The clear focus for all of the HFES Healthcare Symposium 2017 attendees — from manufacturers to consultants to FDA reps — is to improve people’s lives through the application of human factors and usability engineering research.
When planning the human factors lifecycle for a product, companies must make thorough and rational assessments about what makes sense to test, to what depth, and when? The FDA issues guidance, but they don’t provide formulas for determining what’s the right answer for your company’s particular product and use scenario. Networking with others in the industry and bouncing questions off the FDA reps in a casual setting (over red beans and rice and jambalaya), reveals a multitude of practical insights that you’re likely to be able to directly apply to your own research.
For conference first-timers, it can be surprising to see FDA reps there in person, mingling with industry. Many have the impression that regulatory bodies feel more adversarial than supportive, but the human factors champions at FDA make it a point to be highly accessible to the industries they regulate. In terms of exposure to the FDA, the opportunities at this conference are great; this is one reason why everyone who works on medical devices should seriously consider attending.
Here are some of our key takeaways from the 2017 Symposium:
1. It’s very possible to avoid repeat summative studies.
This was a hot topic of conversation at the conference. The FDA reviewed existing guidance and provided statements regarding things to remember if you want to get your study approved the first time and avoid conducting multiple summative studies. Some of their recommended actions included:
Mitigate risk early in the product lifecycle by conducting iterative testing.
- Gather input from key opinion leaders early on.
- Conduct several smaller formative tests in advance of your summative.
Initiate open and early discussion with FDA.
- Set up a meeting with FDA to discuss your validation study plan well in advance of the study. FDA reps stressed their readiness to talk to companies early on in the submission process to help scope out the necessary human factors processes and assessments.
- Submit your summative study protocol to the FDA ahead of time, so the FDA can review and provide feedback on what needs to be changed before you run your study.
Ensure that your summative study protocol does the following things.
- Defines what success on each task looks like.
- Includes a Use-Related Risk Analysis that defines critical tasks, associated use errors, harm, and severity level. Don’t focus on the likelihood of occurrence of errors, because the FDA is more focused on severity of error.
- Includes known problems associated with your product. Even if your product is one of a kind and completely novel, you should be able to find similar products made by your company or other companies. Check the MAUDE Database.
2. You’re not alone.
One great thing about going to an industry conference is finding out how many of us – even reps from large companies, who sometimes seem to have it all figured out — often have the same challenges and questions. It’s useful as an industry to recognize these struggles, so we can all put our heads together to navigate them.
Here are some of the shared issues that stood out:
Gaining access to healthcare institutions to conduct contextual inquiry studies. This is possible, but not easy. Hospitals seeking to minimize liability might lean toward keeping researchers out, while manufacturers may need access to perform research in the environments where their products are intended to be used. But we all have the same goal – to help people by improving the safety and efficacy of products. So, how can we get past the sticking point of access to healthcare institutions, and how can we make everyone (e.g., hospital Institutional Review Boards, human factors consulting agencies, healthcare personnel) satisfied that it’s being done right?
Recruiting participants with rare diseases. Several interesting ways to recruit ‘hard to find’ low-incidence populations were discussed at panels and among colleagues. One unique recruiting method we heard was to work with a manufacturer’s clinical trial team to try and get those patients to come in for human factors testing. Another interesting recruiting method mentioned was to partner with a non-profit foundation and conduct your study at the same location where they have a big conference. The foundation will help spread the word about your study and you’ll have access to numerous potential participants in one location rather than having to travel to several cities.
Structuring differentiation tasks to avoid them becoming reading tests. While FDA and industry recognize the importance of ensuring a product is able to be easily differentiated from similar looking products, some have found the specific recommendations for how to conduct this differentiation testing frustrating. This was particularly true for recommendations that suggested participants should read a product name on a prescription and then find the product that matches that name. Some contend that in real life, patients are not merely matching product names to a product, but rather selecting a product to use based on their experience of having interacted with it one or many times before.
3. Instructions for Use (IFU) may need their own instructions for use.
There was plenty of discussion about Instructions for Use (IFU). One recurrent theme we heard was to submit IFUs early in the process, so the FDA can make its recommendations/edits before validation testing.
The FDA emphasized that they prefer to see both an editable Word version of the IFU so they can make track changes, as well as a PDF version so they can see the overall layout. Submitting your IFU before summative testing will save you from the sticky situation where the FDA reviews and makes changes to your IFU after a validation study, giving you no chance to test the changes (without increased time and cost).
A surprising proposition related to IFU design was presented by Dr. Eric Shaver, who suggested that IFUs (and all labeling) should consider avoiding using symbols that do not contain text. This is initially counterintuitive, because the purpose of a symbol is often to replace text. However, as Dr. Shaver discussed, a recent study by Seo, Landoni, Brunk, Becker, & Bix (2016) demonstrated that only six of 38 standardized ISO 9186-1:2007 symbols were successfully identified (defined as >85% correct response rate) in a sample of 86 healthcare provider’s — and five of the six symbols that were successfully identified contained text!
4. There’s a new guidance out this year on generic combination devices.
In order to be classified as a generic device, the generic needs to be so similar to the reference listed drug (RLD) that a user could pick up the generic device at the pharmacy without knowing it had replaced their brand-name product, and then go home and use it without any training.
To ensure a generic device will be able to replace a RLD, FDA guidance recommends the manufacturer conduct a threshold analysis, to evaluate external critical design attributes that may differ between the generic and RLD.
A threshold analysis includes:
- Labeling comparison: Compare the generic and RLD Instructions for Use, and identify any differences.
- Comparative task analysis: Walk through the steps and sub-steps required to use the generic and RLD, identifying where differences occur.
- Physical comparison: Examine the physical features, colors, and mechanisms of the generic and RLD and highlight any differences.
As you can imagine, particularly when the RLD you’re working toward pre-dates the industry’s focus on device usability, it can be challenging for a company that wants to improve the RLD’s user experience (and potentially fix known problems), yet still get qualified as a generic.
After conducting the threshold analysis, the manufacturer should determine if there are no changes, minor changes, or major changes. If major changes exist, design modifications should be considered if a product wants to be classified as a “generic.”
5. Every industry needs at least one Dr. Rafael Grossmann.
Dr. Grossman is an expert surgeon and “healthcare futurist” who loves technology. Among his claims to fame is that he was the first person to wear Google Glass while conducting live surgery. You can learn more about him by watching his excellent TEDx talk, “iPod Teletrauma: the $229 130 million sq. ft. Trauma Room.”
We learned and were inspired while listening to Dr. Grossman’s opening plenary talk about new, once-considered-futuristic technologies, that are now being used to help train for and conduct medical procedures. His talk had us dreaming about a future where simulated-use formative and summative testing could be done using holograms.
Here are some of the technologies he introduced us to that we’re excited about: FlexDex Surgical; Clarius Wireless Ultrasound; Touchsurgery App, a free mobile app with step by step instructions for surgical procedures; and Axon VR Suit.
And our last takeaway? New Orleans is a fantastic place to hold a conference. The excellent food — perfectly crispy blackened fried chicken; sweet potato beignets with a chicory coffee sauce; shrimp and grits…. yum! And, of course, the live music everywhere made the time spent away from the conference as enjoyable as the program itself.