A Conversation With David B. Nash, MD, MBA: Game Changers for Population Health
A Conversation With David B. Nash, MD, MBA: Game Changers for Population Health
Bundled payments, big data, and advances in personalized medicine will change the way primary care physicians and others practice
David B. Nash, MD, MBA, envisions walking into his primary care practice and accessing an array of tools designed to help him provide the best, most cost-efficient care possible to engaged patient partners — and then getting paid a premium for achieving great outcomes. It’s a scenario that’s on the horizon, he says, thanks to payment reforms that reflect his “no outcomes, no income” mantra, the potential of use of big data to understand socioeconomic factors of health, and information technology that could enable the creation of comprehensive patient registries and the ability to better benchmark care at the physician level.
“I am definitely excited about where primary care is going,” says Nash, who is a board-certified internist, the founding dean of the Jefferson School of Population Health, and the Dr. Raymond C. and Doris N. Grandon Professor of Health Policy at Thomas Jefferson University. Also a government and private-sector consultant, Nash chairs the Technical Advisory Group of the Pennsylvania Health Care Cost Containment Council and is a member of the boards of directors of a variety of health care organizations, including Humana.
Nash has published more than 100 articles and edited 22 books. He is the editor-in-chief of four national journals: American Journal of Medical Quality, Population Health Management, American Health and Drug Benefits, and P&T, a sister publication of Managed Care. He earned his bachelor’s degree in economics from Vassar College, his medical degree from the University of Rochester School of Medicine and Dentistry, and his MBA in health administration at the Wharton School at the University of Pennsylvania. He spoke recently with John Marcille, editor of Managed Care, and Sonja Sherritze, editor of P&T. See the unabridged version on video below:
John Marcille: It’s been about three years since our last interview, when you mentioned that pieces of the Affordable Care Act that were not getting attention at that time, such as the CMS Innovation Center and the Patient-Centered Outcomes Research Institute, were really exciting pieces of the legislation. They are in place today. How do you feel about how they are operating and how they could be improved?
David B. Nash, MD, MBA: The Center for Medicare and Medicaid Innovation [CMMI] has made tremendous progress in the last three years. Dr. Rick Gilfillan was the inaugural director and is no longer at CMS, but CMMI is largely responsible for the nationwide effort at implementing bundled payments. We’ve got probably 30 million Americans in some type of bundled payment experiment across the whole country, largely for procedures — hips, knees, open-heart surgery, and so on. That’s a real important accomplishment. It’s too early to tell the progress of the Patient-Centered Outcomes Research Institute, or PCORI. Certainly, they got a lot of funding support. They’ve moved into beautiful new offices, they have all full-time staff now. They brought people from all over the country, from Kaiser, from Aetna, and elsewhere. They have a national advisory board. But it will be two to five years before we have any feedback as to the work that was accomplished under the grant award system.
Sonja Sherritze: A spokesman for PCORI has said that it was intended to be comparable to the National Institute for Health and Care Excellence [NICE], the patient outcomes institute in the U.K. Do you think that will eventually be a fair comparison? Obviously, it’s not right now.
Nash: NICE is a wonderful British idea for making sure that if we are going to put a drug or device on a national formulary it has good cost-benefit and cost-effectiveness ratios. Nothing like that exists in the United States. The FDA is basically safety and efficacy, and that’s where the conversation stops. In the current state of events, no way will PCORI become NICE. It might become NICE lite, but as we know from the Affordable Care Act passage, the Republican party insisted that PCORI’s work not be tied to any aspect of reimbursement. It’s a statutory prohibition against linking PCORI to any aspect of payment, most especially for pharmaceuticals.
Here’s where we’re headed: We will see the FDA asking more questions about whether we really need that 10th beta blocker and that 14th ACE inhibitor — good questions that they ought to ask — but it will be a long time before we are even ready for NICE lite. If there’s going to be a NICE lite, PCORI will probably fill part of that void.
Marcille: Do you see the possibility of a parallel institution growing up — perhaps not involving government — that might take on that function?
Nash: We can’t continue to put every drug willy-nilly on every formulary. That’s a big challenge both at the hospital level and at the managed care insurance company level. So we are going to need some better information about the economic consequences of these products. With regard to who will generate that information and who will be responsible for its implementation, at this moment, it’s a toss-up. The work of the Academy of Managed Care Pharmacy, the dossier system — standardizing the approach to how we make a formulary decision — that’s a step in the right direction. But I would not want to see a new federal agency of any type being involved further in the conversation about which drug gets approved, which drug is on which formulary.
Marcille: When we talked three years ago, you used a phrase that I thought was pretty great: no outcome, no income. How have we made or not made progress toward this goal?
Nash: “No outcome, no income” has been our mantra since March of 2010. Maybe we should have copyrighted it. We’ve got $750 billion, give or take, of waste annually in the system. Waste of resources. Waste of overutilization. Waste of inappropriate testing. Waste of using the wrong drug. We’re talking about hundreds of billions of dollars. It’s almost hard to get your arms around it. So if we could find a way to use these products more effectively and get paid, at least in part, based on the outcome that we’ve achieved, that’s what we mean by “no outcome, no income.” Another way to think about it is value-based payment or pay for performance. These are some of the synonyms that are in the marketplace. But it is all about: We would be happy to pay you — we would even be happy to pay you a premium — if you get a great outcome. But if you get a poor outcome, especially if you didn’t follow evidence-based practice, you are not going to get paid.
We have good evidence of that today, in terms of not paying people for readmissions within 30 days. That’s a failure to coordinate care. We’re not going to pay you certainly if you have a sentinel event. We’re not going to pay you if you have a central-line-associated bloodstream infection. We’re not going to pay you if your urinary catheter leads to an infection and prolonged hospitalization. And my view of that is, Great, you ought not to get paid. In no industry other than health care are you going to get paid if you don’t deliver value.
Marcille: And we’re making sufficient progress toward this end?
Nash: If you look at the evidence since 2010, especially nationally, it’s very clear that hospitals are improving their performance against all kinds of national benchmarks, in part because of the stimulus provided by no outcome, no income. We are definitely seeing a decrease in readmissions. We are definitely seeing an improvement in Core Measures. We are definitely seeing a reduction in hospital-acquired conditions. The state of Michigan is a great example. They have driven central-line-associated bloodstream infections down virtually to zero statewide in a major cooperative effort funded by the Agency for Healthcare Research and Quality. The big question is, Would we have done this without an economic stimulus?
Sherritze: You had also predicted a return to capitation. Do you see movement in that direction, and how does that fit in with what you just described?
Nash: Capitation is a dirty word, isn’t it? And it brings with it all kinds of emotional baggage, from Hollywood to Main Street. “Your doctor can’t do what he or she wants to do for you because of the big, bad insurance company.” All of that, of course, is mythology. Let’s look at the facts: There is no question that from 1990 to 2003, managed care, through capitation, bent the cost curve. That is irrefutable. People got tired of the gatekeeper model, doctors got tired of their perception of fighting with the managed care plans, so capitation went to the wayside. But really we are coming back toward a bundled payment and pay for performance. These are capitation in drag. Call it whatever you like, but here’s the issue: The research evidence is pretty convincing. If you pay providers to do a better job, you are going to get a better outcome, probably at a lower price, probably with fewer mistakes, probably using pharmaceutical agents in a more cost-effective manner.
A bundled payment is all about, Hey, we’ve got to work together — subspecialists, hospital, primary care doctor, nursing, pharmacy. All together we have to figure out how to allocate the pieces of that pie.
We’ve got good evidence that this really does work. Here’s a key difference that makes it more palatable today: Capitation was a gatekeeper-focused tool. It really didn’t engage the subspecialty community. A bundled payment is all about, Hey, we’ve got to work together — subspecialists, hospital, primary care doctor, nursing, pharmacy. All together we have to figure out how to allocate the pieces of that pie. That’s a core difference between 1998 capitation and 2014 bundled payment.
Marcille: Capitation has a reputation for being a perverse incentive not to provide care. You think that the new systems take care of that problem?
Nash: There is little to no research evidence that capitation caused a withholding of care. In fact, once again, if you look at outcomes in the ’90s, they were no better or worse than they are today. And by the way, we spent a whole lot less money. I’ve never been convinced that capitation led to any underuse. That would argue that we have a pretty clear pathway to bundled payment, value-based payment, whichever phrase you like. If you pay providers to do a better job, they are going to find a way to do a better job. And the hope is to achieve greater value for the money that we’re spending.
Marcille: You gave a talk, and we published a transcript in Managed Care, in which you envisioned a bright future of personalized medicine. But you are the founding dean of the School of Population Health. How do these two concepts work together?
Nash: My conversation about personalized medicine was in the context of the Lennox Black Symposium on the Thomas Jefferson University campus. We had some of the top people from around the world talking about the progress we’ve made — or lack of progress — in personalized medicine. From the work of Eric Topol and others, my sense is personalized medicine goes something like this: I have access to your genetic code, and I know from a buccal smear and an analysis of your genes that you’re susceptible or not to a whole series of diseases. I understand your mutations, and using that knowledge, I can personalize my therapy for you.
Personalized medicine might mean that I can tell a priori which anti-hypertensive is appropriate for you based on your genetic makeup so I can have more targeted therapy. I may not need to do screening for prostate cancer, as an example, if I know what your personal susceptibility is. For women, I might know your genetic risk for breast cancer, and I would not have to do mass mammography screening leading to all kinds of false positives, extra biopsies, and so on. So on my right hand, we have this miracle of personalized medicine, and in my left hand we have the current state of affairs of mass screening of populations. I’ve called this the paradox of population health. We have every incentive today, where appropriate, to do mammography, PAP smear, colonoscopy, screening for high blood pressure, appropriate prostate screening, and all the rest. That’s very expensive and not targeted. In the future, personalized medicine may obviate all of that. Whether that will exist in my primary care practice lifetime, I don’t know. I hope I’m around to see it; it’s incredibly exciting.
Sherritze: Dr. Topol says that this is going to result in a complete change in five years in the way that we do medicine. Can you discuss how that will happen and what the converging factors are?
Nash: Eric’s book, The Creative Destruction of Medicine, is a fun read. He is part of the quantifiable self movement. These are folks who, with new technology, are regularly measuring their heart rate, the number of hours they are in REM sleep, the number of calories they expend every day exercising, how many calories they consume every day in food. They are quantifying what they are all about on a physiologic basis. Eric’s thesis is that this group will be engaged in their own care in a way that most average patients are not. That’s going to lead to a creative destruction of the old model and greater patient engagement in their own care. He’s using this term from economics, creative destruction. That’s certainly one piece of it. Another aspect of it is personalized medicine. Eric’s contention is that if we keep moving the discoveries down the field and making progress in personalized medicine, we won’t need widespread screening. No more PAP smears for everybody, no more mammography, no more colonoscopy — only for targeted patients who are at genetic risk. Pretty cool.
Marcille: Can you give us a brief description of population health?
Nash: The literature says population health is based on three things. One is the recognition that there are outcomes of medical care that are unevenly distributed in the population — morbidity, mortality, and quality of life. The second tier of population health says those outcomes are based largely on the social determinants of health — where you live, your socioeconomic status, your level of education, and where there is poverty, crime, and pollution. And the third component of population health says that laws and policies like the Affordable Care Act are also in the mix. Here’s the punch line: 85 percent of a population’s well being, its quality of life, is due to factors other than medical care — 85 percent. What we do in health care accounts collectively for 15 percent of society’s well being. Population health is all about what goes on outside of the four walls of the hospital and four walls of the doctor office — the social determinants of health and the policies that drive it.
Sherritze: You have talked about Humana, Cigna, and other insurance companies harnessing big data and organizing it at the individual patient level. How do you see that playing out?
Nash: Big data certainly is getting a lot of attention. There was a recent Harvard Business Review cover story on big data. There’s a brand new journal out there on big data in health care. Pretty incredible. My understanding of big data goes something like this: If you really want to understand the health of a population — let’s go back to those social determinants and connect the dots — if obesity and smoking and socioeconomic status and income disparity and all of those social determinants are important, well, that’s a lot of data and you are going to have to find that in all different sources. Then we are going to connect all of that to actual clinical outcomes — utilization of pharmaceuticals, length of stay, appropriate testing — so we are going to bring together the clinical stream of data and all this community stream of data. When you put those together, it’s synergistic, it’s exponential. That to me is big data. So I can really only understand my population of patients with diabetes if I understand where they live, whether they smoke, what schools they have gone to, and what their tax returns look like. A subsidiary of big data would be what the exchanges are trying to do: check your tax status, your previous employment, all of that, and then determine if you are eligible.
Big data is big and going to get bigger.
The payers are very interested in harnessing big data because they want to answer the questions, How do I achieve value in the marketplace? Where will my investment yield the greatest return? Is it on pediatric asthma hospitalization in these five ZIP codes? Is it on flu vaccination in these 10 ZIP codes? Is it on open-heart surgery in these three ZIP codes? We are really just at the very beginning of this national movement to assess the impact of big data. Then, imagine if I am an Aetna, Cigna, Humana; I could drive patients, based on outcome, to certain providers. We are already doing that to some extent with centers of excellence, and so on. But I want to make sure that I am sending my Humana beneficiary to the Florida hospital that is going to give the best outcome for the money that I’m spending. And I may only be able to understand that outcome if I have big data to support it. All of the for-profit insurance companies will be very interested in the impact of big data.
Sherritze: So big data is big?
Nash: Big data is big and going to get bigger.
Marcille: When we had our last interview in 2010, you talked about the use of a registry to organize data and how useful this would be. There are registries in use all over the place for clinical conditions. What exactly is your vision of this larger registry? Is this everything about every person, in a three-dimensional grid that you can slice and dice this way and that way?
Nash: This is a very exciting area. In my primary care practice, for example, imagine if I could use registry-like software — and there’s a bunch of companies doing this already, Synaptic, Phytel, and others — I want to be able to say, “Hey, how am I doing in the care of my patients with congestive heart failure? Let me see their functional status, their readmission rate, their average weight loss under my diuretic therapy and their blood pressure.” There would be my population, with all of that data displayed. And then, a comparison population, holding severity of illness constant, so I could compare my performance on an outcomes basis with other doctors, other provider organizations. The registry function will enable me to practice population-based medicine. It will enable me — across all my patients in given clinical conditions — to understand how I’m doing relative to a benchmark group. That’s the vision I have from a registry function. The challenges are great. You’re not going to get a registry out of just a plain, vanilla electronic medical record. We need to graft on top of the EMR a registry software engine that is going to pull data from all kinds of sources, not just the inpatient arena. We’re going to need to link outpatient and inpatient at a minimum to give me a robust registry function. The good news is, these products are currently available on the marketplace. Lots more are coming.
Marcille: It is true, isn’t it, that physicians do respond positively when they are presented with actual data on their performance.
Nash: There is no question, based now on 25 years of research, from Brent James’s original experiment in Salt Lake City all the way through the Institute for Health Care Improvement, the work from our school, and hundreds of others: You give providers good information in a timely way from a reliable source that is non-punitive about how they can improve what they do, and then you give them the skills to improve. You’d better get out of their way, because they are going to stampede to improve. But let’s go through those steps. First you’ve got to close that feedback loop in a timely way; we rarely do that. You’ve got to give me good information based on my patients, because my patients are way sicker than yours. It’s got to be delivered by somebody I trust. It can’t be Uncle Sam just sending me an e-mail. It’s got to be face to face. I want to understand how I can do a better job. You’ve got to do all that. Then, you’ve got to give me tools: quality-improvement skills, a master’s degree in health care quality and safety, the ability to analyze my data. And you’ve got to protect my time to do that. Once you’ve given me the good information and you’ve given me the skill set to improve, and then you economically incent me to improve, I’m going to knock your socks off about how fast I can do a better job. That is the key.
Marcille: My general internist is a solo practitioner. He’s not going to be a solo practitioner in 10 years, is he?
Nash: There’s not much future left in our world for the onesie-twosie office struggling with an electronic medical record, trying to achieve meaningful use, under-resourced, not part of a registry, unable to benchmark their performance. I think, and most of the experts in this area would agree, that we are going to see the rapid demise of the onesie-twosie private practice primary care doctor. The whole reimbursement model is going to be around infrastructure and improvement, not individual doctors doing the same thing they’ve been accustomed to doing principally since the Second World War.
Marcille: Thank you.