The Future of the Healthcare Sector: Richard Migliori, Keynote Speaker

We share with you the perspective of UnitedHealth Group on innovation in healthcare, and it's probably shared by many peers, but I would contend that it's also shared across our industry, including [inaudible]. Innovation in the healthcare space is our lifeblood, particularly at UnitedHealth Group. It's our basis for maintaining an adaptive capability so that we can weather the storm and also anticipate future trends and then be proactive and help shape trends. For us, the innovation has to have three components. First, it's got to be strategic. What I mean by that, it's got to be able to improve clinical or economic outcomes, improve access, or improve patient experience. Second, it's got to be compliant, and I'm not just talking about regulatory compliance or compliance with information. It also has to be compliant with evidence-based medicine; after all, we're in the healthcare business. Then the third component, it has to be scalable. We serve 70 million people; there are 313 million Americans. In order for these types of programs to have merit, they can't be nurses on phones. It's got to be something that can spread and to create. The final aspect of things, if we're going to be true to our soul, is that we have to leverage our core competencies, of which we have three. One is the ability to organize and to manage within a healthcare delivery system of providers; we work with 85% of America's healthcare delivery providers, hospitals, physicians, pharmacies, etc. The second, it can leverage technology for transactions. We manage some 52 billion transactions a year. So, we have to be able to have that kind of ability and that experience could help drive some of our thinking. And then the final component, it has to be able to leverage knowledge. It's no longer just good enough to keep all of your claims in a shoebox so that you can pass a claims audit. Those claims tell you something; they tell you about consumption in healthcare; who's getting it, what are they getting, and how effective it's being delivered. But let's get to that first point. Let's be strategic.

Healthcare reform brings us great promise. It already has a clear pathway to expanding coverage to more people. What it doesn't have to date is a pathway to sustainability, and that sustainability is only going to come by the ability to manage costs. This is work that comes from Eliot Fisher, it's out of the Dartmouth Group, and it demonstrates for you a couple of things. First, that there's a dramatic variation, disparity in the cost of care per capita across the United States. The fact is that there's a twofold difference between Minneapolis and Manhattan, and they can see scattered across the country different levels of expenditure. That expenditure, first and foremost, is not just a reflection of the unit cost per service, it's far deeper than that; but in all areas, it continues to grow at some 3 to 4% per annum over the course of this 14-year period. The driving focus of that comes from this dynamic. It's the decisions made on a day-to-day basis in the exam rooms, in the emergency rooms, and the operating rooms across America between these two people, and that decision that they're about to make is fully dependent upon the quality of information they have about the patient, the physician's awareness and willingness to employ evidence-based medical standards, and third, their ability to track progress or even compliance. When you start looking at the American healthcare system's ability to make decisions, however, you find the source of the variation that Eliot Fisher was showing. This is work that goes back to Jack Wennberg and it gets repeated on a regular basis. The only thing that changes about these charts is that the absolute number of cases per year grows. These are America's ten most common surgical procedures, and what they represent are regional variation in the utilization rates of these procedures where this is a community or hospital referral region in Louisiana versus in Minneapolis for doing angioplasty. I know this data is a little dated; this is getting repeated year over year. There's a [inaudible] difference here in the likelihood that this procedure will be performed. That is not driven by differences in the presence of disease or the epidemiology of those communities. That 12-fold difference is small area variation, driven by the decision-making processes that are different in those environments. Let me say it a little differently. In this country, you are more likely to have your plan of care influenced by your zip code than the stage of your disease. We have an issue with utilization, we have an issue with utilization driven by decision making. When you talk to Eliot Fisher, his estimate is that 30% of our expenditures could be reduced if we would have more of a normalization of this type of variation.

Now, some other evidence about decision-making quality in the country is here. This is Elizabeth McGlynn's work where she looked at tens of thousands of decisions in her clinical practice and found that 45% of the time, the decisions made by the practitioner had no basis in science. In fact, in 11% of those decisions, they are actually potentially harmful. Here's the other person in the room; the behavior of the individual. We talked about the 55% being coherent with medicine, however, 1 out of 5 people handed a prescription never fill it, though they come back and tell the physician that they did. Of those people who do fill it, 50% of them will stop taking it within six months, particularly if it's for an asymptomatic cause. And they can see that the expenditures, and this is probably an underestimate of hospitalizations in this country because of poor compliance; then we have this influence on the decision-making processes. About half of the growth in the healthcare expenditure over more broad indications of price than CPI comes from new innovation, but unlike what you probably would anticipate coming out of me, this is an absolute necessity. We cannot continue to sustain the healthcare system without the introduction of these new technologies. We have to be able to achieve outcomes at a lower cost, and you're only going to get there with better tools than we have currently, but it does provide a new air of complexity for people making decisions. So, as we started looking at creating solutions around this area, we recognized we needed to achieve three things. One of them, we had to make the healthcare system more connected; it's far too fragmented and I won't go into much more than that. Certainly, we have to make decision-making processes more intelligent; better information and I'll get into that in a second. And certainly, we need to have all participants, whether they're consuming care, delivering care, paying for care, or adding new devices or pills or whatever to the system, or even regulating it, getting people aligned is going to be ultimate. So let's go back to that exam room. There are going to be three things that we felt that we could bring to the table, one of which is if we could use our data gathering capabilities to help inform this dyad better so that they know what's going on in this guy's background. Second, if she's out of medical school more than five years, medical information has probably doubled. While she's been working on her patient base, she may not be keeping up with her knowledge base and we can automate that for her. And finally, the ability to track what's happening because we saw the issue [inaudible] dietary restrictions, he may not be taking his pills. In order to do that, we built out a system, a platform called eSync. What eSync does is it looks at our entire population of 70 million Americans and collects all data that we have on them, whether it's claim data, whether it's pharmacy data, whether it's clinical laboratory data, whether it's medical notes, etc., and builds for each individual a composite view of their lives. And then we use analytic tools through our rules engine to first organize that data to infer what conditions this individual may have, building out their problem list, and then examining using 950 evidence-based rules, what we—for each condition will look at what kind of care they had delivered and where, versus what was expected on evidence-based rules. And then for errors of omission, errors of commission, and build out for each individual something called a personal action plan, which shows the opportunities to develop more conformity with evidence-based standards. Well, in order to get that plan to do something, we ended up finding that we are most effective if we were sharing that with the consumer themselves, things for them to do, but also share that same to-do list with their physician, all of them, and if we're involved because we have nurse managers and the like, then with their own staff, so you have everybody working on the same concurrent platform. Everybody on the same prioritized to-do list. When we did that, as we demonstrate in blue, the likelihoods of conformity to evidence-based standards improved an incremental 20%. So we can influence the power of a decision. The issue that we face, however, is getting people to adapt it and to adopt those changes so they can adapt their behaviors. People are different and one of the things that we recognized is that we need to become more of an integral part of people's lifestyle. We have all of this data, but to get people to look at it, we realize we have to be different. Whether we are reaching out to them by having our nurses reaching out to our sick and most ill people, or gaining other people who may be interested themselves in portals or getting into their lives with mobile applications, or for those who still find themselves a victim of snail mail, to be able to reach them that way so that everybody gets to see that personalized to-do list. Let me give you a couple of examples of that. This is the nurse's desktop. When they go in in the morning, they get to see who their patient is and then they start looking at some opportunities to change behavior. The type of behaviors that we look for are things they can be doing about are they seeing the right type of doctor, are they on the right medication, are they on the right care platform, or even are they living the right lifestyle. In this case, let's look at an example where the nurse decided that we're going to look at the physician that they've chosen for the problem. This person has diabetes; clearly this person did not have an internist. One of the things that the system does is to generate internists who are nearby and also provides quality ratings, based on our experience. What's this physician's natural history of conforming to evidence-based standards in their practice and doing it efficiently? And so selecting physicians and helping the patient by providing them mapping instructions, and then we'll even go so far as to make the appointments for them. But if we're going to do that, we also need to be able to provide that in the hands of individuals who do this themselves, who can only reach out to as many as maybe 5, 6, or 7% of the population who are driving 40%. For the rest of the population, it's self-service and encouraging people to do so, requiring the way in which we present data online. One of the things we've done is to be able to show people that their personal action plan—by showing them where they stand. What is their medical history and putting in front of them in the [inaudible], the things that they need to do to correct it, and the layer called 'I do', and then what we're seeing more and more from employers and even governments, is applying the layer that shows tracking with financial incentives and a variety of other incentives based on some sound behavioral economic theory to get this 'I am, I do, I get' philosophy so we can drive behavior. And it's had some rather encouraging results, but we also have to be able to get to people and share this same personalized action plan, things they can do to make better decisions by putting it in their pocket. One of the things to do is to take that very same physician identifying thing, and put it right into a mobile application. In fact, this is available for you if you want, just merely load it down in your app store, whether you're in a Droid or a Blackberry or an Apple phone, this application is free. It will show you whether you're looking for a physician, facility, or a healthcare center, the list of physicians by specialty and their [inaudible] will make the appointment for you if you're one of our members. But we also wanted to be able to get into lifestyle and this is a new platform that we've created again being driven off those same personalized action plans, it's called OptimizeMe. It is also free, you can download this thing, it's on Droid based as well as on the Windows Phone 7; it's a way in which people socialize. One of the things we recognize is that if people's relationships are important, social distances are important, the Framingham study even showed it; well it showed that if you have a first-degree contact that is a family member, or a close personal friend who becomes obese, your chances of becoming obese go up 57%. This isn't that fat people get together, this is that if there's a change in behavior in somebody in your group, there's a 57% chance change in that behavior. It goes likewise the other way. People losing weight will create a collective. So what this is, is a way of creating for people challenges that's a four squares game, it's a platform that shows a variety of challenges and badging and a variety of other things, but is also a means by which people can administer incentive plans through their employer for wellness activities. The final thing that we're doing is moving into this space, create a greater connectivity between patient and their physician, with the evolution of a variety of the different tablet devices and smart phones; the ability to connect patient-centered activity with physician under the patient's control where they can release data going forward. As this evolves, it gets us to go to one last place that I'll finish up with, is the ability to take this intelligence and technology and use it to solve a problem with access. We are going to be 45,000 primary care physicians short in this country by 2015, according to the American Association of Medical Colleges. There's not enough healthcare access to go around. Look at what's being reported now in Massachusetts. Massachusetts Medical Society recognizes now that when they increased coverage around this plan, one of the things that happened is that the waiting time for physicians extended dramatically and emergency room utilization went up. It now takes 49 days to get a primary care appointment. The healthcare system cannot grow fast enough to accommodate the new demand that's going to be uncovered with the expansion of coverage. We need to find new solutions. One of the new solutions is to take people with current clinical licenses and increase their game. How? By sharing information; by sharing technology using mobile applications, using telemedicine solutions as we've started, and as well as even making these things as a rotating service. The ability is to provide a remote clinician at a given level with support of a more specialized clinician in a more centralized setting and by using telecommunications to spread that intellect.

[inaudible] opportunities for collaboration of partnership between medical device and payers. In the background here, we have a speaker that comes to our class. He worked on the payer side then he moved to the medical device side, and he used to say—his joke that when you're on the payer side you look into the medical device industry and you're thinking that oh geez, here's another expensive technology that is going to drive our costs up. If you are on the medical device side, you are looking to the payers and you're thinking, oh boy, there are other people out there with all these rules that are making things difficult for us. So, [inaudible] actual collaboration as opposed to the traditional [inaudible] between payers and medical devices [inaudible].

Well, I would agree with everything that John has said. The needs and the necessities and the point where we are now, what we got to create is sustainability for this healthcare modernization are required of us of sitting on the same side of the table. It's not just, can we get along, and it's an absolute necessity. We do not have the capabilities and skills that we need or the technologies that we need to go even further. Look at any other industry, the investment in technologies to make it go better, faster, more effective and more timely, etc., are things that medicine needs right now. Let me give you a case in point. There's a drug called Gleevec. It costs $36,000 a year. However, what's happened is it's taken a disease that affects 4,500 Americans called chronic myelogenous leukemia and is essentially curable, provided you take the pill every day. The problem is that when you look at a $36,000 drug, if you had just a plain old view and say how can you spend that much on pharmacy, you start looking at it from a very different perspective and we start looking at what happened to the total cost for these individuals. People who are not on the drug cost about $110,000 a year and they died. Their costs were going into ICU visits and were going to bone marrow transplants. When people are fully compliant with the drug, the cost of the care was cut in half. Now granted, $36,000 is going to [inaudible] but that's okay because the cost of care and the survivorship for the individual is dramatically improved. We now have people dedicated to making sure that anybody saying no is on the damn drug and is taking it.

One of the—that's a great question. So, how do physicians react to having somebody helping them make decisions? What we've learned is that we really had to take different perspectives ourselves. We used to inspect decisions in our industry, now what we're doing is trying to support them. We're trying to be a good [inaudible] to that dynamic that I was pointing out. One of the things that we do for physicians is to provide them with accessed information that would otherwise not have. Their information in their exam rooms is going to be far more detailed than we would ever have, but what they don't have is did that patient go to the pharmacy or is that 1 in 5 that never fills or is that 50% of people who stop taking things. By providing them with that information, the decision to stop a medication or not is at least something that's visible, and it's something that is deliberate, and that it's not an accident of bad behavior.

Thank you for the question. One of the things—we're still in a learning phase here. Our first—this has been going on for about 24 months, the last 18 months or so, we've really been defining what the scope of practice of telemedicine is; working with Cisco, working with AT&T, Verizon and a few others, we've been able to start looking at how much of medicine can be practiced here. The second place thing that we've done is looking at what kind of installations make sense? One of the places where it makes immediate sense we're using ourselves, and that's for onsite clinics for large employer facilities, and what it does is allow a nurse to be mindful of a practice, do acute care and follow-up care and even some wellness care, but when they need it to be able to summon a physician in variable cost basis, to come and co-manage the patient. Another application is what we're doing in Colorado in where we have four rural clinics that hook up with the Centura Health System in which the primary care physician can get on the device, on the telepresencing capability, supplemented with the digital diagnostic equipment can summon any type of specialist they need so that they can have a three-way consultation with a patient, the primary care physician locally and the specialist virtually. What this has allowed for is for us to be able to recognize when there's a need to transfer the patient, when there's a need to be able to retain somebody at home, or certainly to bring in that intellectual capacity. The economics of these models [inaudible] the end is too small, but certainly we see broader and broader applications for this. One simple application is putting this through a secure web enablement to allow families to access this on an on-demand basis and a self-pay basis out of their home.

It's a great question. If you look on a macroeconomic basis, if you look at doing the things that we talk about doing within healthcare, we are going to see a reduction in utilization, is largely going to hit the hospital the greatest because what we're talking about doing is minimally invasive interventions to manage coronary artery disease, preventing readmissions for the 1 in 5 Medicare beneficiaries that gets readmitted to a hospital within 30 days because of some error at the time of transition to the home setting. It's going to be on relying more heavily on experts in given areas of surgical complexity so that you can mitigate the risk of complications and the like, but when you look at the system in general, you will look at the stuff that Eliot Fisher was pointing out with those surgical differences, it's going to hit the hospitals the greatest, and we can't have that happen in just a linear fashion. It's going to call for us to be looking at reimbursement models that are different, more episode-based payments for care, so that way, just as Michael Porter talked about in his redefining healthcare is that when you start to build an episode-based reimbursement, those people supplying care, whether they are the professional clinician, the institution, the pharmaceutical agencies, the medical device manufacturers, if they can share in the savings that are created by the investments they're making, we're going to have for a better system. This pay-as-you-go, pay-for-complication system we have is something that's going to be a deterrent.

Okay, preventive care—one of the innovations that we've had, and it's been a focus, is to get people engaged. If people understood four numbers in their life and they control four numbers, you'd see a dramatic improvement in health status. Those four numbers are body mass index, blood pressure, LDL, and fasting blood sugar. As John had mentioned, there are 25 million diabetics. There are more than twice as many pre-diabetics; 11% of those people convert to full diabetes. One of the things that we put into the system for our own employees, we test everything on the 90,000 of our employees and the 130,000 dependents, is to put in a program in which your payment for your health coverage, your premium that you pay out of your paycheck for your health coverage, is influenced by whether you had the testing for those four items, and if you had those four items, if any of them are abnormal that you're in a program to mitigate that risk. When we did it, we found some very interesting things. First, we found that we had 7,007 diabetics out of 128,000 covered lives. 77,000 adults, okay. 443 of them didn't know they had it. As a result of doing that and getting people into weight loss programs of which we have a 600% increase, because they know if they didn't join, they're going to spend $900 more dollars next year for their health insurance, we end up having 45% of those people lose on average, 5% of their body weight, we end up spending 19.6% less that year for diabetes expenditures, we had 10% fewer myocardial infarctions, we had a 30% increase in people going to their primary care physician, we had a 15 to 30% increase in people getting the various types of cancer screenings that were indicated.

Maybe if I just ask a question of the audience, and raise your hand, how many of you know your four numbers? So it's about 10%, 20% of the audience, so the challenge is not, think of us as we're not your average consumer of healthcare, because virtually every one of you being here, we're above average in terms of our appreciation, not only of the issue, but our overall, whether you call it economic or educational status, and the challenge is in this room, only 20% of you know the four numbers. The challenge is how do we get the vast majority of patients and people across, not only the U.S., but globally, understand their four numbers; I understand only 20% of you do.

So let's start off with the premise first. Per capita expenditure is one of the highest in the world, but it's not the highest. When we talk about efficiency, we also have to talk about what expectations are for the community that we're being served, but one of the things that would make the change to become 'efficient' is to change what American society expects out of healthcare, and that's a societal question that I don't think has been wrestled to the ground yet. We can become more sufficient, but to become like somebody else, from our own experience and our own work, we have 10% of our employee base is out in the rest of the world providing consultation to health systems that think they need to improve. I won't be in a position where I would naturally impugn what we're doing in this country as Americans compared to what's happening elsewhere. There are things that we need to fix. One thing that we are bringing to life is something called comparative effectiveness research. What people actually get to see, not just whether or not a device or drug is safe and effective, but how it compares, and what's the comparative cost? Now, one of the things we could do in this country is to do what happens to some places in Europe where they draw the line for you and below that line you cannot have access to that item. I don't think American society will tolerate that. I think what they'd rather do is to be able to make choices based on what's available and to use that as a basis; let's use comparative effectiveness for transparency, rather than for limitation and restriction.

We have a wealth of unused expertise in this country because we do not ask non-physician clinicians to practice at the full capability of their license. This kind of capability is something we think that can help address it. We will not be able to train primary care physicians fast enough to keep up with demand. We really do have to change these models. Some of the things you're seeing around these accountable care organizations and primary care medical homes, they're great means, but what's fundamentally different about them is that the work process and the coordination of care is something that is a more shared practice where people get to maximize their privileges with what they're given and their licensed accomplices. There is plenty of capacity within the system, but it causes us to change the model and we very much encourage it.