design, then protective measures, and only then safety by information.

In usability terms, a 150-page instructions-for-use booklet that nobody reads does not meet Annex I §22 for a lay-user device. The test that Tibor applies, and that a notified body will often apply in practice, is to give a recruited representative user the device and the instructions, offer no coaching, and watch them try to complete the task. If they cannot, the instructions failed and the design has to change, not just the wording.

The Subtract to Ship playbook for satisfying Annex I §5 and §22

Step one. Write a precise intended user description before writing requirements. The regulation requires the manufacturer to consider the technical knowledge, training, and physical condition of the intended user. That consideration has to start on paper before it can shape the design.

Step two. Write an equally precise intended use environment description. If the device will ever leave a controlled environment, the description must capture that. The outdoor-use wheelchair story is a reminder that environment considerations missing from the use specification lead directly to hazards missing from the risk file.

Step three. If the intended user group includes lay persons, flag Annex I §22 explicitly in the usability engineering file and plan the summative evaluation to recruit lay users. Do not rely on clinicians as proxies.

Step four. Use the risk-control hierarchy honestly. Design it out first. Protect against it second. Warn about it third. Train as the last resort. Notified body reviewers read risk files in that order and look for evidence that the hierarchy was respected.

Step five. Test the instructions for use the same way the device is tested. If a representative user cannot complete the task with the device and the instructions in front of them and no coaching, the instructions are not fit for Annex I §22 and must be redesigned.

Reality Check

  1. Does the technical documentation contain an intended user description that is specific enough to shape design decisions, or is it a generic phrase like "healthcare professional"?
  2. Does the intended use environment description cover every place the device will realistically be used, including edge cases like outdoors, low light, one-handed use, or gloved operation?
  3. Has Annex I §5 been addressed explicitly in the general safety and performance requirements checklist, with evidence tied to the usability engineering file?
  4. If the device reaches lay users, has Annex I §22 been flagged explicitly and has the summative evaluation been planned with lay participants?
  5. Can the risk management file under EN ISO 14971:2019+A11:2021 demonstrate that the risk-control hierarchy was applied in the correct order?
  6. Are the instructions for use written at a readability level appropriate for the intended user, and has that been tested with a recruited representative user?
  7. Has every hazard-related use scenario been traced into the risk file rather than left only inside the usability engineering file?
  8. If a notified body asked "where in the file is Annex I §22 addressed?", could the regulatory lead point to a specific section?

Frequently Asked Questions

Does Annex I §5 apply to Class I devices? Yes. The general safety and performance requirements apply to every MDR device regardless of class. A Class I manufacturer must demonstrate compliance with Annex I §5 even without notified body involvement.

What counts as a "lay person" under Annex I §22? A lay person is a user who is not a healthcare professional. That includes patients operating devices at home, caregivers without clinical training, and members of the public using a device in a community setting.

Can a Class IIa device for home use rely on instructions for use alone? No. Annex I §22 expects the device itself to be designed for lay users. Instructions are part of the solution, not the whole solution. A design that only works because the user reads a 40-page manual is unlikely to satisfy the regulation.

Is usability testing required for software-only medical devices? Yes. Annex I §5 does not distinguish between hardware and software interfaces. Software user interfaces are subject to the same ergonomic obligations as physical ones.

How does Annex I §5 relate to EN 62366-1:2015+A1:2020? Annex I §5 sets the requirement. EN 62366-1:2015+A1:2020 gives the manufacturer a practical process for meeting it and, when correctly applied, provides a presumption of conformity.

What happens if a manufacturer skips the ergonomic consideration entirely? The notified body will find the gap during technical documentation review. Nonconformities in Annex I §5 are among the more common usability findings Tibor encounters, and they typically trigger a design-level corrective action rather than a paperwork fix.

Sources

  1. Regulation (EU) 2017/745 on medical devices, consolidated text. Annex I §5 (ergonomics), §22 (devices for lay persons), §8 (risk control), and §23 (information supplied with the device).
  2. EN 62366-1:2015+A1:2020. Medical devices, Part 1: Application of usability engineering to medical devices.
  3. EN ISO 14971:2019+A11:2021. Medical devices, Application of risk management to medical devices.