The Role of Data + AI in Healthcare Equity

Read Interview

Speaker 1:
Welcome to Champions of Data and AI, brought to you by Databricks. In each episode, we salute champions of data and AI, the change agents who are shaking up the status quo. These Mavericks are rethinking how data and AI can enhance the human experience. We’ll dive into their challenges and celebrate their successes, all while getting to know these leaders a little more personally.

Chris D’Agostino:
Welcome to the Champions of Data and AI series. I’m Chris D’Agostino your host for today’s episode. In this episode, I’m joined by Slawek Kierner, SVP and chief data and analytics officer at Humana.

Chris D’Agostino:
Healthcare is increasingly turning to data to improve the services it provides to its customers. Slawek provides his personal views of the role of data and AI in healthcare equity, how the Lakehouse architecture can enable key insights while still protecting patient information, and finally, on a personal level, he’ll share his views on how leadership coaching helps him and his team achieve big goals. Let’s get started. Slawek, welcome to Champions of Data and AI.

Slawek Kierner:
It’s a pleasure to be here with you, Chris.

Chris D’Agostino:
Yeah. So when we met the first time, it was really fascinating because we talked a lot about coaching and mentorship, in addition to just the big challenge you have at Humana, to run the data and AI function for the higher organization.

Chris D’Agostino:
Wanted to learn a little bit more about your passion around coaching, and mentoring, and developing other leaders within the company, and how do you tackle that on a daily basis?

Slawek Kierner:
It’s a great question. And as you know, our domain is evolving so fast, but we all need to have a learning mindset. Nobody knows it all anymore. I mean, the whole space of AI and data and cloud got so massive that we just need to appreciate [inaudible 00:02:08] amount of skill.

Slawek Kierner:
And at the same time, we as leaders for this space in our enterprises have this role to help others grow and evangelize. And when it comes to that and to the leadership style, I’m always a believer in the servant leadership, in promoting by learning culture. And finally, the fact that leadership actually equals results.

Slawek Kierner:
So I use various formats. We started simply by a lot of reverse and personal and team mentoring, including all of our management team. So they all got sessions with me personally, but I leverage, of course, my team to deliver a lot of this learning.

Slawek Kierner:
Then I think a lot is done by symbols that leaders being. So literally, the first few months, when I came to Humana, I took my team and my peers to all kinds of conferences, at that time still in-person, the Build, the Data and AI Summit, the important moments where data science community meets, and just encouraged people to get on this learning journey.

Slawek Kierner:
And now we matured a lot. Now it’s all different. We have bespoke trainings where they’re built for various career paths for analysts and data scientists. We run our regular talent reviews. We have internal conferences actually, that bring thousands of people. We have competitions, hackathons. We have a rotation of our analytics talent, a bespoke program to promote rotation between different units. So the data scientists and analysts exchange perspective.

Slawek Kierner:
And we promote. This is an important thing. That’s part of our values and this is how we reward people. And this kind of technical mastery is important.

Slawek Kierner:
And then the final point is I also stretch myself to run a very flat organization. So I have many direct reports and they have bespoke time in my calendar every single day, to connect with them, and to also connect with my peers.

Slawek Kierner:
Because I learned through my time that technology’s an important aspect of course… I just talked about that… But the people aspect is so important. And the soft skills that you need to have to ensure that your team is motivated and moving on and growing. And frankly, also that you have permission space from all of our executives to drive a transformation that we all need to drive in our enterprises.

Chris D’Agostino:
Yeah. And I’ve talked to a lot of leaders. There’s definitely the role of when you’re leading the transformation… I think if it is managing up and managing down in terms of the people that are your peers, or senior executives, that you need to continually update, and convince them that things are on the right track, or make investments if things are not moving as you want.

Chris D’Agostino:
But I love what you’re talking about, where you’re working with your direct reports and the broader community within your team about up skilling, building the confidence of the team. The rotation sounds like it’s an interesting model to give people a different perspective on the other work that’s being done.

Chris D’Agostino:
And oftentimes gaining that perspective, they’re willing to be more collaborative because they have an appreciation of the difficulty of some of the other tasks.

Slawek Kierner:
100%. And this collaboration is so fundamental, because, as we all know, data models and reports, analytics becomes so much more relevant and precise when it’s based on more different types of data.

Slawek Kierner:
So if you bring this data democratization, the culture and understanding of what data is available also through rotation, you catalyze this movement. So that’s part of what they do.

Slawek Kierner:
And I actually even went on a bespoke coaching training at some stage, went through all of the certification to learn how to coach people. And I use it quite a bit also to help me advance this type of culture.

Chris D’Agostino:
From our last conversation, you actually have a coach yourself. And why don’t you share with the audience a little bit about that? I mean, you’ve gone through the certification. You know the process and the tools that are used. You apply them for your own staff and other members of the organization. But then at the same time, you take time out of your schedule to make sure that you’re getting feedback and some coaching. So can you share a little bit about that?

Slawek Kierner:
Yeah, 100%, that’s what I do, just because I’m a believer that all of us have an opportunity to develop and grow. And how it works, we talk a lot. So it’s clear. Every two weeks or so my coach gives me a lot of homework, so it’s time with him, but also, ideas emerge that motivate me to take maybe a bit different action than I would do otherwise.

Slawek Kierner:
And it’s all about getting faster to results. I deeply believe that through these conversations, through helping me see the world… Thanks to the questions the coach asks me, it helps me see the world through different perspective and broadens my understanding, and helps me accelerate on the mission that we have.

Slawek Kierner:
So essentially, I’m a deep believer that it is not important what level you are in organization. I think coaching could help you, and looking at yourself from broader perspective, getting this kind of confronting questions from the coach gives you additional perspective and ultimately leads to faster, better results.

Chris D’Agostino:
Yeah. Prior to Databricks, I was at a Fortune 50 company, and as an executive there, I had a coach assigned to me, and he was fantastic. And I think the thing that I learned, as someone with a background in computer science and engineering, learning some of the soft skills and how to better work with other executives in the organization, translating the culture into how I communicate and how I behave within the organization was important.

Chris D’Agostino:
And just some really good professional lessons learned, not quite at the stage of being able to certify myself as a coach and apply it, but I certainly try to mentor people. And here at Databricks, there people within our field engineering group that I spend time with mentoring, to be able to have more elevated conversations with executives like yourself.

Slawek Kierner:
And I believe it is so important, because you bring this very relevant experience from an enterprise, but as a user of technologies that Databricks now develops. So having someone like you internally in the organization must be super helpful to the product teams and leaders that need to develop new features, help prioritize, and understand what actually is relevant from enterprise point of view.

Chris D’Agostino:
Let’s switch gears now a little bit. When we talked last time, you said something really interesting, and this was true in my past role, where the primary function of a company was not directly in the technology space, but yet technology was so critical to its success that we described the company as a technology company first, that happened to offer up a service or a product that most people wouldn’t normally identify with as technology.

Chris D’Agostino:
And Humana’s the same way in that you describe it as a technology company, but when somebody goes to the website, they’re really there visiting to figure out what type of health products are available to them.

Chris D’Agostino:
But can you talk a bit about that distinction where you’re really trying to drive the mindset within the organization to think from a technology standpoint?

Slawek Kierner:
Yeah, well, it’s a good question. And truly so, we of course are a healthcare services company, and a company that has a culture of innovation. If you look into a history of Humana, so much change has happened over the last 40 years or so.

Slawek Kierner:
But for now, similar to, I believe, any company out there, in order to be successful, we need to be on this journey to transition to be more technologically intense, and infused with data, with analytics, to be able to, in the end, achieve our mission to help our members live better, healthier lives.

Slawek Kierner:
So technology’s super important, and I think it brings value on a couple of levels to the company. First for members, it enables that there’s differentiated, personalized, simplified healthcare experience. And that helps us achieve some awards in terms of a better experience perspective.

Slawek Kierner:
Then for associates, there’s so much opportunity. And we see it in a couple of new models and proof points and processes we’ve already transformed, to eliminate burden, to eliminate mundane tasks and replace that with AI.

Slawek Kierner:
Simple things like we get thousands of faxes every day with people’s medical history, to substantiate the approval process, but most faxes so far have to be manually attached to cases by nurses. So nurses… And we of course are facing a shortage of nurses in our society… To have to spend time on this kind of mundane administrative tasks.

Slawek Kierner:
Well, right now we have an AI algorithm that digitizes those faxes, scans, does optical character recognition, then matches to the right case. And a lot of this is eliminated.

Slawek Kierner:
So associate’s get better experience, and finally, enterprise accelerates. And also, an important stakeholder for us, of course, are our investors, and we observe a market and see how those companies that are technologically advanced are getting a better evaluation.

Slawek Kierner:
Because ultimately, the market believes, and I do as well… I’m sure many of us here and the listners also do… That there is all the future in front of data and AI, and those companies what master those will eventually have a faster and more profitable future and more growth.

Slawek Kierner:
So that’s how I look at this. And I believe we have to move aggressively. We are moving aggressively into this space. And there’s a lot that technology can bring to healthcare.

Chris D’Agostino:
Yeah. And if you think about healthcare in general and the role that data and AI play in terms of equity… I think of health as one part genetic, one part environmental, one part lifestyle. There are other factors, but those are the ones that come to mind for me.

Chris D’Agostino:
And now most people have access to devices that can track the number of steps that they take, their heart rate. Curious about what you think the role of data and AI is for healthcare equity, and what some of the challenges might be.

Slawek Kierner:
Yeah, that’s a very, very important topic, especially in this moment of transformation. Setting these things right is such a critical component. And as I said, AI and machine learning and data can deliver more easy and seamless healthcare experience, which we all want from our healthcare providers, but it can also reflect and magnify the unintended biases that are there in the historical data.

Slawek Kierner:
So it’s important, I think, to tackle, but then I’m a huge believer that it is way easier to remove unintended bias from an algorithm than it is from thousands of individual decisions of different constituents, providers and humans that are making these decisions today.

Slawek Kierner:
So for this reason, Humana is committed to developing and fostering this more equitable and inclusive AI by a couple of things. We are examining our models for bias and we’re taking appropriate actions, actions that include our principles, which we have formal six principles of our AI development. Everybody can see on our websites. So they are public.

Slawek Kierner:
In mid 2020, actually, we also signed an equal AI pledge. That’s a pledge that holds us accountable to make sure that every model is tested for unintended bias, and that we take appropriate actions, and that we even expect our partners, our vendors that help us deliver our service, to do the same.

Slawek Kierner:
So that’s key to our mission. And maybe one more point is, it’s not just important to do it once. Of course there is ongoing monitoring that’s going on of all of these models. Because drift can come in and the unintended bias can creep in as we are in development.

Chris D’Agostino:
That sounds like a really nice, transparent way of doing the underwriting process when you’re assessing individuals for healthcare coverage and things like that.

Slawek Kierner:
It is, yeah. And that’s one part of it. There are a lot of decisions, also Chris, that happen in the company, that are around our proactive care management and outreach, which by the way… I think maybe we have a chance to talk about this a bit later… Is the greatest opportunity, the move from reactive to proactive care, and therefore, this huge role of AI to decide to whom to outreach and for what reason, and how to help the person. So these actions need to be monitored for bias.

Slawek Kierner:
And I know there is a lot of history in the financial services, but what I learned in healthcare is that this topic is very different in this industry. So every company needs to answer it for themselves, whether you’re technology providers, the topic of AI equity is different in financial services.

Slawek Kierner:
I know that a lot of work is there around removing those potentially important features from the models, like age, or gender, or race, that they shouldn’t be steering your underwriting or your credit decisions.

Slawek Kierner:
In healthcare, we cannot do that. I guess a lot of your healthcare diagnostics, progression of disease and the way you are treated actually does depend on age. We treat young person differently older people. They do depend on race, frankly. There are some diseases that actually progress differently depending on that. And they do of course differ depending on gender. So not simply from the same diagnostics.

Slawek Kierner:
So we just cannot remove those. We need to learn how to deal with results of the model and make sure that everybody has equal opportunity, while we still take into account all of these important facts.

Chris D’Agostino:
Yeah. And so that raises the next issue for me, or topic, which is around data architecture and the role the architecture plays in governing data. As you say, you’ve got to include a lot of this very sensitive information that many organizations and other verticals really aim to remove from the datasets, to prevent bias.

Chris D’Agostino:
But in healthcare, these factors actually contribute to, as you say, disease progression, the protocols that are used. So can you talk a little bit about the architecture? We spoke a bit about the Lakehouse architecture of combining the governance and structure associated with data warehousing, with the flexibility of a data lake and looking at different data types.

Chris D’Agostino:
How does the Lakehouse architecture, in your mind, help contribute to governing the data and making sure those datasets are good quality and used for the right use cases?

Slawek Kierner:
So of course, as we talked earlier, I’m a deep believer in investible architectures for a couple of reasons. So one, we of course have a lot of data already, and there will be more and more, as you already indicated, Chris.

Slawek Kierner:
There is devices that people carry, the virtual care, telehealth growth, and all of this will continue to more and more datasets, which are best captured for machine learning purposes within data lake type of architectures.

Slawek Kierner:
But then we also quickly find out that, for the purpose of runninganalytics, KPIs, managing the various processes internally, you do want to have structured data products, which are governed, and you need to be able to marry both, because for the purpose of data quality, you don’t want to have separation.

Slawek Kierner:
I believe that actually having, in one architecture, both a data warehouse and a data lake, and from the same datasets operational reporting which gets looked at by executives and from this gets this feedback loop of quality.

Slawek Kierner:
If the same data set has then used for machine learning model, of course, we kind of have better guarantee that those models are relying on high quality data. So for this purpose… And I see a lot of future in continued development of those Lakehouse type of architectures… That’s what we have right now at Humana. And we see tremendous pickup of it.

Slawek Kierner:
So almost a hundred machine learning models that are way more precise. We have almost 20,000 users of our analytics reporting platform, that sits on top of data warehousing capabilities within this structure. So huge future in front of it.

Chris D’Agostino:
Yeah. And as you look back on your career, and having been involved in different transformations… I was on a call earlier today, and one of the executives said they feel like they’re in the next wave of innovation in architecture, where enterprise it was data warehouses in the 1980s, and then it was Hadoop based ecosystems in the mid 2000s. And now this Lakehouse concept.

Chris D’Agostino:
And said there always seems the cycle of the architecture is evolving. And he said he wonders how long before it will be the next thing. So the technology definitely keeps changing, but when you see these transformations take place, can you talk a bit about how the organization, the non-technical aspects… Has there been consistent things in the transformations that you’ve been a part of or have led that you see as being critical to companies being successful?

Slawek Kierner:
Yeah, for sure. I see a lot of concepts actually repeat themselves, and I saw from my past from more than 20 years ago, where I was at Procter & Gamble, and had the ability to bring that technology transformation to supply chain and factories.

Slawek Kierner:
This is when we were moving from pneumatic to database control and bringing some first albums where a lot of these learnings about process management, and feedback, and data quality, and signal, and distinguishing signal from the noise are useful right now.

Slawek Kierner:
And then I was running biggest database in Europe at that time for a longitudinal view of consumers. We had more than 80 million of those, with very interesting of course, progress at that time when Europe was making privacy and this legislation.

Slawek Kierner:
So again, that is useful to have, and this concept of how you think about data protection and how do you work with privacy executives, with risk, with your view, your seesaw. That certainly is something that I use quite a bit.

Slawek Kierner:
And then with Microsoft, the transformation that Microsoft went through is of course amazing. It was great to be part of it with a company under the leadership of Satya, and be able to co-create some of the fastest growing platforms and products, and learn how to think about agile software engineering, the KPIs, the metrics that you use for platforms, the network effects in platforms.

Slawek Kierner:
And we use a lot of this right now, internally at Humana, as kind of a Lakehouse concept. So what we call internal enterprise data platform and enterprise analytics services, as well as our machine learning platform.

Slawek Kierner:
That becomes our internal platform, that creates the network effect of getting more data and for this, better quality insights and better precision of models, and therefore more intervention as models drive and more data comes to the platform itself.

Slawek Kierner:
So to your point, I think concepts change and architectures change, evolve, I think, for the better, because technology progress of cloud computing has enabled that. But at the same time, a lot stays the same in context of transformation.

Chris D’Agostino:
So Slawek, I know from the last time we talked, you have a very important job. You’ve had a career at Procter & Gamble, Microsoft, now Humana, probably pretty some stressful days.

Chris D’Agostino:
And I know sailing is an important thing for you, to go out and enjoy being on the water. Can you talk a little bit about the type of sailing? Because I think the audience will find it fascinating.

Slawek Kierner:
Right, Chris, so you’re right. And I cannot wait for the time when we can be with bigger crowds of people in a more confined space, like a sailing boat.

Slawek Kierner:
I use it number of times in the past, where I took my team on the water, on the lake, on the sea here. But it’s something I haven’t been doing for the last two years. So really looking forward to that. But I understand that you also like all kinds of sports outside. So is this boats, or what would you like to do, Chris?

Chris D’Agostino:
Well, I love boats, and I guess since you’re in Boston, there’s probably, what, like two weeks out of the year you can sail, that it’s warm enough? Otherwise it’s too cold.

Slawek Kierner:
Maybe four.

Chris D’Agostino:
Maybe four. Perfect. So let’s hope everything is kind of back to normal as soon as possible so that you don’t miss those four weeks of warm weather there.

Chris D’Agostino:
For me, I love being on the water as well, of course, but flying and being up in an airplane and piloting a plane is really fun. It’s interesting because, similar to our jobs with trying to build systems that are precise and produce good quality results, the style of sailing you do, and the team-oriented aspect of that, or in my case, the idea of piloting a plane and following a flight plan, and needing to stay compliant with all the rules and regulations for altitude and direction and all these things, it keeps you on your toes, and it’s a lot of fun, and it’s certainly a stress reliever, for sure.

Chris D’Agostino:
So let’s shift into wrapping up here. I think there are two things that I’d love to hear a little bit more about. The first is, given that coaching is so important to you, and the success of your team, and the success of your career from a personal level, is there something you can share with the audience about something that you’re continuing to work on?

Chris D’Agostino:
This continuous improvement is always a key theme whenever you and I speak, so share with us something that you’ve sought after some coaching and really are trying to work on and focus on in 2021.

Slawek Kierner:
Sure thing. And there’s probably not one. There are many things. So let me give you two or three. So I think one thing consistently from my career… Because I was born as an engineer and a data scientist… It was about communication. And that probably is even amplified because, as you of course all can hear, I speak with an accent in a foreign language.

Slawek Kierner:
So I spent a lot of time on developing my ability to communicate at various levels, from communication in bigger forums, smaller forums, with a board, with other executives. So that has consistently been a part of my coaching, and training and development.

Slawek Kierner:
Then technology is always on the list. I always try to spend some time in an interesting conference, some time reading, actually including also reading journals and a couple of articles a month, help me keep my knowledge current around the developments in the space.

Slawek Kierner:
And then finally, my biggest fascination of the last two years has been healthcare. Healthcare was totally new to me. It’s a fascinating field. It’s way more complicated than I thought, coming in here.

Slawek Kierner:
So I’m still… Might be crawling, not yet running, but I think I know enough to be dangerous and certainly have a lot of fun and pleasure spending time both learning and working with my colleagues that are some of the best healthcare professionals in the US.

Chris D’Agostino:
Yeah. I mean, I like that you talk about stepping into a role, and such a big role at that, where maybe you’ve walked into that role feeling pretty confident around your technical acumen, but maybe the vertical itself was a little bit scary because you didn’t have a long history in healthcare.

Chris D’Agostino:
So that seems to me to be a really nice… This willingness to take a risk, and knowing that you believe in yourself enough that you can rise to the occasion, I suppose. Can you talk a bit about what advice you’d share with people that are aspiring for a role like yours inside of an organization?

Slawek Kierner:
Yeah, so what I would say… You want to be in a place that helps you practice at the top of your license, which is a term actually, which is very common in healthcare.

Slawek Kierner:
And what it means for data science and analytics professionals is you want to be in a place where there is a lot of data, where there is a lot of information to be processed, where there’s access to modern cloud technology, and some form of permission space, to advance methods and new sophistication and innovation.

Slawek Kierner:
So Humana is a place like this, and I’m sure there are a few other as well, but choose for a place where actually you have access to a large amount of data.

Slawek Kierner:
And then finally, another thing is, you want to be in a place that has good values, that has good ethics, that has good purpose, that motivates you to spend the extra time, go an extra mile to learn, grow and create change, create transformation with your data analytics, data science, because you clearly can drive.

Chris D’Agostino:
So to sum it up, head someplace where there’s a lot of data so that you can really gain some good insights, find an organization or a cause that you’re really passionate about, and motivated to put in the extra time in order to maybe change the world a bit.

Chris D’Agostino:
And I would say in 2021, being in healthcare, there probably hasn’t been a more fascinating time to be a technologist in this space, for sure.

Slawek Kierner:
I agree. And if I can just add a tag on, I think only started. Honestly, 2021 was this kind of inflection point and acceleration. It gave us more data and more recognition in the industry on how important real time data is.

Slawek Kierner:
It changed member habits, consumer habits. There’s more of a digital and virtual care. It then positioned us to create more precise models that can turn healthcare from reactive to proactive. And this is where the fun starts.

Chris D’Agostino:
Slawek, thank you so much for being on Champions of Data and AI. Appreciate your time. It’s always great talking to you.

Slawek Kierner:
Thank you, Chris. It was a pleasure. I hope to see you somewhere on a boat, or from a boat in a plane. That would be [crosstalk 00:32:20].

Chris D’Agostino:
I’ll be flying overhead.

Slawek Kierner:
Yeah. Take care.
Speaker 1:
Thank you for joining this episode of Champions of Data and AI, brought to you by Databricks. Thousands of data leaders rely on Databricks to simplify data and AI, so data teams can innovate faster and solve the world’s toughest problems. Visit databricks.com to learn how data leaders are unlocking the true potential of all of their data.