A documented selection scheme with rationale for which hazard-related use scenarios are included in summative evaluation, and the results of applying that scheme stored in the usability engineering file.
AMD1:2020 restructured and clarified the three selection options; updated ISO 14971 cross-reference for severity determination from 2007 to 2019 §5.5.
Documented exclusion rationale for any scenario not included in summative evaluation — undocumented exclusions and excluded high-severity scenarios are the two most frequent clause 5.5 findings.
Maps to
IEC 62366: §5.5 Select the HAZARD-RELATED USE SCENARIOS for SUMMATIVE EVALUATION
ISO 13485: §7.3.6 Design and development verification
ISO 14971: §5.5 Risk estimation
Requirement text
The manufacturer shall select the hazard-related use scenarios to be included in the summative evaluation. The manufacturer shall select either: (a) all hazard-related use scenarios; or (b) a subset of the hazard-related use scenarios based on the severity of the potential harm that could be caused by use error (e.g., for which medical intervention would be needed); or (c) a subset based on severity and other circumstances specific to the medical device and manufacturer. A summary of any selection scheme, the rationale for its use, and the results of applying it shall be stored in the usability engineering file.
Why this clause exists
Clause 5.5 exists because summative evaluation with representative users is resource-intensive, and the standard recognizes that not all hazard-related use scenarios carry equivalent risk of serious patient harm — but the selection cannot be arbitrary. The scenario selection step is the gating mechanism that ensures higher-severity scenarios are always included in summative evaluation while providing a documented, auditable basis for any scope reduction. Without a formalized selection step with documented rationale, manufacturers either test everything (which is sometimes impractical for complex devices with many scenarios) or selectively test without justification (which creates risk that high-harm scenarios are excluded). The clause’s requirement to store the selection scheme, rationale, and results in the usability engineering file is specifically designed to make the selection decision visible and auditable: a notified body or FDA reviewer examining the summative evaluation plan can verify that the scenario set is not biased toward low-difficulty, low-risk interactions. FDA’s human factors guidance consistently flags summative evaluations that excluded high-risk scenarios as a major deficiency, because the evaluation cannot demonstrate safe use if the most hazardous interactions were not tested.
What changed
IEC 62366-1:2015+AMD1:2020 CSV clause 5.5 was substantially restructured by AMD1:2020. The 2015 edition listed three selection options in a simpler format; AMD1:2020 reformatted and clarified the options, explicitly noting that the severity of potential harm from use error is the primary selection criterion for subset approaches. AMD1:2020 also clarified that the selection scheme may additionally depend on other circumstances specific to the medical device and manufacturer, beyond severity alone. The ISO 14971 cross-reference for harm severity determination was updated from ISO 14971:2007 to ISO 14971:2019, §5.5. The core obligation — document the selection scheme, its rationale, and its results in the usability engineering file — is unchanged.
Common gaps (what we see in audits)
- No documented selection scheme or rationale for scenario subset selection — Manufacturers include a subset of hazard-related use scenarios in summative evaluation without documenting the selection scheme or the rationale for exclusions. FDA reviewers and notified bodies treat undocumented scenario exclusions as a major deficiency because the evaluation cannot be assessed for completeness without a documented selection rationale.
- High-severity scenarios excluded from summative evaluation — Some manufacturers exclude hazard-related use scenarios with high harm severity from summative evaluation — often because the scenario is operationally difficult to simulate in a controlled test setting — without documented rationale or alternative evaluation evidence. Excluding high-severity scenarios without justification is a direct violation of the clause and a major finding.