Regulatory glossary

Human Factors Engineering: Use Errors, Critical Tasks, Validation

Human factors engineering, also called usability engineering, is the discipline of designing a device so the people who use it do not make errors that harm patients. FDA treats it as a safety question, not an ergonomics one.

The governing standard is IEC 62366-1. FDA's expectations are set out in its 2016 guidance, Applying Human Factors and Usability Engineering to Medical Devices, and the work feeds design validation under 21 CFR 820.30(g).

The idea that drives everything

A use error is not user error. The term is deliberate. If a nurse misprograms a pump under time pressure, the finding is about the interface, not the nurse. Your obligation is to design out the error or mitigate its consequences.

That framing decides how the whole file is written. "Trained users will read the instructions" is not a mitigation.

Use-related risk analysis

The analysis starts from tasks rather than components. For each task a user performs, you identify what could go wrong, what harm would follow, and what in the design makes the error likely.

From that you identify critical tasks: tasks where a use error would cause or contribute to serious harm. Critical tasks are the ones validation has to cover, and getting the list wrong invalidates the study that follows.

Common sources of use error, worth checking against your own device:

• Similar-looking controls or connectors that can be interchanged.
• Alarms that are ignorable, ambiguous, or indistinguishable from each other.
• Default values that are safe in one context and dangerous in another.
• Instructions that assume the user has the manual open, in daylight, unhurried.

Formative and summative

Formative evaluation happens during development. Small, iterative, on prototypes, with the purpose of finding problems while changing the design is still cheap. There is no fixed participant count and the results are not submitted as proof of anything.

Summative evaluation, also called human factors validation, is the study that goes in the submission. It tests the final design, with representative users, under realistic use conditions, on the critical tasks.

FDA expects at least 15 participants per distinct user group. Distinct means genuinely different: a home caregiver and a trained clinician are two groups, and two groups means 30 participants, not 15.

Every use error and difficulty observed has to be analysed, not just counted. A single failure on a critical task requires an explanation of why the residual risk is acceptable.

Combination products get caught here

An autoinjector, a prefilled syringe, an on-body injector, a nasal spray, an inhaler: these are drug-device combination products, and the device part carries the human factors obligation while the sponsor is often a pharma company with no device group.

This is a recurring theme in Complete Response Letters for emergency-use products, where human factors deficiencies appear alongside clinical or pharmacokinetic ones. The drug can be fine and the application still fail because nobody proved a panicking bystander can operate the device.

If your product has a delivery mechanism, plan the human factors work at the same time as the clinical programme. Retrofitting a validation study at the end is how a submission slips a year.