← Back to Insights

How Many Participants Do You Need for a Usability Study? FDA Sample Size Requirements

For a summative (validation) usability study, FDA expects a minimum of 15 participants per distinct user group. For formative studies, 5–8 participants per user group per round is the working standard. The total study size therefore depends on one question teams consistently underestimate: how many distinct user groups does your device actually have?

Where the 15-participant number comes from

FDA's 2016 guidance, Applying Human Factors and Usability Engineering to Medical Devices, states that validation testing should include at least 15 participants from each distinct user population. The figure traces to research on problem-discovery rates in usability testing showing that 15 users uncover the large majority of usability problems for a given user profile, with sharply diminishing returns beyond that.

Two important clarifications:

1. 15 is a floor, not a target ceiling. If your risk analysis, device complexity, or population variability justifies more, use more. FDA reviewers question undersized studies, not oversized ones.
2. It applies per user group, not per study. This is where budgets surprise people.

What counts as a "distinct user group"?

A distinct user group is a population whose differences could affect how they interact with the device — different roles, capabilities, training, or use contexts. Common splits:

• Role: patients vs. caregivers vs. nurses vs. physicians
• Training status: trained vs. untrained users, professional vs. lay users
• Capability differences relevant to use: pediatric vs. adult users; users with condition-related impairments (dexterity, vision) that the intended population includes
• Use environment, when it changes the interaction: home use vs. clinical use of the same device

If the groups would receive different instructions, perform different tasks, or face different risks, they're distinct.

Worked example

An on-body injector used by patients, supported by caregivers, with initial training delivered by nurses:

Add a planned 15–20% overrecruitment for no-shows and screen failures, and the recruitment target is roughly 52–54 people. This is why user-group definition belongs in the budget conversation, not after it.

Sample sizes for formative studies

Formatives optimize for iteration speed, not statistical coverage. The standard practice:

• 5–8 participants per user group per round — enough to reveal recurring use problems
• Multiple small rounds beat one large round. Two rounds of 6 with a design fix in between outperform one round of 12 watching the same problem repeat.
• Early concept work can go smaller (3–5) when you're testing one design question.

When you need more than 15

• High-risk critical tasks where the consequence of undetected use error is severe
• Heterogeneous user groups you've chosen not to split (e.g., a wide age range with capability differences)
• Statistical claims: if you intend to make quantitative claims about task success rates, 15 provides very wide confidence intervals — discuss with FDA in a pre-submission if quantitative criteria matter to your case

Frequently Asked Questions