April 26, 2021
In 2020, and for the past few years, our team has published blog posts summarizing the latest expectations that US Food and Drug Administration (FDA)’s Human Factors reviewers have shared during the Human Factors and Ergonomics Society (HFES) Health Care Symposium. Here’s Part 1 of our 2021 edition, focused on what we learned from individuals representing FDA’s Center for Drug Evaluation and Research (CDER) division. A similar blog focused on key points from the Center for Devices and Radiological Health (CDRH) will follow shortly.
Key takeaways from FDA’s 2021 CDER-led workshop
Below we summarize this year’s key takeaways related to CDER, collected primarily from the CDER-led workshop, as well as from their inputs provided during panels throughout the conference.
- A robust use-related risk analysis continues to be pivotal – the “backbone of the entire HFE process.” Every year, the FDA provides consistent feedback regarding the importance of use-related risk analysis. This year, CDER emphasized the need to utilize correct risk analysis terminology, quizzing attendees on whether harm and risk refer to the same thing (they don’t!), and encouraging companies to rely on the regulatory definitions of harm presented in FDA’s guidance documents. CDER places emphasis on evaluating any potential for harm, including harms associated with compromised medical care. Unlike CDRH, CDER is focused on any potential harm, rather than on potential harm associated with a serious or greater risk.
- High-risk use conditions are not necessarily “worst-case” scenarios. In the context of a use-related risk analysis, CDER reminded companies that they should pay particular attention to higher-risk use conditions. Such conditions might result from lack of user training, negative transfer resulting from users’ experience with similar products, and impairments that users might experience due to their particular medical condition(s). These scenarios were historically referred to as “worst-case,” but that term is now considered outdated because these scenarios are realistic and expected with combination product use.
- Tasks outlined in the use-related risk analysis should align exactly with steps outlined in a product’s Instructions for Use (IFU). The agency emphasized that the intent of an HF validation study is to validate a product’s user interface and, as such, they expect to see evidence that users follow the IFU steps as written. CDER expanded on this point with an example, stating that if an IFU for an injection device specifies a required 10-second injection time, the associated risks in the use-related risk analysis should match this 10-second expected duration. If a manufacturer can provide justification or evidence that all medication is delivered in a shorter injection time (e.g., 5 seconds), then they can present this rationale as part of their residual risk analysis. In short, it is inappropriate to set one expectation for use in the IFU (e.g., a 10-second injection), and a different expectation in the use-related risk analysis (e.g., a 5-second injection).
- HF validation study participant training must be justifiably representative of real-world training. There was a lot of discussion about the product training that might be provided to HF validation study participants. CDER recognizes that user training might range anywhere from unavailable or unavailable consistently, to consistent and intensive, in-person training. To include only trained participants in an HF validation study, manufacturers must have defensible “administrative controls” to ensure all users can and will receive training. Given the known variability in training for most combination products, having all untrained participants – or an untrained “arm” to complement a trained “arm” for each user group – is most likely the acceptable path forward from CDER’s standpoint.
- “No product design will be risk free.” We frequently counsel clients on how to interpret and react to the findings from usability tests we conduct, especially when it comes to the pivotal HF validation study. It’s always comforting to hear the Agency reinforce that there is not a “zero-tolerance” policy when it comes to observing findings during an HF validation test. The Agency expects that medical and pharmaceutical products will carry some risk, and that use errors and other interaction difficulties might occur. The key is to minimize the risk to a justifiably acceptable level through good user-centered design and (ideally, design-based) mitigations.
- The validity of remote testing methods for HF validation is unknown. One of the most common questions we fielded from our clients over the past year is whether usability studies could be conducted virtually or otherwise remotely to enable development efforts to proceed despite the restrictions COVID-19 placed on some in-person testing methods. Our response has been yes, in some cases, for formative testing and maybe, in rare cases and with explicit Agency approval, for HF validation studies. The Agency explained that do not yet feel well-equipped to make a declaration on the validity of remote testing. However, they are eager to review protocols and proposals regarding how to conduct such testing, and to hear from Sponsors and industry what might be feasible, presumably to help inform future policy.
- An HF validation study is one of several user studies expected for Over-The-Counter (OTC) products. In a shift from prior symposia, CDER spent some time outlining the expected user studies for OTC (i.e., non-prescription) products. The agency described four different studies: Label Comprehension Study, Self-Selection Study, Actual Use Study, and likely the most familiar for those in the HF field, HF Validation Study. Whereas the first three studies are quantitative in nature and call for very high sample sizes (~400+ participants), the agency confirmed that HF validation studies of OTC products are similar in their scope, objectives, and more qualitative nature to HF validation studies of prescription products. Among other details, CDER expressed the importance of including individuals with low health literacy in all studies, suggesting that health literacy levels be measured via the REALM test.
As always, FDA representatives encouraged manufacturers to contact them or the FDA project manager via appropriate channels with any questions or to seek input on a specific situation.
The Emergo team welcomes any outreach to discuss any of the above topics further and consider how the FDA’s current stance might affect a manufacturer’s HF strategy or plans.
Allison Strochlic (Research Director) and Laura Birmingham (Managing Human Factors Specialist) are both a part of Emergo by UL’s Human Factors Research & Design division.