"We need to be cautious in the use of technology to prevent increase in total costs of care"

Welcome to the big unlock podcast where we discuss digital transformation and emerging technologies in healthcare. Here are some of the most innovative thinkers and leaders in healthcare and technology talk about how they are driving change in their organizations.

Paddy: Hello everyone. This is Paddy and welcome back. It is my great privilege and honor to introduce my special guest today Karen Murphy, Executive Vice President and Chief Innovation Officer of Geisinger Health. Karen, welcome to the show.

Karen: Thank you Paddy and thank you for having me.

Paddy: You’re welcome. Karen as I understand it you are the Founding Director of the Steele Institute for Health Innovation for Geisinger and also, I think the first Chief Innovation Officer. Can you maybe give us a little bit of background for the benefit of our listeners and how it came about and what are the objectives of the Institute?

Karen: Sure, I’d be happy to Paddy. So, I was working as the Secretary of Health with the Commonwealth of Pennsylvania. And I met Dr. David Feinberg and I was working on an innovative payment model called the Pennsylvania Rural Health Initiative and engaged stakeholders across the Commonwealth to work on this payment and delivery model with me. So, when I met Dr. Feinberg shared with me that he had this exciting new role at Geisinger called the Chief Innovation Officer. And just a little bit about my background, I started my career as a registered nurse, I’ve been a President and CEO of a health system. Then I went into public service to the federal public service at the Center for Medicare Medicaid Innovation and then subsequently went to the state as Secretary of Health. So, I have a diverse background in understanding health care delivery plus the policy piece and really a passion for innovation. So, Dr. Feinberg invited me and said would you like to come to Geisinger and talk about the Steel Institute for Health Innovation. So, the Steel Institute for Health Innovation was created by the board of directors really in honor of a previous hospital health system CEO who is Glenn Steele and Glen needs very little introduction. But Glen was the CEO for almost close to 20 years here at Geisinger and really raised the profile in terms of the national stage for Geisinger being recognized as one of the most innovative delivery systems across the country.

Paddy: That’s really interesting. And I do want to spend a few minutes talking about some of the initiatives that you’ve got going on at the Innovation Center. But to start at a high level, for health system today what do you think are the primary competitive forces that you have to deal with and how does an Innovation Group align itself organizational object in that context?

Karen: Thanks Paddy that’s a great question. So, asking the question of a competitive environment how does an innovation team help. So, I think we are competing in all markets in both quality and price. In terms of the quality of the services we deliver the price that we deliver at is certainly under scrutiny and I really don’t think I think we’ve kind of exhausted the toolbox if you will and trying traditional methods of lowering cost and improving quality. I think we’ve made maybe marginal advances in quality but I mean not in the actual delivery of health care but also when I include quality I examined the patient experience of care, the patient’s ability to access services, the need, the way, the patient’s ability to understand what they’re paying for and what cost they’re responsible for. So, when I say we’re marginal I mean we’re marginal whereas an industry really marginal around that very complex ecosystem. So, let me define first innovation at Geisinger, so we define innovation as a fundamentally different approach to solving a problem that has quantifiable outcomes. So, innovation at Geithner means we are going to focus only on problem. We must take a fundamentally different approach to solving that problem and we must have quantifiable outcome to measure for the innovation. So, I think unless we take a fundamentally different approach to price and quality, I don’t believe that we’re going to move the needle sufficiently. So how does the innovation team attack those two very large problems. Certainly, in collaboration with the clinical enterprise and the whole team across the enterprise. But I think our goal is to spike to very specific initiatives that move the needle in a meaningful way.

Paddy: And that’s a really interesting definition for the works you do in fact the notion of fundamentally adopting a different approach to solve an existing problem is what I hear the most when others I talk to about digital health innovation and you also alluded to this today in healthcare which is access, affordability issues for consumers. Do you care to share one or two examples of how you actually rolled out innovation programs that address these challenges at Geisinger?

Karen: Sure. So, I should say Paddy too that the Steele Institute for Health started July 1st of 2018. So, we’re just celebrating our very first anniversary. So, I’ll give you a couple of examples of how we’re using fundamentally different approaches. So, you may have heard of our fresh food pharmacy where we identify food insecure type 2 diabetics and we prescribe the patient to go to the fresh food pharmacy. And when the patient goes to the fresh food pharmacy they’re provided with fresh food. And the way fresh food diet counseling actually monitoring of preventative services all circling this patient to improve the health of the type 2 diabetic. And I can tell you that we just expanded to three sites because our first site was really very successful. The patients that have engaged in the fresh food pharmacy have seen very positive results some more than you would think then even through medication. And they’ve also demonstrated positive outcomes on their health maintenance exams and some of the other health indicators that we ordinarily would coordinate with better healthier lifestyle. So, we’re really excited about that and that is a fundamentally different approach. We’re currently working on developing a new model of care for patients with chronic diseases. So just to Geisinger alone about 30 percent of our patient population has one or more chronic diseases. And as you know Paddy most of them have more than one. But this system that we’re using for chronic disease management right now is extremely labor intensive requiring a one to one intervention in most cases. So, case manager or community health worker and we look at the diseases in a very siloed fashion. So, we develop chronic disease management programs that are for cardiac, chronic disease management programs that are diabetics, chronic disease management programs for COPD. But the fact of the matter is most of these patients are all three so they could have congestive heart failure, COPD, and diabetes. So, we’re looking at developing a more holistic approach and over the next three years we’ll be developing new care model that leverages artificial intelligence and machine learning along with remote patient monitoring and patient reported outcome to number one. As I said before looking at the patient and more holistic fashion. Number two to really slow down the progression of the disease. And number three prevent disease exacerbations that would require a higher level of care such as the emergency room or hospital admission. So, there’s just a couple of examples of where we’re working really hard at lowering cost, improving quality for the overall welfare of the members and patients we serve.

Paddy: That is such a fantastic example. You know when I talk to folks very often the notion of innovation is somewhat conflated with technology and sometimes you see a lot of startups out there and I’ll come to that a little further on in our conversation. They launch solutions that are fundamentally different ways to approach a problem but there’s no validation in terms of the people process and change management aspects of it. It’s sounds like you’ve covered all of that in the fresh food pharmacy. The cold concept that you just described. I think that is just a fantastic case study. Thank you for sharing that. Now you alluded to a couple of other things when when you talked about that case study. One of them was artificial intelligence. Now we’re seeing a lot of new AI enabled solutions to hit the market. More recently I’ve seen lots of announcements about AI enabled chatbots, symptom triaging and things like that. These are not necessarily just traditional health systems there are a lot of non-traditional players also coming into the market. So, I guess switching gears here a little bit, you care to your thoughts on how health systems could be leveraging technology more to drive the primary care experience and to address emerging competition in the context.

Karen: Sure. So, I think that you know others have used Paddy you raised chatbots within the Steele Institute we have a hub for what we’re calling artificial intelligence automation. I think the use of chatbots certainly are appropriate in a couple places I think those that where we have repeated questions. I think it’s perceived by the patients that the bot does a better job in terms of efficiency and information that I think that’s appropriate. I think the other place where bots are going to be very important and are going to be those tasks that humans do repetitively that we really don’t require a human to do that. But I think we have to have our intelligence bots. So not just a robot that can keep doing the process over and over again but one that learns from the activities that they are doing and I think the part that I think is important about them is that it will help us decrease the overall cost of care because obviously the use of bot is much more efficient and those use cases are really infinite and other industries have demonstrated they’ve done that really well. I think the big part we have to remember about when we talk about artificial intelligence, machine learning in any case is one that certainly there are a whole host of ethical issues that we have to be of course cognizant of. There’s a whole host of especially in clinical medicine if we’re changing the way that we practice. So, there’s a whole, there’s certainly a side to that that we have to be very careful of. But really even and the most simplistic side like all technology there is many things the data scientist can do particularly in the predictive arena. But we need to be aware that in particular in the predictive arena that the data is usable so that we don’t increase the total cost of care. In other words, I’ll give an example highlighting people who are predictive first stroke and there are 30000 of them. Very hard to do an intervention with technology with to notify 30000 people that you don’t actually have to hire other individuals to do it. So, I just want to be I know I’m a little bit always cautious about adopting new technology that we don’t end up increasing the total cost of care as opposed going in the other way because as we began this talk it is critically important for us to decrease that total cost of care and increase the quality.

Paddy: Right. And I think you made some very good points there and we’re all aware that the FDA is now looking at how to really understand and offer some level of oversight on how artificial intelligence rhythms are deployed in the context of healthcare and what is the level of regulation that is required or not. And so, I think the debate is still ongoing, but I think you made some very good points. And just to follow up on that I wanted to ask also about the data sources right. So, for AI algorithms to work well, the more the data and the more the data sources, the better they are at accurately predicting let’s say disease progression like stroke that you mentioned. So, can you share a little bit on how are harnessing emerging data sources such as social determinants or genomics for example.

Karen: Sure. So Geisinger currently has we have about 23 years of electronic health records and unified data architecture. We also have other sources of data in that unified data architecture the social determinants of health. We’re actually not only taking in from external services but we’re looking at now not only screening for social determinants of health but also identifying community resources that we could immediately connect the patient to when we recognize what the challenge is, what the social need is.

Paddy: So, let’s talk a little bit about the fact that Geisinger there’s actually a health plan and a health system and you’re unique. And so how does your innovation model balance the need to build the health system and the health plan and prioritizing your innovation investment. How do you pick your big needle movers as an example?

Karen: So, I would say a couple of things to that. So, the first is that as I said long before I got here Geisinger is a very innovative organization. And there are many operators that I would say are truly innovating here at Geisinger. So, the innovation that occurs is not just limited to the Steele Institute. So, when we pick projects or select projects for the Steele Institute, we really try to select like I referenced before a new model of care. We really try to select those initiatives that have perhaps a larger, more far reaching, ROI in both cost and quality that will benefit more broadly the organization.

Paddy: Now switching topics here, lately there have been several announcements about you know large health systems or health plans are turning innovation programs as commercial entities. In other words, an example of that was a recent announcement by Highmark which is looking to commercialize their innovation program and offering up the data to startups and researchers to test out new solutions or products and really creating additional revenue streams. I was just curious. I know its early days yet for your innovation program but is there a long-term goal to sort of maybe try and commercialize some of this?

Karen: We definitely are exploring commercialization in the common development realm and what that really mean is if there is not currently an application on the market to solve the problem that we’re trying to solve that we would open up and invite companies to come in and work with us and we would co-develop. So there definitely is an interest there. I would say the difference Paddy is that our approach is exactly that. Here we have a problem that we want to fix and we want to work with it and this is how we’re going to work as opposed to just opening it up more broadly we just want to be sure that we’re answering for problems of the future, of the present, and the future.

Paddy: So, when we talk about innovation we’re also talking about digital transformation. You actually provided some great examples of how you’re already on that path in many ways. In the way you select projects, in the way you’re rolling out the projects. It’s of course very difficult for a single innovation group to meet the entire needs of an enterprise especially if you want to accelerate digital transformation. There’s a huge ecosystem of startups out there who could potentially help you, they’re funded by billions in VC money. But it’s also hard to determine which ones of them are validated, what’s the risk involved and so on. So, can you maybe share your thoughts on how you harness external innovation today and what do you think are some of the risks that need to be managed.

Karen: Sure. So, I think you’re absolutely right in terms of there just by definition of what they do is they’re unproven. Right. So, by being a startup early in business not a lot of experience all the risk that comes along with that is inherent to that selection. And I would say quite honestly that we’ve experienced both really good experiences and had some experiences that didn’t work out so well. And I would have to say that they didn’t work out the failures were just as important as the successes because we learned. In this new territory of startups we learned what we would do differently and if you call me in a year I’ll probably say the same thing that we worked with a lot of startups that we had some successes and failures because I think that’s the nature of the work.

Paddy: Well I think we are coming up to the end of our time here and I want to ask you about you know you’ve now you’ve been in both public health and in private institutions and most recently before you came to Geisinger as you mentioned. You were Secretary of health for the State of Pennsylvania. How does a public health experience inform and influence your work at a private institution and what is your advice for someone looking to make the transition either into public health or into private sector?

Karen: So I have said this before Paddy when I was interviewed that I spent most of my career in the private sector but just had the magnificent opportunity to work in both the federal and state government. And I always say that I wish I’d found public service much sooner in my career what I would have done is moved in and out of public service and private sector because I think the best public servant is really one that is in touch with the private sector and understand how the private sector operates. But at the same time understands what it is to have the denominator be the federal, or the population of United States, or the population the state you are serving. So, I would advise anyone that had the opportunity to work in either federal or state government or any public health initiatives for really go for it because it’s a tremendous experience and it’s very much unlike the private sector.

Paddy: That’s great. That’s wonderful. So, we are at the end of our time. Any final thoughts before we conclude the podcast Karen. 

Karen: No thanks very much for having me and I really enjoyed talking to you Paddy.

Paddy: Thank you much.

We hope you enjoyed this podcast subscribe to our podcast series at www.thebigunlock.com and write to us at info@thebigunlock.com