Last week I attended the sixth annual HFES International Symposium on Human Factors and Ergonomics in Health Care in New Orleans, LA, USA. The event attracted 460 attendees from industry, academia, consulting firms, government, and healthcare.
The day before the symposium, the FDA gave a workshop titled, “Product Development and Human Factors Considerations: Navigating the Halls of FDA/CDER.” The workshop was given by CDER’s Quynh Nhu Nguyen, Associate Director for Human Factors, and Mishale Mistry, Acting Director.
The workshop was very well organized and provided key insights into CDER’s latest thinking on the validation of combination products. It was lively and interactive with lots of question-and-answer throughout; presenters even asked the audience some pre-planned questions about our best practices.
Here are some of the important things I learned:
Use-Related Risk Analysis: CDER is now using the term Use-Related Risk Analysis (URRA), rather than the more common Use Error Risk Analysis (UERA).
Critical versus essential tasks: When asked about categorizing critical versus essential tasks, Quynh Nhu Nguyen said that in submissions for review, CDER would like to see a list of all tasks with an indication of which the manufacturer thinks are critical versus essential, so the Agency can make their own determination. In other words, walk them through how you classified severity so they can decide if they concur; don’t discuss only the critical tasks.
What to submit for an HF review: The presenters also showed a slide listing what to submit when requesting a review of a human factors validation protocol:
Samples of the product (preferably 5)
The URRA
Intend-to-market labeling and packaging
The IFU (in editable Word format, not PDF)
The HF validation protocol
Summary of preliminary analyses and evaluations and resulting UI changes
Pre-filled syringes with no new risks: Regarding combination products with pre-filled syringes, if the syringe is already marketed and there are no new risks, manufacturers can submit an argument and rationale as to why additional HF data for nurses is not necessary (patients are a different story, however). This obviously has an impact on sample size.
Recruiting patients with rare diseases: We discussed the difficulty of recruiting enough participants for validation studies involving patients with rare or orphan diseases. It is extremely challenging to get 15+ participants in each user group when the disease is rare by definition. Quynh Nhu Nguyen confirmed that in some cases it may be possible to use “surrogates,” or people with the same range of characteristics as the target populations, but that manufacturers must provide a justification for why these surrogates are the same or similar. [Note that Jacquelyn Wan and I co-authored a presentation on this topic that Jacquelyn gave later in the week. The slides are here.]
Summary of known use problems: CDER wants manufacturers to include a summary of known use problems when submitting a HF validation protocol for review. This analysis should be done at the start of the product development process to inform the entire URRA, but typically it isn’t summarized until the final HF/UE Summary Report at the time of submission. This new statement means manufacturers should submit it earlier.
Standard decay of training times: There was a long discussion about simulating training and training decay times in validation studies. The presenters made an important statement near the end: CDER expects to see one standard decay time for all participants. This has implications for studies in which participants are trained one day and return another day for their simulated use tasks. In the past these decay times may have varied, with some participants returning one day later and others several days later. Apparently for CDER this is no longer acceptable.
These are only a few of the things I learned during the workshop; there were many more. Thanks again to Quynh Nhu Nguyen and Mishale Mistry for being so open with industry and for clarifying the ingredients of a successful human factors submission to CDER!
Kommentare