Device regulation is a hurdle for any medical tech company, but doubly so for startups hoping to strike it big with their very first digital health product. With limited resources and experience, pushing a device through the FDA’s door and onto the market is a multi-year gauntlet of validation, documentation and, unfortunately, frustration.

To help young companies take a bit of the guesswork out of the endeavor, Dr. Steve Pham, VP of clinical and research affairs at Eko, shared some of the lessons his digital cardio monitoring startup learned first hand when submitting is first digital stethoscope and portable ECG.

At the top of his list was the need for startups to build a clear narrative for regulators about the device in question — or, who it’s for, what it does best and how it should be implemented in the real world.

“Our product is a stethoscope and it can probably do 10 different things, but in any submission that we do it needs to be very clear what is the use case,” Pham said this weekend during a health tech regulation panel at the American Heart Association Scientific Sessions 2019 in Philadelphia. “When you make these things clear, it doesn’t muddle the narrative. So, for instance, we make claims about ejection fraction using a single-lead [ECG], but the important thing to note is that we are not an ultrasound machine. We’re never going to be a diagnostic, but instead should help clinicians make a determination when they should use an [ECG] or ultrasound test.”

In building that product narrative, it’s likely that a startup will stumble upon the shortcomings of their device and the question of how to address it. In general, Pham said the FDA is more likely to appreciate and engage with a startup that isn’t trying to sweep bad news under the rug.

“There must be some kind of weakness that you should be able to anticipate, and rather than avoiding the narrative of the weakness you should really be diving ahead into it,” he said. “When we submitted for one of our algorithms, we anticipated that weakness and already started collecting data to study that weakness even as we submitted it. So it’s kind of a continuous delivery model where you’re looking at data, you’re anticipating the weaknesses and you’re studying your own weaknesses while you’re submitting and talking to the FDA. And certainly in our last conversation with the FDA, they really appreciated us not ignoring the problem.”

This sentiment of open communication was echoed by Dr. Linda Ricci, an assistant director of digital health at the FDA’s Center for Devices and Radiological Health. Instead of being scared off by the prospect of red tape and strict ultimatums, startups are better off reaching out to the agency directly so that they can understand what’s expected of them early on in the process.

“We always say ‘Come early, come often.’ It’s never too early to come and have a conversation with the FDA about how your device may be regulated,” she said during the panel. “You may be surprised as we walk through what you intend to do to find out that it may have a very light regulatory touch. But oftentimes we can help small companies understand the type of data that they need to collect before they start to collect it so they don’t waste any money.”

Sometimes it’s not just about the benefits of early regulatory work can bring to a company, but the roadblocks it could create for competitors following in its wake, Pham said. Alongside being the first to market with a particular technology, a startup speaking with the agency has a chance to further protect its business by setting the stakes for competitors looking to develop their own alternatives.

“I think a lot of clinicians don’t see this side — that if you do engage early and often, you may be able to sort of set the stakes for future companies,” he said. “It’s a very strategic way to think about building products and algorithms. … Especially with the 510(k) process, you’re looking at other companies that have done what you [are] doing before. If you’re that first person, then every other 510(k) cleared device has to look at you and what you did and meet that minimum bar. So I think there are a lot of strategic elements I think are left out of the conversation that are very interesting. And if you’re at all interested in thinking that way, engage with the FDA as soon as possible and get really smart minds to think about strategy behind it.”

That being said, receiving that approval isn’t the end of the road for a device or software maker, Pham noted. The last few years has seen the FDA draft and release new guidances outlining regulatory processes for monitoring post-market performance and enabling refinements for algorithms and other software-enhanced products. This means that startups should be prepared to maintain the infrastructure for testing their products, and to keep up their relationship with the agency well after their device has hit the market.

“This is a lesson that most device manufacturers know this but a lot of health tech startups probably don’t: [after] you’ve already submitted or been approved … there needs to be a machine or process to continuously improve the algorithm,” he said. “We run our trials for highly curated data almost on a continuous basis. We have not stopped any of our trials for the last two or three years. So, it’s just something to think about. If you can do it cheaply and can do it continuously, you should continue to do that and engage with the FDA with your validation trials.”