Datica Podcast

December 1, 2015

Robert Wachter, MD - Professor, Department of Medicine at UCSF

Catalyze: I didn’t realize that you were the person that started the “hospitalist” movement, which is not a movement any more but a standard.

Dr. Wachter: People say I invented “hospitalist” and I say, Al Gore invented the Internet. It happened organically, but I coined the term in 1996; I just got the sense that this was going to be organized and it became the fastest growing specialist in history.

Catalyze: I’m sure UCSF has hospitalists tracks now, right?

Dr. Wachter: And programs, because I’ve got 70 hospitalists that work in my group and what’s fun about it, now that we are 20 years into it, they are becoming the leaders in the organization. Nationally, too! The Surgeon General is a hospitalist, the top doctor at Medicare is a hospitalist, so the field by establishing itself and focusing on how do you make systems work better is really matured into a very important specialty in American history.

Catalyze: You are still a practicing hospitalist, right?

Dr. Wachter: I am.

Catalyze: So that brings me to where you found the time to write a book?

Dr. Wachter: Well, I’m a practicing doctor, but in sort of the air quotes way, because I’m chair of a very large department; I sort of call myself a witch doctor these days, I tell people which doctor they need to see if they need help. The biggest challenge for me was not my clinical practice; it was my academic and administrative role. Part of it I wrote at night and part of it came from taking a six-month sabbatical where I went to Boston. I figured the best place to be to talk to those about this issue was in San Francisco, and the northeast. So that’s where I divided my time writing the book.

Catalyze: That makes sense. There certainly is a decent size healthcare/medical complex in the Boston area.

Dr. Wachter: Yeah, I just didn’t want healthcare/medical people, although that was important, but a part of what I wanted to do was to understand the history of digitization with other industries. I interviewed people in artificial intelligence and people who knew about how digital happened in manufacturing, travel and other industries. That was a very rich experience, but those seemed like two good outposts to start from.

Catalyze: I think one of the things about EHR is that they are this center of clinical workflows and documentation today, but they’ve also created challenges just by their ubiquitous nature. I’ve read studies where ER docs, on average click 4,000 times within an EHR during a 10 or 12-hour shift. Can you talk a bit about some of those challenges in terms of using the EHR and in communicating with other physicians and patients?

Dr. Wachter: Part of the reason I wrote the book was I saw those challenges, and nobody was talking or writing about them. When I spoke to doctors about it, often they’d say, “Yeah, this is really hard and I’m thinking about retiring, depending upon how my Roth IRA is doing.” I heard that from everybody. So I realized that there was something ubiquitous going on.

I think part of the challenge was that we implemented this technology without thinking hard about what we were trying to do and about the impact of the technology on people that were going to use it. In retrospect, all of that was obvious. It should have been obvious that if we built this technology that required doctors to document 47 different check boxes and provide high quality care that, a) they weren’t going to look patients in the eye any more, b) they were going to be incredibly unhappy, and c) it would distract them from what they were really trying to do. All that was obvious in retrospect, but it’s never obvious prospectively. Prospectively you think about your iPhone, you think about how slick and magical it is, you think about how it’s just going to make stuff better and the interface will be wonderful.

What I came to learn in researching the book is that this is a relatively typical problem in the digitization of complex industries, which looks easy as you enter it and looks great on the PowerPoint slides and then you get into it and you say, “Wow. That’s not working as we’d expected.” Then what you ask is why not? What do we have to do to fix this? Part of that is technology. Part of that is re-imagining the work. Part of that is asking some central questions about all this data we are collecting. Now we are asking for new data because we can. Which piece of data are really important and which pieces of data are simply busy work and don’t add any value to the patients, and if we can, let’s get rid of some of that. So we are at that early stage where we are saying, “Wow, this is tougher than we thought.” Now we have to sit back and do the hard work of making things better.

Catalyze: I have read a few things about you stumbling into digital health or digital medicine and EHRs. It’s not an area that you knew much about previously, but I’ve also read that you wouldn’t go to a hospital or facility that didn’t have an electronic health record system. Is that true?

Dr. Wachter: That is true. It was kind of an important point for me to make in thinking about this and writing the book. I am about as far from a luddite as you can imagine; I blog, I tweet and I’ve been an early adopter kind of person, but technology has not been my field of study. What I’ve studied and written about is the organization of care and the quality and value and patient safety — all things that are tightly linked to technology. But, I wouldn’t know an API from an APB. The nuts and bolts of how tech works is not my expertise. It became increasingly clear that tech was incredibly important to meet other crucial goals like providing better, safer and more satisfying care and engaging patients, cutting the cost and all that kind of stuff.

The tension here was to honestly understand why we were somewhat off the rails and what we nee ded to do to take full advantage of technology. I have people come up to me periodically and say, “Thank you for making that point. We should bring back the three-ring binders, clipboards and bring back the fax machines.” Those people are crazy. That’s part of the point that I wanted to make. I really do believe that my hospital is significantly better and safer now that it’s digital than when it was analog. It’s really easy to romanticize how great it was. It wasn’t great when you were trying to read the doctor’s handwriting, or we used post-it-notes as our main vehicle of communication. That’s ridiculous, but we have to do it better and smarter than we’ve done it so far.

Catalyze: Do you see a path forward for what EHRs become in the practice of medicine in the next 5 to 10 years, or maybe how a practice adapts to a more digital age?

Dr. Wachter: One of the more fun parts about writing the book and thinking hard about this for close to a year was I came into it skeptical, disappointed and a bit angry. I left it much more optimistic; there’s a chapter toward the end of the book that some of the readers said sounded as if it was written by a different person because all of the sudden, it’s like Wow. This is going to work out. My effort wasn’t to have a Disney ending. I can actually see how we are going to get there.

Is it a 5 to 10 year journey or a 10 to 20 year journey? That has to do with the kinds of questions that we ask and the choices that we make. What I came to understand is what we’ve seen in medicine is quite typical. As a very big, complicated industry shifts from doing business one way to another way they never get it right in the beginning. There’s the concept known as the productivity paradox which is that you expect that this may be wonderful, all the consultants and pundits tell you that it’s going to be wonderful — you implement it, then three to four years goes by and nothing happens. Everyone is scratching his or her head and then it starts getting better. It starts getting better because people start saying, “Why is this not working?” They begin building better tools and displacing the legacy companies that just happened to be there first.

That all starts to happen, but I think the more important piece is this re-imagination, which is that young people come into the business and say, “Why does the note look like this?” It looks like a digital piece of paper, but haven’t you seen a Facebook wall or a post, or other ways of documenting one’s experience that has audio, video and weaves it together. Have you seen Wikipedia?” There’s this thing called collaborative documentation. The answer is, “Yeah. But what we did was to digitize the paper note. Let’s come up with a better way to do this thing that we are all trying to do.”

That’s when things get massively better, and what’s so remarkable about this is that it’s at the early stage of digitization. That’s incredibly weird because every other important industry went through this stage 20 years ago. But healthcare is a funny industry and the way we get paid and the regulatory environment means we were late to this particular meal. I think we will get there. I don’t know if it is a 5 to 10 year journey. I’m guessing that it’s more like 10 to 20, because it’s complicated.

Catalyze: You did work with Google, right?

Dr. Wachter: I was on Google’s health advisory board about 10 years ago.

Catalyze: I don’t know how long ago it was but I remember you talking about Google Wave as a potential EHR, because it was this collaborative way to document things. So as we think about re-imagining the practice, people are seeing the limitations in the tools that aren’t meeting the requirements or needs of providers. Where are those changes taking place? Are they taking place at organizations like UCSF or at huge healthcare organizations like the Tenant’s, HCA’s and Providence’s or small group practices and specialties? Where do you see that happening? Maybe it’s everywhere.

Dr. Wachter: I think it is everywhere. It’s important to reflect that we are in the midst of two seminal changes within the organization of healthcare. One of them is that we’ve essentially gone from analog to digital. That’s massively important over the next 10 to 20 years. In some ways, we are now under pressure to figure out how to deliver better, safer, more satisfying care at a lower cost. Again…WOW, you are just doing digitization of healthcare? You could have just as easily said, “WOW, you are just now creating a business environment where the winners of the organizations will be providing a better product at a lower cost? The answer is yes. That’s how I decide whether it is worth $3.45 to buy a tall mocha every morning, but not how we have decided to pay for healthcare. That’s changing very rapidly. It’s a combination of those two forces hitting each other — the value pressure that organizations are feeling and the digital revolution that creates the opportunity for tremendous innovation and improvement. Where is it coming from? I think we are seeing it everywhere.

At a place like UCSF, a big and terrific academic medical center, we had to go through the evolution of first putting in an electronic health record, and getting digital to work; we stumbled around like everyone else has. Our first digital system was one that we ended up pulling out and replacing with another. Today, we now have young people coming in and saying, “Why are we still using pagers?” Let’s figure out some new way of communicating with one another.” One of my young colleagues came in and developed such a tool and we just sold it to a company that’s going to commercialize it. That sort of thing where young, smart people are coming into the industry is happening. Obviously, I live in San Francisco and we are seeing it all over the place. Silicon Valley has finally discovered healthcare.

As you mentioned, I advised Google on their health strategy 10 years ago — don’t hold it against me, but it was a flop. You know when Google can’t figure it out it’s probably really hard. Google, Apple, Microsoft all tried little things in healthcare and then went to the sidelines. They are all back in now. Capital is flowing into healthcare, and enormously. There are kids in garages all over San Francisco trying to develop new apps and tools. And yes, HCA and Tenet and Kaiser are all developing their own tools or looking for tools that will help them digitize, not for digitizations’ sake, but digitize to meet the imperatives to deliver better, safer, cheaper care.

So, [healthcare] is 18 percent of the gross domestic product. We are in an industry that people care about deeply and the answer is now everybody is going to contribute to this. The answers are going to come from different places. Some of them will come from traditional companies, some of them will be from upstarts, some of them will come from companies in the digital space that didn’t do healthcare before, some of them will be from organizations themselves. But, the exciting part is that we are getting toward the middle or end of that foundational stage where the wires are there, the tracks have been laid, and you now have the environment where you can innovate and come up with cool ways of solving tough problems.

Catalyze: Do you see the value-based movement as a bigger driver over the next three to four years?

Dr. Wachter: I show a venn diagram and say, “These are the two transformational changes in healthcare — value pressure and the digitization of healthcare.” So when you ask me today in my role running a fairly large unit at UCSF, what I’m obsessing about is the value pressure; that’s my obsession. We have to figure out ways to provide better, safer, and more satisfying care at a lower cost. What’s amazing about that is that’s a new set of pressures. Previously it was just organized around keeping the building full. That’s the bigger deal today. What I say today is that 10 years from now, the bigger deal will be the digitization of healthcare. The reason for that is Disruption Innovation 101. The value pressure is real and causing important changes in the way we think about our work and organize ourselves. We are still doing it in relatively traditional ways. Upstart companies are developing a new primary care model. But a legacy organization like UCSF, it doesn’t challenge us to our core. We now need a Chief Quality Officer. We need to use data to understand the quality of what we are doing. We need to train our students about the cost of care. Those are big changes for us, but they don’t violate our paradigm about how we do our work and how we organize ourselves and what the org chart looks like and all of that. It’s just a new set of targets.

The digitization pressure will force us out of the box and unless we are completely different than every other industry, once the industry is digitized, that’s when Airbnb happens. That’s when Uber happens. It’s almost become a cliché, but once we are digitized, the legacy organizations, whether it’s UCSF or Aetna, or Walgreens, they’ll not be able to do their old things the same way or maybe in a slightly different way. There’s going to be upstarts that come in that say they aren’t playing by those rules and are going to do this thing completely different. Patients are going to control things that you, the legacy organizations used to control and unless we completely violate the rules of digital gravity, that’s going to happen and that’s where this is all going to get turned upside down. I think that’s probably 10 years away before we are wired enough to make that real. But I think we are beginning to see the early seedlings of that.

Catalyze: As we are seeing more solutions for value-based care, whether they be patient portals, scheduling systems, patient-reported data, Fitbits and all those sensors and recording more continuous data, you see the digitization happening in that 10 year period, one based on the traditional EHR and this whole other body of data that’s going to come from these other services and at that point, you’ve reached that digitization stage where you can have the Airbnb’s and those building completely new models, because at that point, that data is there and you can totally rethink the way we deliver care and services we deliver to patients, is that a good summary?

Dr. Wachter: I think that’s what I’m thinking, as well. My analogy here is the Transcontinental Railroad. What the HITECH Act, Meaningful Use and what the federal incentives did, what the federal incentives did was to get one set of tracks to be built — that’s called electronic health records, those are your prized-based hospitals, clinics, doctor-centric. They support the way we’ve organized healthcare. Traditionally, they provided a vehicle to get data in, and maybe eventually to get data out. That’s one set of tracks.

The thing that is in some ways more interesting is another set of tracks is being built and that is the Silicon Valley, consumer-facing Apple Watch kind of thing, which today is being overhyped. I don’t think we know what to do with a lot of that data yet. Companies in that space are claiming a lot of things and we’ll see. It’s happening and eventually it will be valuable. But that’s a completely separate set of tracks. At some point, someone lays the golden spike and that is one system, but to get the patient data to flow seamlessly, that act is no different in some ways than the patient going to see the doctor and answering a few questions. Then the opportunity for someone to tap into that system and show a completely different way of organizing is exciting. What typically happens in every industry is that process ends up democratizing the enterprise tremendously. More and more work is pushed away from professionals and big boxes, whether you call them stores or hospitals and toward people or families to do more things themselves.

Medicine has a peculiar flavor there. I just gave a TED Talk where I argued that the doctor as God is not the right model, but the patient as doctor is not either. This is not the same as managing your own financing or making your own plane reservations. You might have cancer; you want to talk to an expert who can comfort you. My son called me a few weeks ago and told me that he had diabetes and glaucoma. I said, “What are you talking about.” He had looked up his symptoms on WebMD and that’s what he thought he had. Of course, he is fine.

There will be a lot of bumping into walls as we get this right, but I think the premise is right. When there’s one stream of data that connects everything, then the concept of the EMR/EHR as a standalone enterprise thing will go away, as will the concept of consumer-based Fitbit data. It will be my data about my health that links in an intelligent way to the data base of understanding of medicine and science that is more accessible to patients in new ways. The healthcare system is not organized in any way, shape or form to deal with that yet.

Here’s where I go a little bit berserk. When I hear people say how wonderful it’s going to be; the patient is going to be wearing something on their wrist or something in their underwear that is going to measure their heart rate, their sweat and that data is going to beam to their primary care doctor who is going to be thrilled to get it, then I say, “What planet is that exactly on?” Because every primary care doctor I know, if they get one more piece of data on one of their 2000 patients will jump out the window. Clearly, we are not prepared for this now; there needs to be smart algorithms; there needs to be a layer of sort of air traffic controls that are monitoring a group of patients; there needs to be health coaches; there needs to be doctors somewhere in this. We don’t have the foggiest idea how this all works out, but I think that the infrastructure is one seamless data stream that covers patients to hospitals.

Catalyze: So we already have this EHR that has a ton of data, but then you start layering in Fitbit data or whatever and you get to a point where no one wants additional data, because there’s no way they can do anything with it in the time of the patient encounter to process it. You think it’s largely going to come down to different types of providers processing that data and also some form of intelligence around what the data actually means?

Dr. Wachter: I think it has to. Even if we’re not wealthy enough as a country to have enough people sitting there watching all these data flows and making sense of them, it’s smart algorithms or companies that will emerge to sit at the interface of all of this, looking at this data flow on a thousand diabetics. They may be able to say that diabetic number three and number 72 and number 126 look like they are going off the rails, let’s first send them a little notice to their smartphones to adjust the insulin or change the diet or something. And if that doesn’t work, they are going to get a call from a health coach. It will probably be a call like this — a video chat to see how they are doing, and if that doesn’t work, it gets escalated; it can’t all be done with human capital because it’s too expensive.

But there’s also this promise of this ubiquitous data somehow being wonderful. At some point it’s not. One thing we need to do is figure out is what parts of it are useful and what parts of it are not. Today, I really do want to know how my diabetic patients’ sugar is doing every hour or every minute, if I can. We know that’s useful and we know that’s actionable. For a patient with heart failure, I really do want to know their weight every day. For a healthy 28-year-old, I really don’t want to know their heart rate every minute. As far as a physician, that has absolutely no value.

Periodically I’ll hear how wonderful it will be that patients will have that information. I take care of patients who are hanging by a thread in intensive care units; they’ll die if they brush their teeth and I don’t need to know their heart rate every minute; that’s not useful. It’s not just managing the data; it’s a weed and chaff problem. Which of the data is useful and which are not? The science of sifting through it and making sense of it is complicated but it’s not that much more work. If Amazon can say that you bought that book like this book, then we can probably say that patients like this whose sugars have gone from this to this, seem to have this kind of problem and need this kind of intervention and that seems to do well. We’ll build that into our system, but there’s no system today that’s even close to that. But that has to be the future state.

Catalyze: I like the analogy that we’ve laid the Transcontinental Railroad but now we have these parallel tracks that are being laid and at some point someone will hit the golden spike and we’ll end up with one track that everybody will tap into….

Dr. Wachter: Let me break in here to say that that somebody is likely to be a company, but the government probably has to enable or make that happen. That will be an interesting question, because in other industries it did happen on its own. Betamax happened and you think about standards in other industries. It’s a complex and interesting question, whether medicine is really the same as other industries that eventually went to a standard and it worked and the government really didn’t have to intervene, where in medicine there are some peculiarities about the way people are paid and the way money flows and the way insurance works. It may be necessary to sort of bang people’s heads together and say let’s agree on a standard. I think that’s an open question.

Catalyze: You mention potentially the government enabling a company to lay the golden spike. Can you talk a little bit about that? How you could see that happening with the government.

Dr. Wachter: What the government’s role is likely to be here is as a convener and maybe a prodder to say c’mon folks, there is a public good in standards that allow patient data and enterprise-developed data to come together. That’s a good thing and yet the market is not making that happen fast enough. We want this to happen and we’ll use the tools of government to provide incentives or make you do it or whatever. They’ll never say, “Here is the company that we’ve selected to do it.” They’ll say, “Here is what we want this to look like.” Companies will then enter that space and vie to be the connector.

I kind of agree with you about Meaningful Use. I think the role of the government in all of this has to be limited and thoughtful. It can’t be that the government has no role. You could say that the government should keep their hands off it. I remember a debate in the mid-90s when Bob Dole was running for president. He said, “I’m going to make sure that government keeps its hands out of Medicare.” I said to myself, the government runs Medicare; it’s a government-run program. What does that mean? The government pays for half the healthcare in this country; the government will have a role in this and will have a seat at the table.

To the extent that they are not making important things happen, the government has a role in making that so. The government had a massive role in the early days of the Internet. The government was DARPA. But then, quite wisely, pulled way back and the government had little to do with the day-to-day running of the Internet. What I think happened here is that the government got deeply involved into the healthcare IT because they were throwing 30 billion dollars into the pot to make healthcare IT happen. That was okay, but now they need to pull way back and ask the question of what is the proper role of government, forgetting about the 30 billion — it’s spent. Now we have a digital system. What should the government do that only the government can do and where should they get out of the way. Meaningful Use, with what the government is asking healthcare vendors and systems to do, it’s way too far in the weeds and way too regulatory and it’s going to get in the way of innovation that we need.

Catalyze: When you get to that point where you are convening those tracks, are those parallel tracks going to merge into a whole new track?

Dr. Wachter: I think it’s anybody’s guess. I think the answer to that has less to do with the technical parts. It has more to do with the larger issues in healthcare and health policy. As the system devolves into a system where patients really are managing their own health to a large degree, so that the idea of patient ownership of the medical record, which sounds good, but what does that mean if I still need to see a doctor to get care?

I’m working with companies and big enterprises that have to troll my data or at least have full access to it. Someone asked that question: Does this world of patient-centric care and philosophy do so much that the thought of an enterprise-based EHR become sort of antiquated. It’s only logical that patients will control all their data because those are the people making the decisions. If I had to guess, I would say that that’s not going to be the future state. I would say that the future state is likely to be that the central organizational entity still is something called an electronic health record but will look very different. We’ll have data collected from patients and it will connect to your pharmacy and will connect to your gym, if that’s what you want it to do. The reason for that has less to do with technology and more to do with my belief that patients are still going to need the traditional parts of healthcare systems that have morphed in ways that still going to see doctors and still going to hospitals and pharmacies and still have health insurance that contracts with some entity that is responsible for their care and wellness. And these changes are of course in the EHR being a vehicle to collect episodic data has to change because the data won’t be episodic, it will be much more holistic. But I don’t think that the fundamental premise of there being some business entity that is responsible for working with you that has accountability for your outcomes and that becomes the fundamental hub of your data. I think that model will be with us for the next 10-20 years. Beyond that, who knows?

Catalyze: Are there any particular technologies or categories of things that you are excited about?

Dr. Wachter: I think the things that I’m most excited about are new tools that begin to sit at that interface between the patient and the system. I’m on the advisory board of a company called Twine that basically has a tool developed out of MIT that collects data on a population of diabetic patients or emphysema and matches you with health coaches. So it essentially has created what I was talking about earlier where the data is being collected but there is a system that has smart algorithms and coaches to basically help people manage their health. But it’s embedded in a healthcare system. I think you can take this patient-enabled and patient-centric idea a bit too far. If I’m a diabetic or an emphysema patient or heart-failure patient, I want to manage as much as I can myself, but I need my system, my smartphone to do that, but at some level, somebody is prescribing my medicines. There’s a hospital I go to if I get really sick. I want a cardiologist in the loop here somewhere. We don’t want to kill this ecosystem that is completely independent of the health delivery enterprise. I want something that syncs up with it. So the apps and the tools that make that real are the ones I’m most excited about.

Catalyze: I just talked to a close friend at Carolinas Health and their using Twine and exactly what you were describing, Twine sort of empowers an entire care team, but sitting between the care team.

Dr. Wachter: Those are very essential elements that if you just start with the data and the collection, the algorithms and you don’t think hard about the workflow and the people and what happens if the patients aren’t doing well and fall off your algorithm, then you’ve screwed it up. You’ve taken the patient part of the way there, but you’re almost better if you hadn’t done that because the ones that now fall off the path are the ones that I now worry about. So the tools have to pay attention...and that’s true in all sorts of other things, to say that we are going to be alerting the doctor when X or Y happens, then you have to look at the life of the doctor. If the doctor is getting a thousand alerts in an hour, they are going to ignore all of them. So it’s sort of a thoughtful collection of the data piece and the people and the workflow and I think Twine is a good example of a company that’s thinking hard about how to weave all of that together in a thoughtful way.

Catalyze: I did have a question related to that. I’m not sure of all the conditions that Twine works with but it’s always made sense to me that you’d have these solutions pop up for diabetes or for asthma or mental health, but how do you manage working with five different care management platforms for different conditions?

Dr. Wachter: Really those are two different questions. One is the system IT piece of that, which is as we bring in new tools, how do they integrate with our central core EHR — that’s a big question and we are still grappling with that. When we decided to go to Epic at UCSF, it was partly because one of the things that Epic sells is: We do everything. And it does, but it doesn’t do everything perfectly well. We now have to move to a world where we are bringing in best of breed products and you have to figure out how they integrate. It can’t be that people are clicking out of your core system to a completely different system that has a different look and feel. It has to feel like you are in the same environment, it’s just that the data is coming in through a different tool. That’s an IT job to make that happen. That’s a surmountable problem where we just figure out the connections that weave this together. That’s beginning to happen.

The second issue is in some ways more complicated. I don’t have enough patients that are cooperative enough to only have one thing wrong with them. They actually have five things wrong with them. They have diabetes, and heart failure and there you have a problem. The tool is just a diabetes tool and that has its own problems of integration but in some ways the deeper more clinical problem is sometimes the recommendations around problem one are very different and conflict with the recommendations for problem four. That’s partly why we still need doctors, because I don’t think the computers are going to get smart enough any time soon to think about the algorithm to use all of that together, but eventually they will. That is a tractable problem. In your big data set, there will be enough patients who have diabetes and emphysema and heart failure that you could begin to look at the best way to take care of those patients. A lot of patients, a lot of data and a lot of smart people analyzing it and we are no where near that now. So there’s a silo problem in both the tools and in some ways the more clinical problem of recommendations with problem A almost conflict with problem C and the patient actually has problem A and problem C.

Catalyze: Do you see any immediate changes over the next couple of years that will be enabled within Epic so that these big enterprise EHR systems will improve the experience within Epic?

Dr. Wachter: Yeah. Epic is getting critiqued these days and partly because they are the winner of the sweepstakes so far, and that’s kind of the natural thing that happens. I spent some time at the headquarters and got deeply into the way the company does its business; they are bunch of smart people and the reason that they’ve won so far is that they’ve made some very good choices. They built the best system, if you are a big complex health system. Most of the systems that have done RFPs and made a choice about which to buy, particularly for big systems; it’s not because of a conspiracy, it’s a business choice. They are smart folks. They all have iPhones and understand the ideas of bolt-ons and they are working on an environment where you can create a new innovative tool and link it in. They understand the risk of being the big gorilla in an industry with other upstarts. They are taking it on the chin politically.

What I’m seeing is that the newer versions of Epic are a little bit more attractive. They are building tools for the problems that we need to solve like, population health. They are beginning to grapple with the fundamental problems like alert fatigue and some of the poor user interfaces. The patient portal is now decent. It’s good, but not too great. It’s not what you are used to in other interfacing portals, but it’s good.

It’s an open question as to where this world will be in 10 years. I think Epic will be doing good in 10 years. I think it’s a smart, innovative company and the barriers to entry, there’s no startup in a garage in San Francisco that’s going to take on Epic for a five billion dollar enterprise like UCSF Medical Center. Places like us have to go with traditional and it’s the regulatory barriers, so what you are going to see are the innovators like Athena who have come up with a more nimble solution for a small practice. That’s where you’ll see more innovations — in small practices and certainly around consumer-facing IT, where it’s essentially been a vacuum. But I think the Epic’s of the world are probably in good shape for at least the next 5 to 10 years. Beyond that is an open question.

Catalyze: We are 10 minutes down the road from Epic's campus, which as you’ve seen, is quite impressive.

Dr. Wachter: It is quite impressive.

Catalyze: They’ve done well over the last 10 years.

Dr. Wachter: Yes, they’ve done well because they’ve made some smart decisions that turned out to be prescient. When the things were being made you could have argued that that wasn’t the right call to try to do everything for everybody and not to do partnerships and not to take on debt, sell stock or not to go public — a hundred calls that in retrospect turned out very well, but God bless them, they are winning because they produced the best product at a time and they got lucky. They had the best product at the time when everybody needed to buy EHRs and their challenge is the challenge of legacy companies that are at the top of the heap, which is they have a big target on their back and their front. The history of this is that big companies like this don’t innovate fast enough as they try to protect their franchise, so what happens will be an interesting story.

Catalyze: Yes, it will be fun to watch. Maybe in your next book you can cover the transition in 5-10 years. Thank you so much Dr. Wachter.

Dr. Wachter: It was a lot of fun. Have a good day everybody.