Datica Podcast

August 17, 2015

Simon Jones - Vice President of HIT Product Strategy at Blue Shield of California

Catalyze: Thank you, Simon Jones, from Blue Shield of California, for joining us today for the interview. Going to be talking about a few different topics related to technology and healthcare behavior change. Then, of course, one of your favorite topics, Cal INDEX.

One of the challenges everybody faces, but obviously, as an insurance company, you guys face it much more acutely, is motivating behavior change in those that aren't already motivated. From your perspective, how do you think we bring in this other cohort of the population that right now isn't necessarily out tracking themselves and trying to make those practical changes in behavior?

Mr. Jones: It's a multi-faceted answer, and if I had the complete answer I'd probably be running a company that knew how to do it perfectly. One of the things I'm seeing is this cohort of folks that are motivated. They're inherently motivated, and they are going to likely do fairly well with or without the device, with or without the Fitbit, with or without a living scale or any of those type of things. It specifically is something that helps them track what they're doing, do a little better with it.

There's a cohort of folks that come sit on the cusp of that but need just a little bit of something, some of the gamification, some of the competition with it. That band of folks is going to benefit from something like a Fitbit or a Misfit or whatever, pick your device. Then there are the bulk of the folks who are not going to be motivated to change their behavior by what those types of devices or those types of things. One of the reasons I think it plays into is just us as human beings, right?

We're programmed to see the tiger that's about to jump out of the bush and eat us in the moment. We don't worry about the tiger that's about to be born that may come out of the bush and eat us 20 years from now, right? Thus we get obesity, hypertension, poor financial planning, a host of things. Folks tend to have different intrinsic motivations for things.

The industry has really tried to take a one-size-fits-all approach to a lot of this, which is, "Hey, we're going to give a Fitbit to people, and we're going to put a competition up online, and we're going to give them some points that they can buy a sweatshirt or whatever at the end of the day, and this is going to work for your Company X." Well, unless your Company X is selling world championship bicycles or something like that, you're not likely to have an entirely homogenous cohort.

How do you get to that intrinsic motivation that people bring to the table? How do you try to effect behavior change in different cohorts differently? You're going to end up with some of the folks like Omada, doing some great work in the pre-diabetes space. There is some discussion, with how much of that giving, it's just because the patients are opting into it or folks that would benefit now. If that's true, that's still great because they're really the company that's showing the most conclusive evidence of true behavior change.

The question now is, "How do you get to the folks that it's more challenging, more difficult for them to continue with a program at start?" I think we're going to take learnings from, in particular, the diet industry. Where Weight Watchers has been much more successful than some of the other programs. We have a long way to go because if you look at financial planning, and you look at weight management, and you look at things that have been around for a lot longer than we've been trying to do, use technology to effect behavior change in chronic disease, they haven't cracked the nut.

We're going to work hard on it, and it's going to end up, in my opinion, being a number of different approaches or, say, diabetes or pre-diabetic management. There's not going to be a one size fits all. There's going to be a multiple that's going to range from very low-touch gamification to really high interaction with individuals. You're seeing Weight Watchers start with the on-line coaching. They had the ads on the telly not too long ago about how we can help you with somebody who can be a coach that you can reach out to you. I think there's a lot of things that are happening in the industry that we could look to, but I don't think anybody has the answer.

Catalyze: Are you looking at it as a best of breed approach where there will be these multiple solutions? For example, some may target pre-diabetics; some may target progressed diabetics that have A1c's eight or nine. Are you looking at different solutions and offering those to members?

Mr. Jones: We are. Nobody is trying to claim that they have a one-size-fits-all solution out there in these spaces, so we're looking at best of breed and trying to look for people who have compelling evidence that is at least an early indicator of potential for success. There's few people who have studies out there, but people who we talk to who say we've been able to get people to sign up, and then we've been able to get people to persist in the program for 30, 60, 90 days. Those are good indicators.

We're still working and having a lot of the discussions around "How do you segment the population? How do you look at what the cohorts are and which are going to be the most efficient things?" We have a couple of years at least of significant experimentation across a breadth of solutions. Hitching your cart to just one horse in this is not going to be an effective strategy.

You made a point about this being a difficult problem. I would actually argue that this is the most difficult and the most important problem we have to solve, because we talk about technology in the sense that technology shouldn't be done for technology's sake. It should always be in support of a business process, a clinical process.

One of the rules we talk about in our group is It doesn't count unless you change behavior. It's not enough to get somebody to adopt something. It doesn't count unless you actually effect the behavior change. It's the last piece of it that we have to work on because, by and large, the technology is there. By and large, the knowledge base is there to do a lot of things. But the hard part is bringing it together in a way that it impacts a patient, a doctor, a person at the end of the day.

Catalyze: It maybe is the most important problem that we need to solve. It's also called the most challenging. I agree that the technology is there. There's nothing right now at least revolutionary from a technology perspective. It's more "How is that technology actually adopted and used by these groups?"

Catalyze: When you look at evidence as you're evaluating these options, you mentioned, people sticking with the program for 30 days, 60 days, 90 days, whatever a period of time it might be, as sort of a form of evidence that people are continually engaged, then looking at actual behavior change. Do you stratify evidence?

The step further is better outcomes and reduced costs, right? There's kind of this spectrum of evidence. Are we just seeing the very early stage, where all we really have right now is engagement metrics and maybe some basic metrics around behavior change?

Mr. Jones: In short, yes. We're seeing the very-early-on. When we look at it, and we look at success criteria of a pilot or a program, we're looking at metrics that are activity metrics and results metrics. When you're looking at the results of the outcome metric, you may be looking at improved quality of care. You may be looking at reduced cost of care. You may be looking at increased engagement, kind of whichever of the typical Triple Aim things you're going to target, maybe all of them.

There are activity metrics that you can measure along the way to start to predict whether you're going to achieve the outcomes metric. That's why we're talking about things like the enrollment, the persistence, those types of things because frankly, some of these things take a year, two years, three years to start to show results.

What we spend some time talking about as a team when we're trying to define something is what do we believe will have a significant, both correlation and causality, in creating a statistical tie between that activity metric and the outcome metric? Because, frankly, there's just nobody out there that is saying, "We've got a seven-year empirical study that shows that if you do X, that it will reduce your cost of care in this population by 3%." It doesn't exist yet.

Catalyze: You mentioned Omada, but there are others like Mango Health, a big group of those. When you look at a company and evaluating it for a pilot, do you want to see some basic metrics?

Mr. Jones: It's one of those things that's really hard to just define globally what your criteria would be. What's the magic thing for a vendor that's going to come in and talk to our teams and for us to say "Yes"? It really depends on the stage of where they are. If you've been in the marketplace with a product for any appreciable period of time, you should have metrics, and you should be able to say, hey, here's where, and also, you should understand clearly where you are on your pathway to getting to outcomes and to being able to demonstrate outcomes.

We talk to companies who say they are a pre-product. "We're in prototype phase," or "we're in our final phase of testing," or wherever they are. We'll look at them with a very different lens. We're looking at them with a lens of, do they look like they have a true understanding of the space in which they're playing? Do they actually have a sales and marketing plan that's going to get this in the hands of people that actually use it at the end of the day? And do they look like they have a credible road ahead of them? And is it smart for us to jump in and start to do a group concept or a pilot with them?

If you have been in the marketplace with a product for a period of time, you really should be able to say, "Hey, we are at this point. We had expected to have this by this point. We have these metrics. We believe it's going to iterate to this, and have a really credible story around it."

I find that folks tend to underestimate the importance of placing their metrics and their success in the context of where they are in the journey, and they can tend to lose credibility that way.

As somebody says, "Well, we signed up 40% of the people that we targeted."

Okay. Well, what does that mean?

"Well, we did really better than the other guy."

Well, okay. Yeah, but what does it mean? How are you going to get from here to there? It's a hard thing to do. It's a hard discipline to create that set of activity metrics that you believe are going to have a causality correlation to get to your outcome metrics and manage it and get better and better over time. It's not an easy discipline to do, and I think folks tend to sell it a little short.

Catalyze: I've never really thought of it that way. Of course, I don't evaluate probably as many vendors or have as many pitch me as you and your group do. It's almost like you need an outcome story, right? There are other indicators, like you said, and view them with a different lens. Whereas, if you'd been in the market for a year or two years, you're going to want to see something that shows, there's some sort of positive outcome from whatever it might be, adoption, ongoing engagement, whatever the metrics might look like.

Mr. Jones: Or even: "We expected to see this. We didn't. We learned, and we're pivoting, and we're going to do it differently this time." That's an acceptable story. It shows self-awareness, and in some ways it's good because it shows an ability for a company to not be so stuck on the pride of ownership.

Catalyze: One of the things I was going to ask about and am curious now that we're talking about not having too much in the way of long-term data, I've tried Fitbit and Jawbone, and I used to have a Nike FuelBand, Living Scale, the whole thing. I'm ashamed to admit it, but I don't use any devices at this point. I use a running app for running, but that's it.

One of the major issues is fatigue, and the bright, shiny nature of some of these devices and technologies and apps. But it's the ongoing engagement, the longer-term metrics.

When you look at engagement or continuing engagement, is there a tipping point where people are using something and it's not a fad anymore?

Mr. Jones: Yes, but I think it's different every time, and I think the tipping point is when an individual begins to see value in it. Let's take your example—and it's funny because I'm looking around at my desk here, and there's a drawer over there, and I'm wondering how many of these devices I tried for six days. There's literally a Fitbit sitting on my desk; it tells me I'm not exercising enough, and I sleep poorly, and here is part of the problem. If there's a problem, so what? What do I do with that? I already knew that there are things that I should do to sleep better. And I already knew that I needed to exercise more. The Fitbit didn't add value there, right?

If we take your example, you're using a running app maybe Runkeeper or something like that that you're tracking your time and your pace, and it's showing value. It may be to you that the value is you're training for a 10K or marathon or whatever, and it's showing you your progression, your pace. The value to you may be that it comes on your headphones and says, you're a mile in, you're going at this pace, and it's an encouragement along the way. But you did not reach a tipping point with a Fitbit, with a Misfit with a, pick your poison, a Pebble or whatever device that you want to choose to say that there was an inherent value for you, and you continuing to use that device, which by the way, is kind of where the tipping point gets to, you typically can change behavior.

Now with your running app, something has demonstrated value there to you, and there's some inherent value to you in helping it change your behavior to achieve an objective likely. So maybe you're running a little longer. Maybe you're running a little faster. Maybe it could be that you're competing with your wife on how many miles you guys can run in a week, and you want that at the end of the week to not get beaten quite as badly as you usually do. [Laughter] But it comes back to kind of the behavior change and the intrinsic, it has to trip an intrinsic value switch in your head for you to continue to use it.

Catalyze: I didn't get much from realizing that I was taking 8,000 steps on Tuesdays and 11,000 steps on Thursdays. And so ultimately, there wasn't a value there. I did have a Misfit and a Pebble. I forgot about those two, as well. They're all scattered throughout my house.

Catalyze: Transitioning from quantified self and wearables to Cal INDEX. I know Cal INDEX, and I think others know Cal INDEX, but outside of California I don't know if a lot of people do. Could you talk a little bit about Cal INDEX, the development of it, the motivation behind, and where it is today?

Mr. Jones: Sure. A few years ago, actually probably four-ish years ago, at Blue Shield we have accountable care organizations. We have commercial accountable care organizations throughout California. We've seen demonstrable success in those changing the trend. In fact, they run typically half or less the annual cost of healthcare, which is very significant, especially since we have ones that are doing it year after year, and so we have sustained two, three, four-year success there.

We had a realization and a recognition that you can, through hard work and alignment of incentives, get only so far with programmatic alignment. Then you start to run out of space. Even if you did everything that you could possibly do manually, as effective as it is possible to do it, there are still things that just are not possible without technology.

Things like a physician office really understanding gaps in care for their entire patient population. Who's had a flu shot? Who hasn't had a flu shot? It's just not practical for one of the people at the front desk to be carrying a list that is manually written on a piece of paper. And checking against that when somebody is coming in for an appointment, and calling them at the end of the month if they don't. There are certain things that don't work. We recognize that you needed technology in general to be able to accelerate the pathway to savings, be able to prolong it and extend it, and to be able to increase it.

That technology would be likely fairly disparate as we talked before. It's a different thing for different applications, and there's no one size fits all. There's no one vendor that you can go out there and go buy a packet from them that says we'll just improve everything overnight.

The heart of it is data. We looked at it, and we said, you have to build the bottom up, data centric strategy because everything is dependent upon sharing the data and having really, ideally, a consolidated view of an individual's data.

Creating that person-centric longitudinal patient record with all the information about that person is important from a physician treatment perspective, so that I have the most comprehensive view of the patient that I can. But also from a tool perspective. If you are looking for gaps in care that you want to flag and send alerts out, or you want to understand somebody's persistence on a medication or any of these things, you have to have that underlying data, kind of a base layer. It has lived in, first, it lived in paper islands, so the paper charts in a physician's office, those giant racks that we see in paper charts.

People would maybe carry that paper chart around with them in a little plastic baggie of medications and that was how the data moved around the ecosystem previously. Then it flips to these electronic islands of information, so again, completely non-interoperable siloed islands, so you've got all the EMRs that sit there that are these great electronic filing cabinets that are really not very good at exchanging information in between.

Then we did HIEs, and the HIEs had typically been provider-driven or government-sponsored pilots. The provider-driven ones said, "Well, for competitive reasons, we're going to only share it with our close partners". You didn't get, "Well, if somebody went to the emergency department two towns over because they crashed their car on the way home from grandma's, that they had an easy access to the information."

Grant funding is, with few exceptions, not a great long-term business model, so they have a hard time demonstrating that. The other piece was payers weren't playing, and payers actually have a significant amount of centralized data. One of the things that was missing was some of the understanding in the clinical community of how much data a payer actually has about a patient.

We said, "Look, we need to get as much data in one place as we can, which means the payers, the providers, everybody is playing in one place. And we needed to take the competitive aspect out of it so that somebody's not saying, "Well, I'm going to wait until this guy goes in because I don't want to put my data in first because it's just, I already have it so it's no advantage."

We also recognized that we represent a portion of a physician's panel, so let's say it's 15%, 20% in a given practice. Nobody is going to change their clinical practices for one-in-five, right? They want it to be a significant portion of their population. We got together with Anthem Blue Cross, which is our biggest commercial competitor in the state, so it was an interesting dynamic in putting that together. We said, hey, we all know that we need centralized data. We know it's better for everybody.

We went out and we talked to our provider partners, their ACO partners, and we said, "We're in this together." We created a not-for-profit company and funded it and have seeded it with the information, the clinically relevant information and claims that we have. We didn't put any dollar amounts in there, so you're not doing any sort of gaming around the dollars or anything like that, but it's the clinical information.

On a medical claim, you have who treated somebody, what the diagnosis was, what the date was, where the place of service was, and then you get the pharmacy information. It's not the richness of an EMR, but it's got a base layer of quite a bit of information. We did that with our biggest competitor. We started it, and it's up, and a participation agreement is completed and starting to get some of the providers signed. We're right on the cusp of that, and it has, oh, gosh, the numbers were, I think it was close to half a billion records in it now in terms of claims.

You start talking about amount of clinically relevant data, about large portions of the population. It's pretty huge. We're starting to get to the point where it's got some real mass, and people are starting to look at it and say, "You know what, that's actually a pretty important set of data for people." The hope is that it not only will be just that source of data, but it also starts to provide an infrastructure that we don't have in healthcare, right?

If you look at why the financial services industry did so well in moving online, well, they had credit card processing and the ability to move money around from bank to bank, and so there was an infrastructure there that they were able to leverage with the Internet and then ultimately, mobile. Retail simply piggybacked on that. You look at these massive digital transformations in banking and retail, and there was an infrastructure in place. We have not yet had that in healthcare, and so we're really hopeful that Cal INDEX can become that digital infrastructure that we need to make things go.

Catalyze: You mentioned grant funding as really a very poor form of sustainable funding. What's the business model for Cal INDEX in terms of how it continues to operate and be funded?

Mr. Jones: We recognized the fact that this was going to take some time to demonstrate true value in the ecosystem. We came together with Anthem and said, "Look, we will fund this for the first three years for all of our members so that Anthem members and the Blue Shield members and any other payers who wants to come on, we hope they will adopt the same philosophy."

We're going to fund this for the first three years of operations, and we're going to fund the operations of Cal INDEX, and we're going to fund the set of initial implementations with some of the close provider partners and some of the bigger systems in California so we can get it to scale, and we can get it to where folks are starting to see the value.

That's a three-year time horizon that we committed to that initial funding for. We don't want to pay for the whole thing forever, and we know Anthem doesn't want to pay for the whole thing forever, but it should be shared across the ecosystem. So we're going to fund it for this seed period of time. During that seed period of time, we need to demonstrate value.

This isn't a grant-funded model that can putter along for as long as you get the grants. We actually have to demonstrate value. So we're in the process now of working with them to set up, what are those discreet measures? What are those discreet metrics that we frankly have seen very little evidence of to date? I mean, there are some studies that are out around how you can save money in the emergency department and eliminate some waste around the use of an HA, but there's nothing really broad yet.

We're working right now on what are those discreet measures that we can, at the end of this three-year period, when we sit down with everybody who is a participant in it, we could say, okay, it's valuable. Here's why. Here's to the extent. Here's to whom the value accrues, and those people should start to pay. We're going to be one of the people who pay some of the break on it, for sure.

Catalyze: You already have the clinical data from claims, like you said. Are there are participants, like ACL participants, now using Cal INDEX?

Mr. Jones: Not yet. We are right at the point where we've got the bolus of data in there. We've got a system that's production ready. We've got folks that are starting to sign up, and then we're going to move from there into implementing the provider data streams, which is, since we don't have very good inter-operable data across all the EMRs in the country, is a massive understatement, right? We have a road of implementation there, so that's part of the reason too. We recognize it's going to take some time to really build it into fruition with everybody truly using it.

Catalyze: In addition to the participation from different groups and eventually being just this record that ACOs can use to get a better picture, and providers can proactively address different groups in their population, do you see Blue Shield and/or others that are participating also building technologies on top of Cal INDEX?

Mr. Jones: Absolutely. There's no question. In fact, we're working on a mobile application that's going to sit on top of Cal INDEX and helps physicians understand what's going on with their patients now. It starts to provide that consolidated infrastructure that you're not having to do this aggregation across huge numbers of areas, something that you're somewhat familiar with.

Catalyze: As you build out these solutions, whether they be for providers or for patients, do you see Blue Shield offering and using applications developed by others that are part of Cal INDEX? Or are you offering your solutions to other ACOs or providers or even payers? That's sort of a whole new world of collaboration that I'm trying to imagine.

Mr. Jones: It is, and it's an interesting dynamic because we are having discussions about that on an almost daily basis. If we come up with this great application, and we think it might provide a competitive advantage to Shield, it may benefit patients or a more mission-driven organization. Also, by the way, providers won't use it for a small sliver of their population generally. What is the model? What is the right way to approach the marketplace? If you take a real marketplace look, you have to look at the supply and the demand of it.

I do think you're going to see Cal INDEX ideally evolve into something like a force or whatever marketplace, like the Apple App Store, the Google Android Store, whatever, you pick your kind of infrastructure component. I can see where there would be scenarios that we would say, "Hey, this is something that we're going to use. It's specific to Blue Shield members because it deals with the administrative, or it deals with this, or it's really specific to a product type for an ACO."

I can see things that we would say, "Hey, this is great for our ACOs, but the providers aren't going to use it for 20% of their population, so we're going to put it on the app store and allow them to hook it up to Cal INDEX and use it for all their patients. But again, there's a fair amount of evolution that I think needs to happen. The one fairly consistent piece of feedback that we get is doctors, nurses do not want to use different apps for different things, that do the same thing for different patients. It just doesn't make sense.

Catalyze: Clinical adoption is hard enough. You don't want to have them have their groups of just Anthem apps and Blue Shield apps and things like that.

Mr. Jones: That's going to be difficult for the industry and for the payers. That's new territory. That will be uncharted ground for all of us.

Catalyze: In addition to Blue Shield or Anthem or Dignity or whoever it might be developing applications on top of the Cal INDEX data, do you also see at some point bringing in external vendors that are going to run on top of that same data or interface with that same data that aren't necessarily provider payer participants?

Mr. Jones: Absolutely. If you look at the health information technology space, right, you just look at the amount of investment and the number of independent companies, and it's not just the dollars that are going in, it’s the minds that are going and just the intellectual capital that's being poured into this space right now. There's so much there, and if you were to limit it to applications that were developed by payers and providers, you'd be taking this thin sliver of the whole, and it would be a foolish endeavor.

I think what we would like to see happen is our headquarters are in San Francisco, and there's this place just south of there that some technology occasionally pops out of. And so what we would really like to see is, we'd like to see those folks having an ecosystem in which to develop and deploy technologies that are total game changers, and they're the likely place for it to come from. These kids are coming out of Stanford or Cal or wherever that just, they don't know what you can't do, and that can be a really powerful force. So having an environment where you are able to protect all the security, the privacy, put it into a managed ecosystem, the potential and the power is huge there.

Catalyze: Is there anything else you wanted to cover?

Mr. Jones: You know, to wrap it up: It feels like we are at a cusp that is exciting, because it's not like we're having to invent any of the underlying technologies anymore. There's so much there to work with. We have all the Lego pieces, and it's how do we put these together? How do we put them together in a way that brings the system together at one place?

Then I'm going to go back to: If it doesn't change behavior, it doesn’t really count. Folks that are working on this, including us, that should be your one point that you can say at the end of the day. It changes behavior in this way, and it leads to this outcome that's positive for the system.

Catalyze: It's fascinating to hear about some of these potential collaborations, because it feels like completely uncharted territory for some pretty big players in the industry. It is exciting to watch. Humorous at times, but pretty exciting to watch.

Mr. Jones: I think we'll have our laughter and tears and everything in between along the way, I'm sure.

Catalyze: I'm sure. Thank you, Simon. I really do appreciate your time. We really, really do thank you so much.

Mr. Jones: Thanks, Travis.