Chapter 07: Developing an Innovation Center at UPMC
Files
Loading...
Description
In this chapter, Ms. Kaul talks about the next phase of her work at UPMC, establishing the Technology Development Center at UPMC. She explains how the national economic climate at the time influenced the projects they took on. She talks about developing a clinical documentation improvement system that saved the institution $28 million.
Ms. Kaul talks about lessons she learned while spurring innovation at an academic medical center. She comments on leadership issues, including team building and providing incentives.
Next she discusses information technology differs from innovation, which includes product development and taking the product to market. She sketches challenges that arise during this process. Then she tells the story of developing a free app that helped helicopters find emergency locations and send EKGs to hospitals in advance of patient arrivals.
Ms. Kaul observes that it takes an organization’s entire culture, rather than individuals working separately, to create innovations.
Ms. Kaul returns to the story of the records coding software and talks about working with the UPMC health plan to develop risk assessment coding. After mentioning a few other projects she worked on, she notes that UPMC wanted to change the focus of the Innovation Center to external investments.
Identifier
KaulR_02_20160425_C07
Publication Date
4-25-2016
Publisher
The Making Cancer History® Voices Oral History Collection, The University of Texas MD Anderson Cancer Center
City
Houston, Texas
Interview Session
Topics Covered
The Interview Subject's Story - Professional Path; Finance, Entrepreneur, Biotechnology; Professional Practice; The Professional at Work; Discovery, Creativity and Innovation; Discovery and Success; Technology and R&D; Fiscal Realities in Healthcare; The Healthcare Industry; Leadership
Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 License.
Disciplines
History of Science, Technology, and Medicine | Oncology | Oral History
Transcript
R. Kaul, MBA:
So when that concluded, I came back to UPMC, and was offered the job, to develop an innovation center that would be developing many of these kinds of opportunities, right? Because obviously I had just built a team to do this, and proved value, so the question was, Could we build a portfolio of those kind of opportunities that would be highly valuable operationally and also commercially viable? I said, Sure, that would be a great opportunity to try. And that started my job as running this what they called the Technology Development Center at that time. So I was employee number one of the Technology Development Center built a small team and started to look for opportunities. And at that time, our risk tolerance was somewhat low, there were some economic downturns. So we started to get involved in sort of what I call low-risk, low-reward initiatives, meaning teaming up with big vendors and forming co-development relationships with those big vendors. But those big vendors would own all the intellectual property, take the product to market through their sales channel, which is a well-established sales channel, and we would receive -- we would help in the development of the product and then we would receive free usage of the product, as well as a royalty stream on all sales of the product. Which, I say, I mean that's pretty low-risk, low-reward, because we're not really paying for the usage of the product, and we're --for any sales, it's somewhat gravy, right? So we formed a couple of those kinds of relationships, one of which was with a company that had acquired A-Life Medical, once we brought it back in they kind of came back to us and said, "Can we do the next product together?"
T. A. Rosolowski, PhD:
What was the company?
R. Kaul, MBA:
0:24.45.8 Optum. So we formed a couple of these kind of relationships, and we -- those were pretty successful, which we proved a new tranche of operational value. So, for example, what we did with Optum was build a clinical documentation improvement system, so starting to then, back to my strategy around natural language processing, we started to think about, well, if we could look for the presence of information, kind of again back to my, can we increase the complexity? Can we also look for the absence of information? So can we improve documentation by looking for what's not there that needs to be there for coding purposes, because again, it was with Optum that was focused on coding at the time. And so we built a clinical documentation improvement system that provided, I believe, $28 million annually of value through, prompting physicians for more information that led to more accurate coding. So, that was successful. We had developed with a company, a telemedicine solution, and deployed that widely throughout the enterprise. That wasn't a commercial success, but it was an operational success.
T. A. Rosolowski, PhD:
What are some lessons that you were learning during this time?
R. Kaul, MBA:
I think a lot about how to innovate in a big academic medical center. It's really hard to get support. It's really hard to drive adoption, to get people to participate, to get people to think differently, to try products that aren't necessarily shrink-wrapped. You know, people are used to buying products that have been in the market for 10 years; they're not used to trying something that's in a prototype stage or in a beta stage that's going to have issues. People aren't used to having to spend the kind of time giving the feedback in the development process. And people don't want to, and generally, incentives aren't aligned. So no one's being incentivized to spend all this time with the innovation team, so it's extra. They have to do their day jobs, and then they have to spend time with us. So part of the learning was also, how do you create soft incentives? Because if the administration's not going to literally change the incentive structure, which most people don't, how you play to the values of what's important to the people who participate. And sometimes you get lucky; there's usually what I'll call "grassroots innovators" in the community that just want to do it because they find it interesting, and it makes their day-to-day more interesting, they want to participate, and that's the best scenario.
T. A. Rosolowski, PhD:
Makes them feel like they have an impact.
R. Kaul, MBA:
Yeah. And, like I said, it's something that they just in their heart enjoy doing. That's the best case scenario, to find people in the clinical and operational community who are just innovators at heart and want to participate. But that's the exception, not the rule. And there are people who do it for the glory, so meaning they get to present on an executive committee. They get to do a press release or a presentation and things like that. So there's an ego component with some people. You know, yeah. There are all different kinds of ways. Some people have other agendas, like projects they want you to do that maybe have no commercial value and you wouldn't ordinarily do, but you figure if I can -- it's worth the investment of, say, spending -- putting a developer to on a kind of pet project for someone for a couple of months to get that done, in exchange for them participating in something. Of yours. So there's a little bit of bartering and trading. Trading on agendas.
T. A. Rosolowski, PhD:
Well, it was interesting because when you were describing the process, you know, of going from that small, focused project, and kind of scaling up and exploring the different options, it was reminding me of clinical research, how it goes through these different phases. And you have to follow one pathway to prove this, and then you have to spend time providing something else, and then you put it together, prove the next thing. So it's very different, like, as you said from just going, okay, shrink-wrapped right off the shelf into use.
R. Kaul, MBA:
Right.
T. A. Rosolowski, PhD:
I mean, it just isn't the way it works. And people don't think about it that way.
R. Kaul, MBA:
People don't think about it, and there's always this notion that people have of, oh, in particular there's always a rub with kind of IT where it's, like, well, why isn't this just IT? Well, it's genuinely a different skillset. Product development, commercial product development in particular, but also product development is different that implementing and supporting off-the-shelf packages. I might make the analogy that in the finance world, you don't think of an accountant the same way you think of an investment banker, or a financial analyst, right? Even though they're both in the financial field.
T. A. Rosolowski, PhD:
Sure.
R. Kaul, MBA:
0:30.53.1 Those are two very different jobs. It's the same thing in IT versus innovation, I mean, like I said, doing deployments and supporting systems is very different than developing systems from scratch, and taking them to market. And, there are different skillsets, and it's different training. So, that was also hard for people to understand. The other kind of cultural thing is, people, hen you say, "OK, you're creating an innovation team," people say, "Well, everything is innovation. We all are innovation." And it kind of goes back to, I think, what we said in my earlier discussion, which I don't even like this word "innovation" because it creates all these feelings that are completely unnecessary. So, agreed. There's an -- innovation can mean a lot of things, and everybody innovates in some shape or form in what they do every day. It's just sort of how we're defining it for this particular function, you know. So, I don't disagree with that statement.
T. A. Rosolowski, PhD:
Well, yeah. I mean, your notion of innovation has this resonance about an impact on an entire system and what the goals of that system are. So it's a whole lot better than, Oh, well, let's make this room blue instead of green, because people like it better. You know? That's my bright idea. You know, it's a whole different order, exponential order of what this new creativity is about.
R. Kaul, MBA:
Yeah. It's different, and there's a lot of lessons learned. The other thing is dealing with these big vendors, when we did some of those earlier ones -- and I'll tell you over time the pendulum shifted and we did more startups. But in the early days, when we were doing these big vendor relations, innovation is really hard and really limited with large vendors. It's really hard to manage these big vendor relationships. I mean, get vendors to be open to new ideas.
T. A. Rosolowski, PhD:
0:33.23.8 Culture just more entrenched ways of doing things? What is it?
R. Kaul, MBA:
And there's also the catering to, and it's like what every textbook tells you not to do, but it's typically what happens, is they hear from their sales people that this or that needs to be done to get the next sale, and so you're always just catering to the next sale, as opposed to the vision in three years, do you know what I mean? So the next sale is an innovation. So the next sale -- is incremental to get the next sale. And it's really just narrowed to what the -- like I said, every book you read on innovation will tell you, the last thing you need to do is listen to what the next sale is. But it's generally what people do, because they're large companies, they have market share, they're driven by next quarter's numbers, right? And everybody works -- I mean, one of the biggest lessons is, people work to their incentives. Right?
T. A. Rosolowski, PhD:
Yeah.
R. Kaul, MBA:
In any context. And when people set up these innovation groups, whether it's here, whether it was at UPMC or anywhere else, one of the biggest things, when I talk to other innovators in the industry, we all talk about is, my goodness, wouldn't it be nice if incentives were aligned? If people were actually part of the incentive system actually was to innovate and to do this kind of work, then we wouldn't be pulling teeth all the time. In the industry, people are incentivized for their next quarter's numbers, which is either making more sales, or if they're consultants driving more billable hours. Nothing says, in their incentives, says, "innovate" or "develop the future," or "make an impact," right?
T. A. Rosolowski, PhD:
Yeah. Or that lip service can be paid to it, but is it part of performance review?
R. Kaul, MBA:
Exactly.
T. A. Rosolowski, PhD:
Is it part of getting your bonus? Yeah, exactly. Yeah, yeah.
R. Kaul, MBA:
Exactly. So working with these big corporations and navigating your own big corporation is not a recipe for acceleration and transformation. It's hard.
T. A. Rosolowski, PhD:
Did the situation change, at least in part, when you started working with startups?
R. Kaul, MBA:
It did. It presents a different set of issues. So we were -- I mean, as much as I just complained, to some extent, about these big companies, we did have some successes. I mean, are they big transformational initiatives? They're OK. They were successful, they were commercially successful, etc. They were better than what we had.
T. A. Rosolowski, PhD:
Well, I have a feeling that as you're talking, you have a continuum. You know, that the projects that really jazz you --
R. Kaul, MBA:
Right.
T. A. Rosolowski, PhD:
-- that you're going to put the seal of approval on are the big transformational, oh my God, we went to the galaxy far, far away, whereas -- but it doesn't mean that small or incremental changes are bad.
R. Kaul, MBA:
It doesn't mean we can't do these other things.
T. A. Rosolowski, PhD:
Right. Right. No, no, so I un-- yeah.
R. Kaul, MBA:
0:36.50.1 You have to earn your stripes with base hits. It's just the way it works.
T. A. Rosolowski, PhD:
Sure. Well, and you also have to work with a client where they are.
R. Kaul, MBA:
Yeah. So over time, we, as we develop things with these big vendors, there was sort of the same conclusion from it, from executive leadership, which was, gee, if we're so good and we have all these ideas, and we're coming up with these ideas that are making these big vendors successful, again, back to the original premise, why can't we develop some of our own, or have more of a stake in the game? So it was kind of full circle back to that same conclusion.
T. A. Rosolowski, PhD:
So just so I understand, at this point, you were helping other vendors develop their products, but you weren't, at that point, collaborating with them?
R. Kaul, MBA:
We were. I mean we would put people on the ground to collaborate, to build a new product with them. But it would still be their product.
T. A. Rosolowski, PhD:
Their product. OK.
R. Kaul, MBA:
We would put 30 people on the ground next to 30 of their people, and we would work as one team to develop it and use our facility as a test bed for it, and we would put developers and designers and analysts and everything on the ground with them. But we didn't have control. We were informing the process, pushing the process. But it's ultimately the vendor's product, right? So it makes it hard. You know. And we had an economic interest, so it was positive, but it was still incremental.
T. A. Rosolowski, PhD:
Sure. So at what point did the executive leadership say, hey wait a minute, we've gotten away from our initial idea, which was to --
R. Kaul, MBA:
About three years in.
T. A. Rosolowski, PhD:
OK.
R. Kaul, MBA:
Three years in we had a real shift. So when we first started, it was your typical life cycle and innovation of people questioning why we needed an innovation organization, not knowing what it was, not necessarily supporting it. And us having to earn our stripes. And we did that with some of these big vender relationships. We also did it with a couple what I would call "skunk works" initiatives; things we built with our own team that were valuable, not necessarily commercial, but valuable operationally that gave us --built relationships, and gave us some traction where people started to think of us as people you could come to to get stuff done.
T. A. Rosolowski, PhD:
I have to ask. Why "skunk works?" I like that name.
R. Kaul, MBA:
Why I call it that?
T. A. Rosolowski, PhD:
Yeah.
R. Kaul, MBA:
Oh, it's just -- again, it's just another common --
T. A. Rosolowski, PhD:
Oh, OK. It's so colorful.
R. Kaul, MBA:
Kind of just, you know.
T. A. Rosolowski, PhD:
I hadn't heard that one either, so...
R. Kaul, MBA:
Yeah, so, yeah.
T. A. Rosolowski, PhD:
You're bringing a whole new vocabulary to the table.
R. Kaul, MBA:
Sorry.
T. A. Rosolowski, PhD:
No, no, it's good. This is good.
R. Kaul, MBA:
So yeah, just kind of, like, little projects. Like we built a -- it was funny, we built this little emergency medical service navigator that gave protocols to EMS personnel on the ground through an application of the iPhone and the Android. And helped the helicopters land in the right place, and then allow, like, EKGs in the field to be sent back to the hospital and things like that in advance of the patient arriving. A little --it took us about a month to develop, nothing -- no big deal. It actually helped us initially just get our own ducks in a row on what our processes were going to be, work with a group that was engaged and they could kind of come to us with this, that they really wanted it, and it would help them, nothing commercial in mind. Nothing, like I said, no big deal, not a lot of intellectual property from the perspective of this -- if it only takes you one month to build, it's not like it's -- it's not something that someone else can't replicate. But again, it got us learning about ourselves, and how we would work as a team, and how we would work with others and build some credibility. So we just did it. And to this day, it's still out there. And it has more than twenty thousand users.
T. A. Rosolowski, PhD:
Wow!
R. Kaul, MBA:
We were, like, shocked. It got really popular.
T. A. Rosolowski, PhD:
Yeah, yeah.
R. Kaul, MBA:
It's free, I mean something -- it just was something we just decided to do. So we did some of those kinds of things that just gave us credibility. We got lucky on that one, I mean, I think the Discovery Channel picked it up and did a whole piece on it. Because again, it got traction in the market. You know, people just liked it and it got a ton of downloads and it became this whole thing. But, and it was just lucky. We just did it because we were just trying to get going. Get some things done. So, we did a combination of those kinds of things over the first three years. And we went through kind of a curve of people not believing in us, people not supporting us or understanding us, maybe all of those three things go hand in hand. They don't believe in you or support you because they don't understand you, and typically there's the notion of kind of being threatened by what this -- that's my, again, back to the name "innovation," it seems to threaten people. And by year three, we had earned enough credibility, had delivered enough value back to the institution that that was sort of a shift, where it became people wanting to jump on the bandwagon and become part of us, versus thinking we weren't valuable. So but it took that long. And it does take that long. Innovation doesn't happen overnight, it doesn't just, like, oh yeah, go innovate and come back tomorrow and show -- you know what I mean? It doesn't really work that way. And it -- not only that, especially in the context of a large bureaucracy. It takes time to even understand how to navigate that. And especially when you have to build from scratch. The -- not just products, but the people. First you have to build the organization, and then you can actually innovate. You can't really innovate as an individual by yourself. I can't sit in my office. Go innovate. It doesn't really work that way. It's about building an organization. So -- and a culture. So about year three, it shifted, I would say. It actually started to become fun. And that's when we started getting a little bit more latitude to work with the startup community. So at that time, we found a startup company -- we were interested in kind of going back to the natural language processing again and seeing if we could take it step further. Of course, that's always like a pet thing of mine, since I'm deep in that space from my experience.
T. A. Rosolowski, PhD:
Right.
R. Kaul, MBA:
So, you know, when you start to take risks, you lean on the things you know best, so you're taking risks, and something you're kind of well-versed in. So one of our first entrees into kind of the startup community is, we, you know, thought about taking natural language processing past just an episode, past just the presence and absence of information, and starting to think about how do you look across episodes at the whole patient, right? And that -- but still staying tangible, still driving a business case, right, because you don't want to get too out there where you can't build a business out of it, where it's too researchy. So we started talking to our health plan, because health plans start thinking about the overall risk of a patient, which means, you know, what does a patient look like over the course of a year, because they're managing populations of patients, and they're managing how that patient did -- you know, they're managing in the context of a year, not in the context of an episode, right? So the way health plans quantify that is in risk adjustment coding, HCCs. So we started working with their HCC -- their risk adjustment coding group, and developed a technology with a startup company to do the risk adjustment coding. And we were able to show, developing a similar computer-assisted coding system from a workflow perspective, but for the purposes of risk adjustment codings for health plans. Because again, it allowed us to push technology to the next level. And we did it with a company that was not well funded, and was on its last legs so that we would have an opportunity to invest. We ultimately showed a $30 million value proposition to the health plan through better risk adjustment coding, and because that drives how health plans get reimbursed by Medicare. So and we took a 60% ownership in the company. So then, we're starting to move along this commercial continuum, to higher risk, higher reward. And that's still an investment that UPMC has today.
T. A. Rosolowski, PhD:
What was the health plan that you worked with?
R. Kaul, MBA:
Our health -- we had our own health plan.
T. A. Rosolowski, PhD:
Oh, you had your own health plan, OK.
R. Kaul, MBA:
We had UPMC health plan. So we were able to go -- because we lived in a payer provider world, it was --we as an innovation group worked across anyone, I mean, we weren't -- from a corporate standpoint, we were aligned to anyone in particular. We were enterprised so we could work with the health plan, or work with -- you know. So that was sort of our first entrée into the startup community. We had some of our own developments that we, when I was there, had planned to spin out. We hadn't done it yet while I was there. But we were starting to get more into the space of looking at startup companies, and thinking about investing. And shying more away from some of these big corporate partnerships because it --you can't move the needle as much. And that sort of, like, brings me to where I exited, which is the -- that direction was so embraced, to some extent, that they wanted to use this as a true revenue diversification strategy for the enterprise, to a point where they set up, upon my exit, an investment fund. They were talking about it while I was still there, to actually move all the way on the continuum to where it's less about internally innovating and developing products and more about investing in startup companies and managing an investment fund, which is where they are today. You know, if you read the newspaper, they've rebranded. They're no longer a development center; they're called UPMC Enterprises.
T. A. Rosolowski, PhD:
Oh, interesting.
R. Kaul, MBA:
0:49.27.6 And that rebranding really does reposition who they are which is ultimately if you look there, making financial investments in companies more than they're doing really any development toward what I would consider innovation.
Recommended Citation
Kaul, Rebecca MBA and Rosolowski, Tacey A. PhD, "Chapter 07: Developing an Innovation Center at UPMC" (2016). Interview Chapters. 1266.
https://openworks.mdanderson.org/mchv_interviewchapters/1266
Conditions Governing Access
Open
