Why Failure Mode Identification is Preferred to Identifying Every Conceivable Use-Error

By Pamela Davol, Clinical & Regulatory Project Manager

A Failure Mode is defined as “the manner in which a device (equipment, software, etc.) may fail functionally.” Depending upon the complexity of a device, there may be many conceivable use-errors for every failure mode, not all of which may be observed in usability testing and thus not necessarily required to be identified in a UFMEA. With complex devices where it is essential to prioritize use-steps for evaluation, rather than spending time conceiving of every possible misuse/use-error for a product, identification of primary failure modes associated with use-steps will identify those use-steps that are critical and/or essential for safe and effective product use by end-users (i.e. failure modes resulting in higher severity of harm ratings, independent of probability of occurrence, or impacting on intended product treatment, whether or not diminished/over- treatment may result in harm to the patient).

Accordingly, all device failure modes should be identified, but not necessarily all use-errors if the latter are not observed in usability testing (i.e. identify only those probable to occur in the preliminary analysis and then add those actually observed during user testing if not previously identified).

Why is this important? In the example below, all of these use-errors/misuses, although falling in a hazard criticality level of “acceptable” under ISO 14971 applied processes will still require justification of residual risks by the FDA in the final HFE&U reporting structure regardless of their hazard criticality (HC) rating (i.e.even if falling in the acceptable level of risk).[1],[2],[3] This is because FDA will focus on severity of harm independent of probable occurrence. Additionally, FDA will focus on the impact on treatment (i.e. over-treatment, under-treatment, or no treatment) even if severity of harm related to the treatment impact is negligible. Calling out use-errors that have not actually been identified in user testing or with other predicate devices/comparable products significantly adds to the burden of usability testing and final reporting.

Key Point: When creating a product UFMEA, do not get creative in coming-up with potential use-errors unless there is a known prevalence (recalls/complaints on predicate devices/comparable products) or actual user testing data to justify! It is much easier to add to a UFMEA (if a use-error is observed) than it is to remove it; the latter requires justification when reporting. It makes the regulatory preparation and submission of a HFE&U report much more cumbersome and likely to go less smoothly during HFE group product usability review (especially any products going through CDER in addition to CDRH; i.e. drugs with delivery devices).

To learn more about Ximedica’s regulatory expertise or about Pam Davol, visit our clinical & regulatory affairs page.

[1] CDRH, Draft Guidance for Industry and Food and Drug Administration Staff: Applying Human Factors and Usability Engineering to Optimize Medical Device Design, June 22, 2011.

[2] CDER, Guidance for Industry: Safety Considerations for Product Design to Minimize Medication Errors (Draft Guidance), December 2012.

[3] CDER, Guidance for Industry Safety Considerations for Container Labels and Carton Labeling Design to Minimize Medication Errors, April 2013.