FDA describes the goal of human factors and usability engineering as minimizing use-related risks and confirming that intended users can use the device safely and effectively in the intended use environment.
Formative evaluations occur during development and inform design decisions. The build needs enough fidelity to present the workflow and interaction questions the team wants to examine.
Before the session
Define the evaluation around the product.
- Intended users and relevant user characteristics
- Intended use environment
- Expected workflow and task sequence
- Known use problems and use-related risks
- Questions the evaluation needs to answer
- Representative scenarios, data, and app configuration
Build readiness
Use a version that can be evaluated consistently.
The agreed workflow should run from start to finish with repeatable screens, interactions, data, and system behavior. The build version, included functions, known issues, and limitations should be recorded before the evaluation.
Client review and testing builds can be delivered through Apple TestFlight. Synthetic or approved nonsensitive data can be used when the scope does not call for live personal or clinical data.
Usability records
Keep the questions, observations, and changes connected.
The usability engineering file can capture the intended users, use environment, task analysis, use-related risks, evaluation plan, observations, findings, and design changes. The records should identify the exact app version evaluated.
Formative findings may lead to interface, workflow, labeling, training, risk-control, or test changes. Those changes should flow back into the applicable product and development records.
Formative and summative work
They answer different questions.
Formative work is used during development to identify and address interaction problems. Human factors validation is performed on the final user interface to demonstrate that intended users can perform critical tasks without serious use errors or problems.
The planned evaluation, participant profile, critical tasks, use environment, product maturity, and regulatory path determine whether a particular build is ready for the next study.
A practical readiness check
Can the team identify exactly what is being evaluated?
Before the first participant begins, the team should be able to name the app version, intended users, use environment, workflow, evaluation questions, known limitations, and the records that will receive the findings.