Lonny Reisman, MD, returns to the start-up life after several years as chief medical officer of Aetna. In his new venture, Reisman is continuing his long-time pursuit of trying to find ways to keep the chronically ill as well as possible.
Steve Krupa: Welcome to the Breaking Health Podcast. I’m here with Lonny Riesman, the CEO of HealthReveal. Welcome, Lonny.
Lonny Reisman: Thank you very much. Good to be with you.
SK: Glad to have you. Good to see you again, talk to you again. You’ve come back to the startup world. A lot of people would say, What, it wasn’t good enough to be the Chief Medical Officer of Aetna and have one prior successful startup? What brings you back, and what were your thoughts about starting a company again?
LR: Well, first, it’s great to be talking to you. I had a wonderful experience at Aetna following the acquisition of ActiveHealth Management, which you were so helpful with. So it’s nice to be talking to you again. I think the experience as Chief Medical Officer taught me a lot. But one of the things that it taught me was that there remains a significant need for innovative ways to not only improve quality in healthcare across the United States or the globe, but that by improving quality, there’s significant efforts that would also contribute to lowering medical costs. So I saw a need and thought that the most effective way for me to address that need would be to go back to my roots as an entrepreneur. So I left Aetna in April of 2014, about a year and a half ago, and it’s been a lot of fun. It’s been challenging, but so far pretty rewarding, and I’m not unhappy with my decision.
SK: Yeah. I do want to eventually talk to you about life as a Chief Medical Officer at a large health plan. But what’s very interesting to me – you know I’ve had the opportunity to talk to you about your new business, so I really want to talk about what got you here and what’s interesting about the business. In reading through some of your ideas, it seems to me that you want to cut a line between the concept of efficiency and effectiveness in the care delivery model. Is that a good way to think about what you’re doing?
LR: I think so. Let me just define what I mean by efficiency as opposed to efficacy, both of which are very worthwhile. I’m basically complimenting efficiency with efficacy efforts. But I would define efficiency as a well-founded desire to eliminate waste in the healthcare system. So the number that you’re familiar with is $750 billion, that resulting from excessive and unnecessary utilization, redundancies, inadequate or less than optimal settings for certain types of treatment, all very worthwhile things. The efficacy part of the equation, which I think has gotten less attention, relates to the application of the best evidence relevant to a patient’s clinical status or physiology at a moment in time when that evidence will have the most potent effect on the patient, and prevent adverse outcomes associated with chronic diseases like strokes and heart attacks and end stage renal disease, and very costly and unfortunate situations for patients. So as we’ve been focused on efficiency, my sense is that we haven’t really been adequately looking at ways or any number of ways to basically bridge what we’ve learned and expressed in the clinical literature to the needs of the particular patient at a particular moment in time, when a particular therapeutic or diagnostic intervention would be most helpful.
SK: So at the end of the day, when we get older and, by the way, one of the most current conversations that I have with people is just how old are we going to get. So I’ll give you an opportunity to answer that question. How old are we going to get, Lonny? Are we going to be living until we’re 100 based on medical technology today?
LR: I think we’ll be living longer than we are. We’ve seen improvements and increases of longevity, so how far it goes, I don’t know. But I’m optimistic, but I say that with some hesitation about greater longevity because obviously you’ve got to cope with the realities of long term care and assisting the elderly in maintaining quality of life. But I think it’s fair to say that there will be interventions that will be developed that will prolong life, and the question is at what cost relative to quality of life and actual economic costs.
SK: Right. Yeah, I mean I’ve noticed since I crossed the 50 mark – or actually when I crossed the 40 mark, but more so when I crossed the 50 mark that there’s a lot of activity when I go to the doctor now. It’s not just come on in here, let me listen to your chest and do a blood panel. There are preventative processes in place that are specific to really the 3 things that affect men: prostate cancer, colon cancer, and cardiovascular disease are the primary killers besides cancer, obviously. And they focus a lot on trying to find that in me when I’m at the doctor’s office. But I think you’re suggesting, if I understand what you’re thinking about is that there’s efficacy that needs to take place when I’m not in the doctor’s office, where I’ve already presented some symptoms that might indicate that I’m on my way towards a severe, chronic disease. Is that –
LR: I think that’s right. So I think there are a couple of things. One is I make a distinction between screening for disease and, as you know, there’s some controversy about the ultimate value of some of the screenings that we’re doing. There have been changes, for example, in the timing and frequency of mammography in breast cancer for women. But without getting into that debate, I’ll characterize that as preventing screening, and for the purposes of this conversation, let’s agree that that’s highly worthwhile. The other type of screening, which I’m focusing on, again is a compliment to what you mentioned, is screening that would identify issues that are imminent. So hopefully if you have any of the diseases that you mentioned, you’re not going to have them for a very long time, and if you do develop them, they’ll be caught early. I’m worried about people with manifest chronic disease where there are certain signals, physiologic signals that portend a very bad outcome. So for example, if someone has a certain type of abnormal heart rhythm, or in the right context, fever or change in oxygenation of their blood or even a change in blood pressure, high or low, that information, that physiologic sort of aberration from their usual pattern may be predictive of a bad outcome. Now where we differ with – relative to what others are doing with predictive modeling is we’re actually accessing data from a variety of sources about that particular patient at that moment in time, and fully understanding the implications and consequences of that physiologic aberration, and actually being prescriptive about diagnostic or therapeutic interventions that might help that patient from evolving to the next stage in their disease, which might be a stroke or a heart attack or a horrible pulmonary infection, whatever the issue might be. So the type of screening I’m talking about I would liken to something like detecting fraud with your credit card, where there aren’t just periodic looks, but your activity is being looked at all the time, and to the extent that there’s an aberration in your usual spending pattern, for example, that aberration is considered in the context of your overall spending habits. And to the extent that there’s something that can’t be reconciled, that’s fraud and your credit card is shut down. What I’m suggesting is that if we can identify a change in blood pressure, say a fallen blood pressure, well, maybe that’s a good thing because you’ve just started on anti-hypertensive therapy; or maybe it’s a bad thing because you have a critical heart valve disease, and that fallen blood pressure represents the moment at which that valve needs to be replaced. So the question becomes is it better to identify the need to replace that valve while you’re basically stable, other than the fallen blood pressure, as opposed to showing up in the hospital when you’ve got heart failure or fainted or had something that’s much more substantial in terms of the risk to you as a patient, but also the associated costs with managing those complications. So that’s the distinction I’d make between sort of traditional screening. What we’re talking about, which again reflects a moment where a bad outcome is imminent, and I make a distinction between saying this is a high risk event; let’s refer to the case to a care management team, as opposed to adding to that equation the ability to be prescriptive about what needs to be done to solve the problem.
SL: And of course, if I end up having some of these things, a stroke, heart attacks, but with stroke in particular I end up surviving the event, then after that moment in time, my life has changed irreversibly, and the cost of my annual healthcare has gone up by probably an order of magnitude or more.
SK: And it would seem that that would then indicated that as we are aging, back to our first question, how long are we going to live, we are managing, and I think it’s fascinating, to keep people with significant disease burdens alive for a very long time. So if we have a population that’s getting older and their disease burden is getting greater, then we’re sort of running up against an ever-increasing cost of healthcare that’s going to be very difficult for us to handle.
LR: That’s right. And there are people, frankly, who would cynically say, well, it’s just cheaper to let people die.
SK: No politicians are going to say that.
LR: Right, and hopefully neither of us will say it.
LR: But the reality is, and this gets us to an end of life discussion, dying isn’t necessarily cheap. And rather than dying suddenly, for example, with the stroke example, you might have years of rehabilitation, years of disability, years of very, very significant requirements in terms of home care. So the point that I’m making is that to the extent that people carry a disease burden, but irreversible damage has yet to occur, if we can intervene between that transition from basic sort of stable, chronic disease to irreversible complications associated with chronic disease, end stage heart failure, end stage COPD, end state renal disease, if we in fact can pre-empt that transition, not only is the life of the patient in terms of quality preserved, but you don’t incur a lot of the costs that will be associated with taking care of these patients with their irreversible complications of chronic disease. And that’s fundamentally what I’m focused on. So if you think about it going now through the efficiency argument, after a stroke, cases management is great, efficient care is important. But I think it’s pretty clear that managing the stroke the day before it happens, so to speak, has got to be a lot more satisfying, certainly for the patient and from a cost perspective than managing it the day after it happens. And using that analogy, if we in fact can detect a signal that is predictive of a stroke, say it’s an arrhythmia like atrial fibrillation or severe hypertension, or even recognizing patients’ symptoms early on, if we can actually intervene at the right moment, and there are very potent interventions for everything from atrial fibrillation to hypertension to actual evolving stroke, then obviously that would benefit the patients and result in a lower cost than anything that we’re talking about with regard to the efficient management of the patient after the event.
SK: yeah. Well this idea is way beyond the state of the art today, I think we both agree. I mean what we have today is sort of, I think, a failed infrastructure for wellness. Would you agree with that statement?
LR: I don’t know if it’s failed, but it’s not optimal, to be sure.
LR: And then the other issue is, again, the real evidence of screening and the benefits of screening and issue around false positives and those sorts of issues. And then the reality, of course, that ultimately life style has a lot to do with the evolution of these diseases, and are we focused sufficiently on life style changes as opposed to screening and hoping to detect things early. So there’s a lot to be done with that. The concern I have is more related to this notion of clinical efficacy. And to the extent that, if you saw the smartest doc in the world for every issue that afflicted you, would you arguably have a better outcome than just sort of going through the system as people typically do? And my real goal is to sort of capture the cognitive skills of that smartest doctor in the world, and to basically create a system where she is basically sitting with you all the time, aggregating, assimilating, detecting signals, and then doing what needs to be done based on the best medical literature. And that’s the sort of environment that we’re trying to create. It’s complex, but I think for something like our health, it’s sort of well worth the trip.
SK: Would the world surveillance be a good word to describe –
LR: I think it is. I’m not sure I’d be politically comfortable, but the notion is if surveillance makes sense for your credit card, if surveillance makes sense for a jet engine, which doesn’t undergo routine maintenance; they’re looked at all the time, our refrigerators are looked at all the time if you have the right devices and programing, you could detect that you need a quart of milk. So to the extent that we’re surveilling all of these issues, taking advantage of technology, the Internet of things, it’s interesting to me and troubling that we haven’t taken advantage of that in healthcare, where we have very rich data in the forms of administrative information, information in electronic health records, and then certainly with the advent of wearables, implantables, home devices, a treasure trove of potential data that can be analyzed, contextualized and reacted to. And that’s what our focus is going forward.
SK: So it’s basically going through the data and getting more data. I remember when you and I were working on ActiveHealth together, we used to just say, Hey, the more data the better. Give me the data, for God’s sakes. I mean get me anything. We were in 1998, we were happy to get claims and labs –
SK: – and pharmacy data. And then well, we get a lot out of that. You want to give me some more data? You want to fill out a health risk assessment? Great, that just gives me more information. So and it’s very clear to me that with EHRs and so forth, we’re moving towards an environment of where their data is there and then the fundamental question is what investments do we need to make to build the analytics infrastructure to make use of that data. So give me a sense. The solution, it sounds to me, or the idea it sounds to me is surveil that data, whatever the best data that we can get, if we could be monitoring someone’s biological outputs, whether it be their heartbeat or blood pressure on a regular basis, that’s sort of the most awesome amount of data that we could get. But the point being let’s look at it, let’s find, I think, a subgroup where we think there’s a significant risk where a major disease progression is happening, and then begin to try to figure out how to find those people and intervene. It seems the purpose, Lonny, is to avoid the progression of chronic disease.
LR: That’s exactly right. And the complications that end up being so devastating for the patients and so costly for society. And I think what’s changed with the advent of wearables and implantables is the notion of complimenting static sources of data, relatively static sources of data like the EHR, which is filled in when you go to the doctor or the hospital, or claims, which are submitted when there’s activity. In the absence of that traditional sort of activity, what’s going on with the patient, day to day, activity to activity that might in fact generate information that would be valuable in pre-empting some of the complications that we’re talking about regarding chronic disease. That’s the idea.
SK: Very cool, very cool. So how do we do it? What’s the process for getting this done?
LR: So the process is to identify people who are at high risk. I wouldn’t want to do this for everybody. And once you’ve identified a high risk cohort of patients, then identify the types of physiologic signals that would be relevant to identify a transition point where something potentially bad and irreversible is about to happen. And then as a I mentioned, instead of designating that patient as high risk and putting them in a care management program, be prescriptive about what needs to be done. Could be thinking of a diagnosis or the institution or stopping of some therapeutic regimen. And I think importantly we need to involve both the provider and the patient much more than we traditionally have. So as we think about this ecosystem, there’s obviously a very, very prominent role for the care team, presumably led by the physician, who will not react to this sort of information with kind alert fatigue that we’ve seen historically. And I think the difference here is that we’re not sending an alert; we’re sending an alert with a context, with a solution. And I think the solution, the prescriptive part of this is what differentiates many other activities where, for example, you could have a pace maker or a defibrillator sending off all sorts of signals. Those kinds of devices in isolation haven’t yielded the results that we had all hoped for, in part because the physician’s left with the burden of having to deal with the signal, theoretically without the context of all the information that a system like ours might have, which can say, well, gee, this fallen blood pressure means XYZ in this patient. So in a word, it’s a requirement to identify high risk cohorts, using static data, track the high risk cohort using the wearables, implantables and home devices that I’ve been alluding to, and then this ongoing analytic capability to say, well, gee, this aberration, say in spending, going back to the credit card analogy, is suggestive of not only a bad potential outcome, but one that can be reversed with the right diagnostic or therapeutic intervention. And then beyond the provider, the patient has to play a much, much greater role. Adherence has been a huge problem, certainly for as long as I’ve been doing this, back to when I was actually a practicing physician. So then we get into the domain of patient education, patient incentives, plan designs that in fact motivate the sorts of engagement and behavior that we’re looking for. But here, the behavior that we’re seeking isn’t sort of general wellness behavior, which is important, but much more specific and sort of honed to the unique clinical needs of that patient at a moment in time, based on the sorts of signals that we’ve identified.
SK: Very good. As I’m listening to you describe this, I’m thinking in the back of my mind, So how do we take this and create a business from this idea?
SK: This business model. And it’s interesting to me; obviously computing is going to play a big role in this.
SK: Because you’ve got to do the population segmentation, and you’ve got to be able to create intelligence over the data so that physicians in the care team can be informed. But what are you proposing as a business model for going into the industry
LR: So the basic model is to go to a risk bearing entity and certainly at the outset, the preferred risk bearing entity would be a delivery system because of the intimate connections not only from an IT perspective, but with the clinical staff of that delivery system that has to exist. So with the advent of ACOs and other types of IDN risk, and the acknowledgement that a tiny number of patients generally drive the bulk of cost – we’ll call it a 20-80 rule. The point here is to go to the system and say, Look, you know that every year you’re going to have a bunch of people who have these terrible outcomes. They will in fact represent the 20% cohort that represent the 80% spend that you’re coping with as you manage your premium. What we’re suggesting is that the capabilities that you have in place now are inadequate for sort of pre-empting that evolution from that 20% to 80%. And what would be ideal would be the ability to in fact identify these patients early on, in fact, identify the 20% when they’re still costing 20 or 30%, and prevent them from ever getting to that 80% threshold. So in order to do that, we need the types of information that we talked about. We need the type of collaboration, cooperation with the delivery system and the care teams that I’ve alluded to, and we certainly need the engagement of the patient. Now beyond the delivery system is certainly a role and a desire to work with managed care plans, whether it’s Medicaid, managed care, or MA or commercial opportunities, same sort of theme. And then finally, employers, and this goes back to our beginning, Steve, with ActiveHealth have always been great sources of innovation and inspiration. And to the extent that we can identify cohorts of patients within these self-funded employer groups, and actually drive and channel volume to providers, who in fact are interested in sort of recruiting this sort of concept. We think the real opportunity is there. So at the end of the day, it’s really any risk bearing entity, and then with regard to our payment, to the extent that we’re just getting started, we’ll be starting with PMPM fees, which is obviously common in the industry. But as we mature and develop a balance sheet that’s reasonably and sufficiently robust, there’ll be a lot of interest in gain sharing and ultimately full risk bearing. For the cohort that we’ve identified at this 5 or 10 or 20% cost phase, inevitably going to 50, 60, 70%, and the risk that we would bear would be our expectation that through this pre-emptive approach we in fact can block that transition from where they are to where they’d be. And if you think about various types of risk adjusted premium models, particularly Medicare Advantage, you an imagine an arbitrage between the premium that you receive for anticipated risks, and an almost inevitable or inevitability to some of the bad outcomes from an actuarial perspective. So people with multiple comorbidities will have strokes or end stage renal disease. What we’re fundamentally trying to do is disrupt that actuarial model and say, no, it’s not inevitable. We’ll take the higher premium, but what we’d like to do is actually redirect the trajectory of these cases so they don’t get to those bad outcomes, and the arbitrage is between lower cost than anybody anticipated against a higher premium paid through the MA plans.
SK: So I have to say I think on the one hand, I think you’ve kind of done it again, if you will, from a standpoint of projecting out a very interesting idea around improving clinical care in the healthcare system, similar to what you did back at ActiveHealth. And if you remember back in those days, we had to spend a lot of time talking to people about the Active Health idea, which is basically today, I think, being copied by probably a couple of – at least 100 companies out there, where you were basically taking health data in, creating individual patient records, and then using an intelligent system to determine whether or not the care that was being delivered was appropriate or not. And in instances where it wasn’t, try to engage in an intervention with the physician and the patient. That was extremely innovative back in 1998, right? Because then it was just all UM all the time.
SK: Today, we’re doing some pretty cool stuff, right? So let’s not say that we’re – I think wellness is getting more personalized, and it will certainly get more personalized as computing begins to deliver individualized wellness programs. We’re doing bundled payments and ACOs, which are effective from a value based reimbursement point of view. But they’re all sort of very contained, I think, inside what you would describe as the efficiency part of this.
LR: That’s right. And I don’t know that those models can be successful unless you address the efficacy issues. So if somebody – pick a bundle, pick a straightforward orthopedic bundle around hip and knee replacement.
LR: Well, if in fact the patient is discharged and ends up developing a fever or deep vein thrombosis, or isn’t compliant with their anticoagulant medications, and they’re basically not mobile, those are all sorts of risks that will accrue most importantly to the patient, but also to the risk bearer. So if we can detect the person who’s about to have a deep vein thrombosis or a pulmonary embolism, or develop a fever and sepsis in advance, early on, by virtue of their failure to be adherent, their swelling of their leg and I mean we can kind of make things up, you have the ability to prevent the bad outcome that in fact will blow up the bundle, so to speak.
LR: So again, I think this is so fundamental. And I would argue that if the basic ingredients of precise diagnostics and therapeutics aren’t there and aren’t delivered in a timely fashion, I don’t know of any efforts that we’re trying to deploy on the efficiency side can make up for the cost then and the cataclysmic results –
LR: – that will result from a lack of real clinical excellence. And again, we get back to this notion bridging all that we’ve learned and all the research and all the publications that, to some extent, lie dormant in guidelines and publications in actually creating that explicit bridge to a patient at a moment in time, not only at the point of care, but at the point of generation of a data point that’s particularly important for that patient.
SK: So it’s evidence, it’s data and analysis, it’s risk stratification, right?
SK: It’s engagement of the provider and the patient. Those 5 elements –
SK: – make for this. And it feels to me like what you’re really – and I’m using this in the language of the day, because my guess is when we talk about this a couple years from now, we’ll be using a different set of terms. But it feels to me that when you do your risk stratification and you find these people that are potentially on their way, right –
SK: – that what you’re really proposing to say is, That is my sort of population bundle. I am going to attack that population with very active surveillance to try to prevent progression of disease.
LR: Absolutely. And as we’ve talked with CMS, I mean generically about bundles, the focus generally has been on procedures. Now CMS is transitioning to chronic disease. And not only will that chronic disease manifest itself along the physiologic lines that I’ve been talking about, but you can imagine also the advent of genetic information, which might be highly important in, again, understanding what intervention at what moment in time will be optimal for that patient. And again, without very, very sophisticated computational capabilities, how can a doctor possibly keep up with all of that stuff? Actually, as we work on what we’re doing now, and look for all the nuances around everything from blood thinners to treatments for heart failure and implantable devices and valves, it is so incredibly complex, and there’s so much nuance, and there’s so many variables. It’s hard to imagine that doctors are able to do as good a job as they do because it is complicated.
SK: It is. And honestly, in the same way, I feel that the work that you did at Active sort of ushered in an industry of data and analytics and interventions. You know, I feel that this idea, which is going to be, I think, unusual to some of the people listening to this Podcast for the first time – not unusual in that it doesn’t make sense, but unusual in that it’s being tried. But what you’re essentially saying is we’re going now down a path where we’re going to use those very same ideas, and we’re going to get into the disease prevention business. And that’s to me feels like it opens up a whole industry segment in and of itself because you’ve got to have the surveillance tools, the monitoring, the literature, the analytics, the interfaces to the providers, the interfaces to the patient. You’ve got to create business models that make sense. I mean one of the issues I know that you like to talk about is how hard it is to engage a patient, right?
LR: Absolutely, absolutely.
SK: I remember we used to say let’s pay them.
LR: And saying it was going to be a huge issue, yeah.
SK: Let’s pay them to be healthy.
LR: No, no, I still think that. In fact, when I was at Aetna, we sponsored a study where we gave people drugs for free after a heart attack, and we actually saved money by doing that because there were fewer adverse events like readmissions or re-heart attacks after the index event. And that was all great; the problem was that we never broke 50% in terms of compliance. So the increment in adherence that resulted in the favorable outcome findings were I think 43% to 49% compliance. So imagine, you know, post heart attack, you’ve had the event, this isn’t a theoretical wellness thing. You’ve sort of lived the experience of almost dying. We give you drugs for free that your doctor has told you will markedly reduce your change of having another heart attack, and for free we couldn’t break 50%. So that is behavioral economics is something that we need to continue to study. I read everything I can get my hands on, and it’s just enormously important. And one could even argue that all the things that I’ve been talking about simply won’t matter if the patient doesn’t adequately engage because, at the end of the day, they’ve got to do what’s best for them. And whether it’s subjecting themselves to a diagnostic test or taking a medicine, it almost doesn’t matter. At the end of the day, they have to do it, and if they don’t, then, you know, none of what I’m talking about is going to be particularly helpful.
LR: Yeah. I’m talking with Lonnie Riesman, the CEO of HealthReveal on his new business around preventing disease progression, which is extremely interesting, interesting to me, sort of the next step in our using computing and digital health to improve the cost and outcomes in the healthcare system. Give me a sense for the following idea: you’ve got a population of people out there that haven’t crossed over, but you know some of them will. For an ACO, for a self-insured employer, for a health plan, when you start to analyze that data, at that point in time, unlike a disease management program, right, where somebody is already chronically ill, that patient is not currently costing them anything, right?
LR: Maybe not much, but they also could have established disease. We’ll sort of take them where we find them as long as there’s an opportunity to help them, and frankly, unless there have been irreversible complications associated with their chronic disease already, which often is not the case, then there’s still opportunity. So even if you’ve had a heart attack, if you’ve survived, you’re still at risk of having another heart attack, arrhythmia, heart failure.
LR: And all of that is exacerbated by kidney failure. So it’s sort of a continuum.
SK: Is there studies that have been done that suggest that you can sort of calculate a reasonable time frame for a payback of an investment in that population? If you’re the ACO and you’re the employer or Medicare?
LR: I think so. And I think what you need to do is look at the clinical literature on a particular topic. So for example, Novartis just published some very exciting work on a drug called Entresto, which is a unique medicine for heart failure. And the study basically showed that over a period of 27 months, Entresto compared to the standard of care, enalapril, 10 milligrams twice a day, Entresto yielded results that were 20% better in those 27 months relative to cardiovascular mortality and admissions to the hospital for heart failure. So again, I’m not a shill for Novartis; I’m just telling you that it’s an example where the evidence is compelling, it was a well conducted, randomized trial that was published in the New England Journal of Medicine. The AHA and the ACC, I believe, are building guidelines around how to use Entresto. But the point relative to your question is that whether it’s stroke prevention or complications associated with diabetes, the literature will speak to the imminent opportunity in these high risk cohorts to make a very profound difference. So the distinction I would make is between a prevention program that might take many, many years in order to yield a desirable outcome, as opposed to what the literature would suggest about these patients at imminent risk, based on, say, multiple risk factors, and the benefit of pre-empting the evolution of that disease state along the lines of what we’ve been talking about.
SK: So it’s interesting. I’m going to get into a question for you here. If you’re – is there a point in time when device manufacturers, pharmaceutical companies, etc. can step into this business with you and begin to – because I mean obviously, we’re talking about an increase in utilization of some of these products that are out there, right?
SK: Step into the business with you and begin to sort of, quote, put their money where their mouth is, if you will, and say, We’re going to provide an up front financial cost that’s lower than perhaps just a straight utilization of our product in anticipation of sharing in the benefits and the cost savings. Are these guys thinking about those models or are they still basically just want to push volume?
LR: I think that they’re thinking about them, and hopefully we will be, I will be successful in convincing them that as we get into this sort of new domain of value based contracting, that just like the delivery systems have sort of gotten off the notion of price and volume, and are trying to deliver value, they ought to participate in ecosystems that again are focused on a particular, say, disease state. So we were talking about bundles before. So imagine that a bundle could be created where not only the delivery system is at risk, but all the components, all the ecosystems, the entire supply chain, if you will, that contribute to the care and the patients in that particular bundle. And to the extent that the device manufacturers and the pharmaceutical manufacturers can play such a magnificent role in improving the health of these populations and can demonstrate real value, perhaps they need to think more about shifting to a paradigm where they will be rewarded somewhat after the fact, based on the outcomes that they’re able to achieve, as opposed to getting their up front payment in a sort of traditional fee for service way.
SK: That would be incredible. I mean and you would imagine though, these are long – I wouldn’t say long term, but there’s certainly – it’s not something that feels like works if you just want a contract for a year. You’ve really got to take a look at these people for some extended period of time before you can say, OK, you’ve made this investment and this is the outcome of your investment. Your example was 24 months on the effectiveness of a particular therapy.
SK: What do you think the ideal timeframe is going to be to demonstrate the recovery of capital?
LR: I think it depends on the issue. So I’ll just give you another example. Atrial fibrillation and stroke. So depending on the patient’s risk for having a stroke, given that they have atrial fibrillation, with the desirability of detecting atrial fibrillation in populations where the prevalence is known to be high, but it hasn’t been detected, the literature speak to 1 to 2 year benefits for high risk people with regard to that sort of an outcome. So there are actually algorithms that are predictive models that will in fact suggest that person isn’t in that risk of, say, having a stroke, and if you believe the literature, and we can translate that literature in a practical, sort of risk bearing mechanism where the entire ecosystem can participate to identify and pre-empt the transition from, say, atrial fibrillation to atrial fibrillation and stroke, the timing is right, I think the economics compelling, and we just need to work together in order to achieve that goal.
SK: Excellent. Well, I have to say I really am looking forward to seeing you succeed again in this particular area. I think this is the key to the sustainability of our healthcare system is figuring out the right economic model and the right sort of technological and clinical approach tackling what really is the problem. Now of course, all of us over here on the VC side, we’re putting a lot of money behind this efficiency question.
SK: But I have a feeling that there’s going to be an effectiveness industry, if you will, and maybe we’ll come up with a cool name for it. I don’t know if effectiveness is cool. I don’t even know if people say it’s efficiency. But it is really. I always say it’s efficiency. Let’s get this thing automated, let’s take this waste out, –
SK: – and then we’ll get the next step is to make the clinical experience better.
SK: I have to say I could talk to you about this forever, but we’re running out of time.
SK: I would ask you one last question, and then let you go.
SK: And it is a little bit off topic, but an interesting question to me. I think a lot of us that don’t work in large health plans look up at the Chief Medical Officer position and try to understand exactly what goes on inside the health plan when you’re in that role. You and I have had a couple conversations about what you were doing at Aetna. But for the benefit of the listeners, just give them a sense for what the role of the Chief Medical Officer is, and what you think it takes to be great at that job.
LR: Well, thanks for asking that. I think the most important role for the Chief Medical Officer relates to medical policy and the sorts of decisions that many people are troubled by regarding defining certain interventions for patients as being experimental or investigational, and some of the controversy about what’s paid for and what’s not paid for. And all I can tell you is from my perspective, my experience, and I was CMO for 6 years, we always made decisions that were based on the best evidence. I always made a point of actually calling the treating physician to make sure that there was nothing that I was missing for some of these highly complex and highly expensive interventions, and that when I suggested or decided that something ought to be paid for, I never, ever, ever got any negative pushback suggesting that I was spending too much money. So I think the thing that I would want to emphasize is the overall integrity of certainly Aetna, and I think this is true of the other managed care organizations, the care that we would take to actually reach out to physicians to make sure that we have a full picture, and the degree to which, by and large, the treating physician on the other end of the phone would ultimately say, Well, you’re right, it is experimental; what else can I do? And the reality is we all wish we could do more, but the role of the health claim is to decide what ought to be covered based on the plan. And as the fiduciary, we had a responsibility to adhere to the best evidence. So I think the one thing I’d want to impart is the care and the integrity with which these decisions are made. And it’s difficult and it’s frustrating, but it’s really important.
SK: Yeah, it’s a tough job. I mean –
SK: – the whole medical infrastructure, we realize that there’s so much that we could try to do for patients –
SK: – but at the end of the day, we could spend ourselves into oblivion trying everything, so we’ve got to –
LR: That’s the point. So we have to rely on what’s known to be going back to the notion of efficacy and efficiency, you know, what’s actually got a chance of working in reality.
SK: Hey, Lonny, thank you very much for your time. I always love talking to you. I will let listeners know – in on one secret. I started in the business sitting on your board, actually.
SK: And you are one of the people, a small group of people, Hal Waxman being another one, that really gave me my education in the healthcare business, and I’m very grateful for that. And my sense is, Lonny, you probably gave some of the listeners an education about where things are headed in this conversation. My guess is I’m going to have a lot of questions. So maybe we’ll get a chance to talk again in the not-too-distant future.
LR: I hope we do. Can’t thank you enough.
SK: OK, thanks a lot.
LR: Take care. Bye-bye.