Corporate Innovator: Roberto Roitz, Director of Innovation at Humana

Roitz has been with Humana for five years and oversees its strategic partnerships

While most entrepreneurs want to be the one to discover the next Amazon or Twitter, oftentimes major technological shifts are coming from the big companies, the players that have been on the scene for years, if not decades. Those companies have survived because they know how to pivot. They're the ones who either seed new ideas or acquire them and distribute them. 

In this column, we talk to those companies and their innovators who are preparing them for what's coming.

Healthcare is one area where advances in technologies like big data, artificial intelligence and machine learning are having a major impact on how the space operates.

Humana is one healthcare company that has been making big strides, through both strategic investments and acquisitions. For example, the company has placed strategic investments in Omada Health, Livongo Health, Aspen Health, ClearData and recently acquired Kindred Healthcare, Inc.

It has also launched programs like Humana Behavioral Health, which helps members receive behavioral health and substance abuse services, as well as Go365, its wellness and rewards program that encourages members to make healthy lifestyle choices by rewarding them for completing certain healthy activities.

I spoke to Roberto Roitz, Director of Innovation at Humana, about how the company views its partnership strategy, how AI has been revolutionizing healthcare and some of the technologies that he sees making a big difference in the next few years.

VatorNews: Tell me a little about yourself and your role at Humana.

Roberto Roitz: I’ve been with Humana for five years now, my role is called Discovery and Strategic Partnerships. Our vision is to become the Partner of Choice through a best-in-class strategic partnership program. From a Discovery point of view, we see it more as focusing on the front end of the venture pipeline, where it’s more about scouting, looking for opportunities, looking for startups and companies that align to our strategic priorities that we can bring to Humana for test-and-learn opportunities. We help entrepreneurs navigate Humana -- find the right internal sponsors, and find ways to determine the right use cases.

On the Strategic Partnership side it’s thinking more about solving challenging problems in new, innovative ways with non-traditional healthcare partners or tech companies. For instance, you have all the big players, like Google, Apple, Amazon and Microsoft that are starting to build a point of view or perspective on healthcare. Some of them have said that healthcare is going to be a big priority for them. As a company, we are looking to find new approaches to develop best in class analytics capabilities, insights and applications focused on health and behavior change. Other opportunities could include, explore ways to deliver personalized health and identify opportunities and insights around AI/ML.

I am in a new area called Digital Health and Analytics (DH&A). The vision of DH&A is to bring an enterprise-wide approach towards analytics by developing new tools and technologies that will enable Humana to deliver a more personalized and differentiated healthcare experience. More data and better analytics are two key differentiators to make care more relevant by tailoring healthcare experiences and interventions. Digital also plays an important role into this equation, as a company we believe that by delivering new digital experience we will be able to empower our members to take better control of their health, and help them slow disease progression. The whole perspective is: digital is coming, and we need to be ready to offer new solutions and services that are powered by analytics and deliver through new digital experiences. It’s about finding that sweet spot where data and digital intertwine to drive higher enrollment and engagement and build more sustainable digital solutions for our members.

VN: When you scout for these companies, what stage are they in? Are they seed, Series A, Later stage?

RR: Most likely these are companies that are between Series B and Series C, so companies with product(s) that likely fit a Humana need, and with intent to build a product “for” Humana that they can commercialize broadly. That’s the type of company we look for when we engage in pilots and tests. On the seed side, it’s more about understanding what trends are coming, what new technology and services are entrepreneurs thinking of, and how we can advocate for entrepreneurs that are working on far adjacent and transformational spaces. Scouting applies to both Series B and C, and even early stage companies.

What it’s important for Humana is understanding the landscape and the changing dynamics of technologies. It’s building a perspective around specific spaces that we might not be seeing, as a company, and that we need to be ready for. The whole purpose of scouting is building points of view and understanding the nascent players, the emerging players, but also understanding which are the more mature players in that specific space.

With seed companies, we want to act as the partner of choice and find ways to help entrepreneurs pressure test their idea. We do that by taking a knowledge base approach. In essence, connecting entrepreneurs with thought leaders, operators, physicians, etc. so that they can understand challenges, pain points, and approaches that could help them inform their strategy and road map. Sometimes we even engage in small pilots to test viability and desirability.

VN: When Humana does decide to enter into a partnership, what are you looking for in those companies that makes you want to partner with them?

RR: On the startup side, the model varies based on understanding the right use case(s). In essence, you might have strategic priorities that you’re going after and you want to find companies that are aligned to those priorities, but, beyond that, we need to define the right use case(s) for us. We start always with the use case. Once that use case is defined, we engage in a test and learn. The use case encompasses not only the problem we are trying to solver for, but also the learning objectives, metrics and expected outcomes we are looking to accomplish from the pilot. Learning objectives are intended to help both parties learn from each other as we engage in the pilot and try to solve problems in new, innovative ways.

Once we roll out the pilot, it’s all about learning. As we learn, as we get new data points, as we get new insights, we then decide whether to move to phase two. You start in gate one where you want to learn, you frame some objectives, you want to prove some hypotheses. If you prove that, if you see some good results, then you think about phase two, where you start thinking about, ‘We have proven that this is successful, how can we think about now making it scalable? Scale is the challenge; you always want to make sure that at least you can support these companies and help them scale. That’s what it means to be a good partner of choice. It’s about solving challenging problems in new ways, in novel ways, that will allow us to reach scale.

If we move into strategic partnerships with bigger companies, that’s a different type of conversation and, to be honest, that’s something we are still framing and evolving on as we make this happen. In essence, it’s no different than when you engage with a startup. It’s really understanding, ‘What is it we’re trying to solve for? What kinds of capabilities are these companies bringing that we don’t have? Which capabilities can we bring that they don’t have?’ Within that intersection, it’s where you start building interesting partnerships because each company has a unique point of view. For instance, we have a lot of experience in healthcare, we have done that for a while and successfully, and these tech companies have been doing great things as well, at scale, in a different industry and market. It’s about understanding, especially as we think about the next wave of Medicare Advantage people, they’re going to be more tech savvy, they’re going to be willing to engage more with digital solutions, they’re going to be using their smartphones more. There is definitely a use case for us to get ready for that future, and how we prep for that future, that is honestly very close and near.

VN: What are some of the technologies you’ve seen changing healthcare in recent years?

RR: It’s an interesting question because there’s a lot happening in healthcare, for sure, and there’s a lot of traction. You see VC investments in digital health and in different areas within healthcare. While there are improvements being made so far in the healthcare system, to say that something has exponentially changed the system, or transformed this industry, I haven’t see that yet. This is my opinion.

Having said that, there’s a lot happening when you think about telehealth, which, by 2020, Medicare is going to be reimbursing these services as they would with a traditional in-person visit with a physician. There’s a lot of companies that are trying to take different approaches, whether it’s by expanding the number of services across Teleheath, whether it’s for behavioral health, preventive care, for coaching, there’s a different model used and applied. That’s going to be transformational for sure, especially as you think about in-home care delivery. As a company, we strongly believe in the at-home space as the focal point for value based care. There is an opportunity for us to deliver better and new services into the home, and that’s going to need a lot of technologies like telehealth, like remote monitoring, which is an interesting space too. How do I capture the right information for members in the home that somehow will allow us to bring and provide the right interventions at the moment that matters for the member? That’s an important thing because folks are not always thinking about their healthcare, but, when something happens, what’s when they need a response, that’s when they need support. So, having that information in place, if you think about the combination of telehealth and remote monitoring, that’s an interesting space for us. There’s a lot of companies trying to solve for that.

There’s also challenges around devices, whether people will use a specific device, whether people will engage, whether people will somehow engage in a way that will keep using that sensor, that device, to monitor their health, those are still things that we need to prove. But there are rich opportunities in those two spaces.

VN: How have you seen AI in particular change healthcare?

RR: As it pertains to AI and machine learning, absolutely it’s also a big opportunity. A couple of things are happening there; first of all, a lot of companies talk about AI and machine learning and they’re not necessarily doing AI and machine learning, it’s pretty much a buzzword, but there are actually companies that are doing it right and at scale. The question for us is, how do we think about AI and machine learning at scale?

There are a couple of models; there’s this trend where companies are focusing on the front door of healthcare. In essence, they are applying AI and machine learning where they’re trying to emulate a doctor’s brain, in the sense that when you ask a question to the member or patient, it’s a chat bot, that is AI-enabled. What they’re doing there is they’re in the process of triaging and understanding where that member needs to be referred to. So, making sure they’re referring the member to the right type of care. That front door of healthcare, where AI is playing a big role, that’s going to be very important because they’re trying to make the system more effective and efficient, and they’re making sure that, based on the information they’re processing, they can refer a patient to the optimal service. For example, ‘Roberto, based on this “diagnosis”, we’re going to refer you to this telemedicine solution because that’s what you need.’ Or, ‘Roberto, you actually need to go to a PCP or you need go to the ER.’ It makes a lot of sense because there’s a prioritization of ER services and a lot of folks going into the ER when there was no need to go to the ER. AI is playing a big role there.

On the other hand, AI and machine learning are playing a big role in everything that has to do with workflow optimization, process optimization, so more on the back end, where we can use AI and machine learning to help care managers or nurses perform at the top of their license. For example, now there is an AI solution allowing them to triage, make better decisions, and more informed decisions. They’re not supplementing doctors, they’re not supplementing nurses or care managers, they’re just providing a solution for them to perform at top of license. Then there’s a lot of AI and machine learning happening on the payment models, claims and prior authorizations. There’s a lot happening in these spaces in order to optimize processes and be more effective and efficient.

VN: Some see this technology as a way to help physicians do their jobs better rather than replace them. Is that how you see it as well?

RR: There’s a lot of debate on whether AI will replace a PCP or replace a doctor, and I don’t think that’s going to be the situation. It’s going to be a supporting engine for them to be more effective and efficient in the way they work.

It’s about making care less variable. There is a lot of variability in the current healthcare system, so, through AI and machine learning, you can reduce that variability. You can facilitate highly complex, evidence-based, medical judgments where AI can help, whether it’s making a diagnosis or whether it’s supporting that doctor to make a better, more-informed decision. In order to make that model so robust and make that model change that subjectivity, it’s very hard because you need a lot of data. As I’m sure you know, data in healthcare is pretty much siloed. The success of any AI or machine learning within the healthcare space is going to be based on the amount of data they can capture and, again, it’s operating at scale; the more data you get, the better you refine your algorithms, the more you can refine your decision trees and all that, to really improve that model.

VN: Tell me about Humana Behavioral Health and some of the initiatives that Humana has started to help its members. How do those work?

RR: On the behavioral side, there is definitely a shortage of psychiatrists and there is a big need. We have a big problem in the country around depression, loneliness, and that’s driving a lot of that need. The fact is, given that shortage of psychiatrists, it’s very hard for people to get scheduled into these appointments; it takes months for them, so there is an access problem and that’s the reality of the system. There are more and more of these solutions that are taking place that are virtual enabled, so they’re trying to add more capacity into the system by leveraging telemedicine services, for example, for behavioral health. That’s one portion; there’s also solutions around leveraging text. They have proven that using text is a good way to engage with members, provide the right kind of advice, help triage as well, and provide that kind of continuous support that these patients and members need.

Again, different approaches; one is around text and there’s companies that have the perception that text is going to be the way to go about this. There’s also those that think virtual is important because you want to see that interaction, the psychiatrist wants to see that member through a video to make sure they understand body language and changes for them to provide the best advice.

VN: What are some innovative growth areas that are priorities for you in 2019?

RR: Telehealth, remote monitoring, AI and machine learning at scale are big priorities. We need to provide a seamless experience because we have members that have multiple chronic conditions. So, if you think about point solutions, that’s not necessarily delivering a good member experience. If I have diabetes, CHF and hypertension, it’s very unlikely that I’m going to engage in three different apps or three different solutions. We need to find a model where there is that seamless experience being provided to the member that makes it easy for them to engage with the solution, where they don’t even notice that they’re interacting with three different apps or services. It’s just an integrated model.

We’re still in this journey, learning and exploring, and trying to find out how to make it happen. We have the right areas in place; it’s just now a matter of executing around these three areas.

VN: Innovation often starts off small, and it doesn't have great impact on a large company like Humana. Have you seen new innovations within the last five years that have had an impact on Humana as a whole?

RR: If I look back five years ago at where we were compared to where we are now, a lot has changed in the way we are organized and the way we’re thinking about bringing new capabilities and building new capabilities. It’s been a success and the way we have framed our strategy is one key leading indicator based on what’s happening in the innovation ecosystem.

Change has happened in the way we built our strategy, and we have a sound strategy. We have built a thoughtful process when collaborating with startup companies and tech companies, which is leading Humana to become the partner of choice to drive better and faster innovations.

I will close by saying, if you think about all these trends around AI and machine learning, all this is going to lead to more precise decisions. It’s going to lead to optimized use of physicians. It’s going to lead to lower costs. It’s going to lead to better outcomes and care. It’s almost like a shift from medicine as an art to medicine as a science. That’s the future. At the same time, the consumer experience will be improved. You’re going to start seeing things like data and peace of mind in care decisions. There’s going to be more consistent intake process. There’s going to be definitely faster, more accurate decisions being made. There’s going to be more information put into context.