Brozinsky: First of all, we chose them because we wanted to have a representative sample of the U.S. health system writ large. We have large academic medical centers and community hospitals. We have some that are deep into value-based care, and others that are more steeped in fee for service. We've got urban and rural, some that have health insurance companies, some that don’t. But the guiding principle was they had to be leaders in their region and not competitors with each other, so that we could facilitate real collaboration. With all that diversity, the problems are extremely similar. How they need to solve them has variances that are critically important. So there might be a core approach that's common, but the nuances matter pretty significantly. 

HCI: But is there a special sauce that makes this work for both the health systems and the startups? Is there a single approach to co-development or do you have to customize it for each one?

HCI: Is there usually a person on the health system end who’s a chief innovation officer or a chief digital officer overseeing this work?

Brozinsky: We have found it is not just one person. All these health systems have chief innovation officers or chief transformation officers, and of course, the CMO is going to be critical. What we try to do is build relationships with all the stakeholders who have a say. We certainly want to get enterprise-wide sponsorship, but then the real magic happens when they can help navigate us to someone who's living with that acute problem. Because the way to really get traction is to have a big vision of what it could look like in the enterprise, and figure out how we help guide our companies to land in the right spot. We want to have the real clinical champions within a service line prove it out quickly and get outcomes. At the end of the day, all of our companies are in the business of selling outcomes. We can go back to those who sponsored it and saw the big vision. They now have the data that that supports rolling it out more broadly.

HCI: Are you regularly surveying these 11 health systems to get a feel for the common top priorities they want to focus on? And is the generative AI trend impacting all that? 

Brozinsky: Our first step when we onboard a new health system in our consortium is to ask: can you share with us your biggest problems and prioritize them, and then your one-year and three-year plans, and your 10-year vision. That gives us context as to what they're worried about, what they're focused on, and where there's white space. Something might be a real big problem and very important for one health system, but it doesn’t lend itself to a successful company. Our work is to understand which of those problems can be better solved by partnering with us. What do we actually bring to the table? What's an Aegis company? And can we see the connective tissue across the industry, and not just in our consortium? By the way, our consortium is not designed to be exclusive, but inclusive, meaning that if you're in it, you're in it because you want to lean in and help shape solutions. We want to build companies that serve the whole market.

HCI: Let’s go back to their priorities, and the focus on AI. Are a lot of them now saying they want to figure out how to use AI for efficiency or for clinical improvement?

Brozinsky: I think it ebbs and flows. Everybody got excited by the zeitgeist moment of ChatGPT and then it went through a little bit of the valley of disillusionment once they started to try to implement these things. But it's certainly the No. 1 thing on everyone's mind. How do you apply generative AI  or agentic AI to workforce issues? We're absolutely seeing that as being the No. 1 priority — how to implement to get efficiency and effectiveness. So far, the focus has been more on efficiency. How do I do what I'm doing now but do it better? We feel there's a lot of opportunity there, but we actually think there are problems that you couldn't solve previously that now you can revisit and use AI to solve. 

HCI: Have you developed a pipeline of tech experts and entrepreneurs who you match up with the problems the health systems identify?

Brozinsky: Our approach is that we get the signals from the health systems, and identify some ideas about how best to solve them. We put that in a hopper, and we're constantly prioritizing them for ourselves, and asking what would make a good company. Can it generate returns? Do we actually add something unique to the mix? Then we create a set of theses. We're operating under five co-creation theses. Right now, we think we have five robust areas that are very fertile to build multiple companies in each one. What we're doing in parallel is building a pipeline of relationships with scientists, researchers, entrepreneurs, founders, and then the broader teams of product folks and engineering. 

The same way we're getting signals from the health system side, we’re doing that on the talent side.

Sometimes someone else has been deeply thinking about it — a scientist or researcher — and we pull that person in early. They may have even built a team, maybe even started a company, but it's early enough where we can co-create it together. Our whole value proposition is we want to compress the time to product.

HCI: I get the impression that many startups have trouble getting their foot in the door at health systems to do pilots so that they can then make a case for their impact. That's why your approach of having the grounding in these health systems, which are willing to co-develop solutions, seems so much more solid of a foundation.

Brozinsky: It’s hard enough to get one health system to do that. Once we align on an idea and decide we're pretty sure we're going to launch something, we go back to all those health systems and say, “Here's why you recommended we think about this space. Now we've got input from the whole consortium, and this is what it looks like.” And we quickly go through prototyping and business planning and and we get their input on that process, too. Then we say, “OK, now we're fairly sure we're going to pull the trigger. Will you make a commitment to use it?”

So we're getting to the point where we're getting health systems to commit to being customers, contributing data and assets if it's relevant, before we start the company. So then if we can de-risk it. 

The statistics on startups that get from seed to series A in the market today is about 7% over a three-year period. Virtually all these companies are failing, and the ones that get there are stumbling to that three-year mark with not a lot of cash. We're not investors. We’re company creators, co-creating with experts and passionate entrepreneurs and researchers and scientists, with the benefit of de-risking it, because we know it's a real problem. That’s really the benefit of the consortium. We're trying to look at it through a founder's lens. So even though we do three to five companies a year, we look at each one as an N of one like the founder does, and we're trying to bat 1000. So far, we're on that track with admittedly a small cohort of five companies. But all of them have gotten past series A except one, which we expect to get there this quarter or next quarter.

HCI: Can you describe one of those companies? 

Brozinsky: Sure. First, the five theses that we're operating on right now in building companies are:
• Reimagining the consumer and patient experience, 
• Automating critical workflows in healthcare. 
• Amplifying and creating new vital signs. 
• Unlocking real world data to accelerate real world evidence for better treatment and care. 
• Helping health systems recapture more share of care. We haven't built a company under that thesis yet, and that's about looking for where there's broken continuity of care, and it also correlates with very high-margin revenue being left on the table that health systems can recapture. 

So I'll describe one in that new vital signs thesis. We've got a company called Optain, which involves retinal imaging plus AI plus the science of oculomics, which is using the eye as a window to systemic disease. It's basically able to look at the vasculature of the eye, and with AI, you can amplify the vasculature of the eye, which maps to the health of the rest of the body. They've taken that camera used by the ophthalmologist and shrunk it down about 80% in terms of cost and size. Now we can put it into primary care or a kiosk in the mall. The company thinks about this as a new vital sign. 

HCI: There was some news that came out today about Aegis Ventures hiring John Driscoll and Chris Ross as partners. Will their role be identifying new opportunities? 

Brozinsky: They are in the role of venture partners. We have very accomplished people who are passionate about healthcare, who bring unique perspectives from their experiences. Chris, having been the longtime CIO of Mayo Clinic, brings a real deep understanding of health systems and implementing technology. John has run numerous businesses as an entrepreneur, and sold his last one to Walgreens and spent time as No. 2 at Walgreens. So he brings the big retail health perspective among others. They join a number of other venture partners that we have such as Steve Gipstein, who used to run healthcare for Apple. They have lots of ideas about what's wrong with healthcare and the solutions.