Last month, I reported on FDA review times and how they’ve increased 34% since 2006. The main cause attributed to the increase in review times is the data necessary for FDA to evaluate the safety and efficacy of the device is not being included with the original 510(k) submission.
Now, let’s look at FDA’s report “Analysis of Premarket Review Times Under the 510(k) Program” to see how the frequency of additional information (AI) requests have increased since 2006. FDA reports that the percentage of AI requests on the first review round for 510(k)s has increased 21% since 2006. In 2006, approximately 56% of submissions resulted in an AI request, whereas 2010 had an average of 77%. FDA also reports that the number of review cycles for 510(k)s has increased from an average of 1.7 review cycles in 2006 to 2.1 review cycles in 2010. FDA looked at the typical AI requests, which include lack of performance or clinical data, failure to follow an FDA guidance document or recognized standard, and problems with the indications for use. These request for additional information correlate with the causes identified for the longer review times. Not surprisingly then, FDA is reporting longer review times, more AI requests, and more review cycles since 2006.
As reported in my previous post, Anson has also experienced longer review times and an increase in the AI requests and number of review cycles so this is not just empirical data for us either; it is real work and real time. Approximately 70% of all Anson-supported submissions between 2006 and 2010 have resulted in at least one round of AI requests. In 2006, 58% of Anson-supported submissions resulted in at least one round of AI requests. In 2010, this increased to 80%.
This begs the question we all want answered; what can we do to decrease the likelihood of an AI request, and therefore, longer review times? Here’s my advice:
- Search for any FDA guidance documents or International Standards that apply to your device type. Read them carefully and be sure that you are addressing all of the requirements or provide a scientific rationale for not doing so.
- Look at what other similar devices have included in their submissions either through a Freedom of Information (FOI) request or by looking at the published 510(k) summaries on FDA’s website. Learn from other’s mistakes and successes.
- Write the 510(k) in such a way that assumes the FDA knows nothing about your device. With literally thousands of device submissions per year, chances are they are not intimately familiar with your product even if its capabilities have already been previewed in testing, pre-market demonstrations, or other means. That said it’s important to strike a balance between not providing enough information and providing too much unnecessary detail which can also lead to longer review times and confusion for reviewers. Clarity above all is key.
- Make it clear in your submission that your device is substantially equivalent to – as safe and effective as – the predicate device. If you provide test summaries or “promise statements” in your submission, be prepared to submit the full report should FDA request them.
- FDA has rolled out several new draft guidance documents recently, including one on mobile device applications. Read them and track the release of new ones carefully. Dr. Shuren has stated in testimony on Capitol Hill that FDA will continue to draft or update guidance documents as a means to be more transparent and clear in communicating it’s expectations to industry, and to make them more applicable to the rapid advancement in new technologies. If FDA provides more clear direction and industry attempts to provide more complete submission documentation based on that direction, we’re confident we will see a reduction in the number of AI requests, and most importantly, faster review times, which means getting products to market quicker