How Clover Health partnered with MindMate app to help elderly with brain games

GamesBeat: Do you see much activity being pushed by the insurance companies in this area?

Wallaert: At Clover we try not to over-anchor on what other insurance companies are going, because other insurance companies — I don’t think that insurance companies have always done a good job of focusing on member-centric profitability through health outcome models. But that’s one thing I like about Medicare Advantage. I do think, at least in the Medicare Advantage space, you will see insurers start to lean in.

I hope they choose to follow our example, because there’s nothing but net good for our society for everyone to get together. Insurance companies, game designers, everyone involved can lean into making people able to live a healthier, better life.

GamesBeat: Like a wellness program.

Wallaert: Yes, but embedding wellness in everything. I think wellness programs, explicit wellness programs, certainly have their target audience. But I have a three-year-old. I don’t get to the gym as much as I’d like. When it’s explicitly about exercise, explicitly about that wellness program, I’m not sure it really gets us there.

Instead, I think it’s incumbent on insurance companies and the medical field to get where people are. Instead of saying, “Come to me,” say, “I’m going to meet you where you are. If this is the game you want to play, play it, and I’ll figure out how to get what I need out of it to be able to help you.” That’s incumbent on us. That should be our burden, not the member’s burden. The goal of Clover, the goal of good progressive insurance, is the idea that people are able to do the things they would normally do, normally want to do, and then we find a way to do them in the way that maximizes their health outcomes.

At Clover we collect personal health goals. In your own words, why do you want to be healthy? What’s important to you? If you say to me, “The most important thing in my life is grandkids,” great. That’s how we’re going to orient your care. We’re doing a thing right now around flu shots. If that’s the thing that motivates you, we’ll harness that motivation for a health outcome that’s good for you and that you care about. We’ll connect those things together.

The member’s self-expressed health outcome, health goal, should be the primary focus of medicine. That’s where we see insurers going, or at least it’s where we’re going. Understanding what people mean when they say they want to be healthy, what their motivation is for that, and then honoring those motivations. Doing the work ourselves to design an environment that takes advantage of those motivations to improve their health, that helps them be where the want to be, how they want to be, in the right way.

GamesBeat: How far do you want to go toward clinical studies showing that this has benefit?

Wallaert: I’d certainly be open. We have a chief science officer, who I partner very closely with, and we’d be open to generating novel clinical data, a novel clinical study. That’s something we do at Clover and that we’re comfortable with. But I don’t think that’s the place to start. The place to start is the piece I was talking about: do members like this? Do they want to play these games? Are they okay with us having this data? Do they want us to give them feedback on what’s going on?

It’s less about, right now, whether this game improves people’s mental health clinically. It’s more about seeing if it fits into people’s lives. If it does, then we can start to understand if it’s good for them, if there’s clinical benefit. But the data is immediately beneficial. That’s something we can take and start to immediately tune into our algorithm to see what novel — is it additive to the model in a way that allows us to take care of someone in a better way? I’m open to a clinical study, but that’s not the standard for me. The standard is, can I derive member benefit from it?

GamesBeat: Something like 60 million people at this point have dementia, right?

Wallaert: Dementia is a very clinical term. All of us inevitably, as we age, experience some of the neurocognitive declines of age. Whether that crosses all the way over into dementia or Parkinson’s, or if that stays on the other side of the clinical diagnosis line but still remains very real — either way it affects all of us. If we’re younger, well, we have family members. We have people in the world that we care about struggling with this. One of the reasons it’s so interesting is because it’s so ubiquitous.

GamesBeat: Is there a trove of data that you have over, say, one year or two years that you’d like to see from a member? How would you be able to make use of that?

Wallaert: Mostly we’ve focused on what we would call within-subject design. That means I’m looking at Dean’s scores over time and trying to attack Dean’s decline over time. Certainly there are between-subject studies you can do as well. As enough people participate and you can watch the decline of various people, that starts to get really interesting. Are there predictable declines? Do they map to demographics? Do they map to signals that MindMate has about people’s activity levels? Do they map to signals Clover has about people’s clinical data? We know what else is going on.

Again, as we talk about this, the brain is not super well-understood. Our ability to recognize and classify these diagnoses is new. By incorporating the data with other data, you can do really interesting things. For example, is dementia worse in the north or the south? Are there demographic considerations? You can get hyper-local. Is dementia getting worse or better over time?

One of the problems we often have is that a lot of data sources are clinical data sources by nature, meaning someone gets a diagnosis of something. But when you do long-term longitudinal–we have this problem with autism. More people today are diagnosed with autism than were 50 years ago. But is that because more people are born autistic, or because we’re better at recognizing what autism is, and earlier?

One of the great things about something like MindMate data is it’s not a diagnosis. It’s a measurement. Let’s pretend we did MindMate for 10 years, you and I, across a large sample of people. I can see, in general, if people’s reaction time is getting better or worse. In general, do they have better or worse memories, and at what point in their lives? Because we’re not relying on a clinical diagnosis. We’re relying on raw data that, in theory, would power a clinical diagnosis. But as you and I know, culture and gender and age and race, all of these things play into how often someone is diagnosed.

GamesBeat: One question some people might ask: If there’s someone with very advanced dementia, a very advanced condition, is it possible to reverse the clock and show improvement that takes them back to an earlier stage?

Wallaert: With the evidence that we have, at least, and the treatments we have right now — turning back the clock, at the moment, is hard. It isn’t something where there’s abundant evidence that we have great treatments. Now, there’s certainly hope. I’m glad people are doing the research. It’s not my research or my interest. For me, right now, I have a plan full of predominantly over-65 folks, in terms of Medicare Advantage. I need to find them solutions for where they are right now, so I lean into things like social support.

Let me put it a different way. There’s one version of the world in which — let’s pretend that mental function, neurocognitive function, can be rated between zero and 100. Zero is bad and 100 is great. When you ask me about turning back the clock — Dean is at 65, can we get him to 65? — that’s one question, and there are great scientists looking into that.

I look at it a different way. Let’s pretend there’s a different scale, which is the quality of life scale, zero to 100. Dean is at a 60. Even if his mental function is declining, if I intervene in the right way his quality of life can go up. If he goes from living alone in a poorly supported system for neurocognitive decline, I can help get him into a situation, an environment, that is going to substantially raise his quality of life. Even if I can’t halt or reverse the decline itself, I can make sure he doesn’t have to have a continual decline in quality of life.

That, to me, is the interesting thing, the thing we can do today. Everyone, everywhere can help make the quality of life for people who are experiencing neurocognitive decline better. We don’t have to wait for drugs and clinical trials. We can do that, all of us together, today. That’s the most interesting thing to me. It’s the dominant aim of my team, the people I work with. Clinical change can take a very long time. Science moves very slow. But we can improve quality of life right now.

GamesBeat: If we take care of this part, maybe that whole immortality thing will work better. If scientists can get us to living a lot longer, it’d be good if they could get our brains to last too.

Wallaert: One of the great tragedies of medicine, I think, is that we’ve made substantial improvements in how long we can keep your physical body in a place where you can have a great quality of life. We need to invest in making sure that your mental quality of life, your brain and your emotions, can match that longevity. Otherwise, I always think about Ray Kurzweil. “I’m going to eat a hundred pills and live forever!” Yeah, but with what quality of life? It’s not just maintaining the ship. It’s maintaining the captain.

GamesBeat: There’s a whole other side of this in all the people, especially in Silicon Valley, who are interested in all that brain-sharpening stuff. Taking their loaded caffeine pills, nootropics and all that.

Wallaert: What was the one they were doing where they were dunking each other in super-cold water? They come up with something new every week. There’s a reason I’m not in Silicon Valley. I was a first-generation college kid from rural Oregon. I’ve gotten to do things in my career I never thought I’d get to do. But there’s a reason I’m at Clover. I’m just more interested in how I can make life really good for the average older adult.

I’m a behavioral psychologist, a behavioral scientist, in tech. There’s very few of us. As you can imagine, I get a lot of weird job offers. People like these nootropic companies come and say, “Be our chief science officer!” No. No interest. I’ll be right here doing this.

GamesBeat: I’m getting more tech pitches that have to do with older people. That never used to happen, which is interesting.

Wallaert: It’s an invisible population. Everyone wants to sell stuff to 18-year-olds on their smartphones. People lost track of the fact that there’s a wider America. Parts of the country still don’t have good internet access. Things are built for Silicon Valley, which has gigabit internet. That’s not the reality for a lot of people. In Oregon I had dial-up internet until, I don’t know, 2010 maybe? We need to work around these constraints.

GamesBeat: I just hope this isn’t all because someone’s marketing deck says, “Look how much money we can make from aging baby boomers.”

Wallaert: [laughs] I think it will be for some. I think that’s where business models come in. When I look at job offers, the first thing I look for is, is the business model aligned with the good of your users? If it isn’t, you’re inevitably going to run into trouble. One thing I love about Medicare Advantage is it’s one of those rare places where everything lines up. We only make money if we improve people’s health outcomes. That’s the only way, literally.

By constructing the business model that way — we watch people come into Medicare Advantage and then leave. If they were doing it only because they thought they could work some math and make a bunch of money, those people bow out pretty quick. The only way to make money long term in Medicare Advantage, because of the way the system is set up, is to improve health outcomes. It’s a special place to be.