Shift Labs (YC W15) Makes Medical Devices for Healthcare’s Future
techcrunch.comWhat's this? A medical device company that publishes prices and actually tries to make low-cost, high-value products?? I will be watching this very closely!
I am in medical school now, but I used to be an EE in the medical device space; my experience was that so many device makers sell through one or several layers of intermediaries that prices have to be jacked up substantially to make any money at all. We're talking selling products with a BOM <$150 for $3000+ (these weren't highly specialized med products either - you'd find similar devices in practically every doc's office). As a result, the companies never moved enough units to justify a refresh of the design more than once every 10-15 years.
I'm very happy to see someone trying a leaner approach, and I hope you guys do well! Sell me some cool stuff in a few years :)
Thanks for great feedback. The lack of transparency is definitely something that keeps the system broken, and we want to drive some change in that direction.
Also, let us know what kind of devices you want to see! We like to build things people want...
CTO of Shift Labs Koji here. Happy to answer any questions!
No questions, just wanted to say thank you. The healthcare industry is in dire need of having its costs driven down, and your team looks like its perfectly positioned to capture a substantial amount of revenue with this.
Thank you! Every day we wake up wanting to work towards a better healthcare solution. There are many ways to solve it, and this is just one of many, but we feel this is where we can make our impact.
Hi Koji,
Co-founder of AsthmaMD here [1]
The first product looks great. If/when you guys decide to venture into asthma care and are looking for partnerships please write me a line, salim at our domain.
Thanks for reaching out! AsthmaMD looks like a great product. Would love to connect in the future.
How is it that no one in the US market competes on price right now?
It's more about positioning of the solution, and thinking of new markets, not replacing existing markets.
We also have a business model that allows us to leverage software in ways that lower costs of doing business -- some traditional areas like sales and support, and some non-traditional ways as well.
Does your device carry the CE Marking?
The device does not currently have the CE Marking, but it is something we recognize as very important to sell in certain markets. We're working on it!
What are your testing methodologies like? Startups aren't what come to mind when I think of rock-solid engineering - but as a medical hardware company I'd expect a thorough and rigorous testing practice. What sort of automated or manual testing do you do? Is there any formal verification of the software?
I can't say the FDA process is the most fun thing I've ever done, but it's been a great opportunity to bring the rock-solid engineering framework to our product development. We use a combination of in-house testing and external testing labs to get the rigor required by regulatory agencies.
Probably the most important part of startup culture that's letting us innovate differently is the ability to create product development processes that prioritize human centered design. It means we develop and test in a different order, and our failure criteria for tests are not just about function, but also about form.
Pretty cool they got the price so low. I've consulted as an Android developer on the app side of medical software and for every day we actually wrote features we spent something like four testing. Unit tests, integration tests, UI automation tests, moving the app up from local dev, to integrated dev, to staging, to production (unused servers), then production (used servers). You can imagine how much that inflates the the time required and cost.
When we started, we really didn't have a good estimate of how hacking in the medical space would be different from other kinds of hacking we'd done.
But it's really not that bad once you get the hang of it. If we could say anything to other people considering taking the plunge, it would be to not be afraid of the medical space. Sure, it's a regulated space and some things are more complicated, but the difficult problems are the more interesting ones to solve!
More along those lines: http://www.shiftlabs.com/blog/on-being-a-medical-device-comp...
Do you work to comply with IEC 62304?
Do you do FMEAs for risk analysis, or some other technique?
Grateful for any insights you wish to share.
We do indeed. And we've done FMEA. We're in the midst of FDA clearance, so all the required quality work for product development is totally part of our process.
We're working on the assumption that regulatory processes establish benchmarks for performance, but they don't need to dictate how one makes things -- ie, design considerations can be front and center, not an afterthought.
hi. project manager at a major medical device manufacturing company here.
the reason medical devices are so expensive is a lot more complicated than what this article suggests.
1) upfront costs. you have to complete the development of the product before you can conduct clinical trials. this is a huge risk: if you say the device is ready, but the fda or eu says it's not, then all that work is gone or has to be redone and you have to foot the bill.
2) clinical trials and animal testing are insanely expensive. I'm talking 50k+ for 1 dog to test your device on. don't even get me started on human centric trials.
3) risk of litigation. look at the current mess with vaginal meshes: boston scientific, medtronic, and others are paying out huge sums of money over faulty tech that was approved by the fda.
4) regulatory overhead. navigation of the fda and ce mark beaurocracies is complicated enough that it is an entire industry in itself. we have a full team working on our project full time just to make sure we stay within the law, to meet with the fda and eu, and to do all our governmental paperwork.
5) ip law. medical companies sue eachother for patent infringement all. the. time. if you aren't checking to make sure you aren't violating patents, you are exposing yourself to huge amounts of risk. hundreds of millions of dollars of risk.
sigh.
plus all the overhead to navigate each hospital's unique purchasing process, dealing with different regulatory requirements in different countries, different radio spectrum requirements (if your tech uses any type of emitors) in different countries, traceability requirements in manufacturing, certified manufacturing sites, and all the rest.
sigh.
but the good news is that I'm an expert at navigating these treacherous waters and am looking for a new job. if anyone at shift labs (or other medical device manufacturer) is interested in a chat, feel free to reach out.
Great points in why things are complex, and why so often complex=expensive.
While each of the items you mention are totally relevant, we've also found that it's possible to mitigate a lot of these risks and costs with careful strategy.
For example, we're building devices that don't require animal or human clinical trials. That saves a tremendous amount of time and money in development costs -- but there are still large, lucrative markets to tackle.
IP is hugely interesting, and our approach to developing simpler technologies means we're not neck deep in the kinds of cutting-edge technology that comes out of research labs. That affects our IP exposure.
The regulatory hurdles are absolutely there, and for a long time we found them daunting -- almost paralyzing. But once we dove in, especially with a great expert on our side -- it became manageable.
This is all to say, you're right: it's hard, and it's complex, and it can be expensive. But we're really enjoying figuring out where the spaces are in between all these hard things and finding new ways to make things happen.
The press release is a bit contradictory. It says the reason things are so complex is because of a profit motive, but also the reason that they will succeed where others did not is because they are working for a profit. Also, they are going to trim all the bells and whistles but they are not going to be strictly functional.
All that being said I do think this venture is fitting into an emerging niche in the market place due to the low cost of designing/re-designing devices. Things like 3-D printers for prototyping, etc. means that the cost to develop a medical device does not require the profit margin of a high-end medical device in order to get back costs.
Functionality will not be made less of a priority, but we are making design and usability equally as important. Doing so helps cut down on training and support required for equipment, which is where many companies make their real revenue (similar to enterprise software charging support subscriptions). Because this is where so many established companies make their margin, there is little incentive to make their equipment easier to use. They also have built up large support staffs for this purpose.
By removing the need to support large support staffs, and reducing our sales team requirements by aiming at markets outside of large hospital institutions, we are able to keep our overhead low compared to the competition.
I hope that makes thing a bit more clear, but happy to elaborate if you like!
Just a quick comment to add: In many ways, what we're doing would probably not have been possible ten years ago. The markets were different, and so was the technology that enables us to implement changes to support and training. In addition, technologies that are facilitating rapid prototyping in other industries are just as essential to cutting down development time and costs for medical devices.
CEO of Shift Labs Beth here as well!
As a Doctor, just have to say that this looks great. I've often thought about how ridiculous our infusion pumps and syringe drivers cost, and have thought about playing around with 3d printers and working out my own. I think it's a great way to drive down healthcare costs. If you have any interest in developing an understanding of the Australian market then please reach out ( I have experience selling apps as a service to the clinical side of the public system here)
robbiep -- thanks for reaching out. It's pretty amazing that there's no middle ground between those super-expensive infusion pumps and hand counting. We used 3D printing for our early prototypes, and it was great for quickly figuring out which physical designs made people's eyes light up. Now we're doing injection molding to get the quality control for the medical sector. And sure -- let's talk about Australia!
Is the drip assist connected?
Not currently. The approach with the design was to focus on basic functionality and ease of use. We wanted to keep things simple. In addition, keeping things simple has allowed us to eek out 290 hours of battery life out of a single AA. Something our users have really appreciated.
We have already gotten strong feedback for more connectivity, so that's definitely something we're working on.