Reward good outcomes to reduce health care costs

The following editorial appeared in the Raleigh News & Observer on January 25, 2014

The contentious debate on the implementation of the Affordable Care Act has focused on health care choice, cost, access and, most recently, the too-low percentage of younger enrollees needed to cover insurance costs. The fundamental need to fix the expensive, inefficient way health care is delivered also needs attention.

Offering millions of Americans the opportunity to access the current “health system” is certainly better than their having no coverage, but it further strains a flawed approach to care. There are three clear solutions to our care delivery problems, and fixing them will make the expansion of health insurance easier to pay for and far more valuable:

• Care should be proactive and personalized with patients empowered to engage in their care.

• Care should be coordinated over time between individuals and their health care providers.

• Critically, reimbursement of care, which drives how care is delivered, must reward effective processes and outcomes rather than the number of procedures delivered.

Currently, health care is expensive and inefficient, in part, because treatment is focused on costly interventions – which are most amply rewarded by our fee-for-service reimbursement system – for often preventable chronic diseases. Care is largely uncoordinated and reactive to disease events and, even with health insurance, individuals are provided little support in navigating complex and uncoordinated services. The astounding increase in unnecessary emergency room use by new Medicaid enrollees in Oregon is but one example. Engaging patients in their care is essential to best outcomes, but it rarely occurs.

A Duke study provides a dramatic example of the perversity of our system. A program that combined preventative strategies, intense patient education and coordination of care for treatment of congestive heart failure greatly improved outcomes and decreased costs by almost 50 percent, but it proved unsustainable because of twisted reimbursement policies. Duke University Health System lost money as profitable hospital admissions were curtailed while unprofitable outpatient support services increased.

Fortunately, the need to change reimbursement to reward good outcomes has received national attention. Astoundingly, two-thirds of health care expenses go toward responding to the consequences of preventable chronic diseases, which have reached epidemic proportions. We can do better because chronic diseases develop over time, are predictable and are often preventable. Exciting new technologies show people what diseases they are susceptible to so they can receive individualized care to prevent or treat conditions most effectively.

As financial incentives help large care providers embrace coordinated care, there is reason to be optimistic. The State of Maryland has just announced bold changes in hospital reimbursement to reward good outcomes and to spur innovation. One major and influential institution, the Veterans Health Administration, has embraced a rational approach that will lead to personalized, proactive health care delivered in a coordinated and compassionate fashion because the VHA budget is aligned to reward best outcomes. Academic Medical Centers are recognizing the need to develop better models of care as evidenced by Duke’s new Institute for Health Innovation.

Improving the delivery of health care should be an overarching focus of the health care debate. Our nation is blessed with outstanding academic health care delivery systems that are capable of developing innovative models of care. Revamping reimbursement to reward good outcomes will spur creativity to develop more cost effective care, which will better enable the expansion of health care insurance.

Ralph Snyderman, M.D., is chancellor emeritus of Duke University and past president and CEO of Duke University Health System.

Industry Voices: Inside Genomics--Q&A with Bina CEO Narges Bani Asadi and Dr. Ralph Snyderman

By Narges Bani Asadi
Published on FierceBiotech (http://www.fiercebiotech.com) January 24, 2014


One of the elements lacking in the personalized medicine discussion today is the perspective of leading clinicians, informaticists and academics working in the field. To remedy the gap, I've asked a series of leaders in the industry to offer up their views.

With the annual Personalized Medicine World Conference (PMWC) right around the corner, it's a good time to reflect on where we stand in terms of reaching personalized medicine. How close are we, actually? What barriers do we have to overcome so that patients can receive the benefits of tailored therapy at their bedside? How much optimism should we have about the current state of personalized medicine?


I recently spoke with Dr. Ralph Snyderman, who will be moderating a panel at PMWC on personalized medicine, about these questions and more.

Dr. Snyderman is chancellor emeritus for health affairs at Duke University and James B. Duke professor of medicine at the Duke University School of Medicine. He has played a leading role in the conception and development of personalized healthcare, an evolving model of national healthcare delivery. He was among the first to envision and articulate the need to move the current focus of healthcare from the treatment of disease events to personalized, predictive, preventive and participatory care that is focused on the patient.

Here's our conversation:

What do you think is the single most prohibitive problem that's keeping us from getting to tailored therapy for the masses? What keeps you awake at night?

This is a very good time to be thinking about progress in personalized medicine. It's roughly the 10th anniversary of the onset of the actual development of initiatives and businesses that focus on the concept of personalized medicine. A lot of what spurred this movement was related to the completion of the sequencing of the human genome, but even before that, "omics" technologies gave rise to thoughts that these predictive capabilities could change the orientation of the practice of medicine--from being reactive and looking backward (i.e., identifying disease and treating it) to being proactive and able to predict, prevent and treat on a personalized basis. So, it's a good time to reflect on how much progress has been made and what the future holds.

My feeling is one of tremendous excitement and gratitude that personalized medicine is now a household name and a major health industry, but I think that the greatest progress is still in front of us. We've just barely walked through the door of what personalized medicine has to offer.

What hasn't happened, I believe, is a shift in the mindset and approach of the practice of medicine--that is, looking at the individual as a unique entity and planning for the most healthful future. For instance, risk assessments and personalized health plans, including therapeutic goals and tracking of outcomes, have not yet caught on in the practice of medicine.

So far, personalized medicine's impact has been almost entirely on more precisely defining and treating disease. This is good, but the major impact yet to come is in identifying risk, being able to prevent the onset of preventable disease, treating disease on an individualized mechanistic basis, and getting individuals engaged in their personalized health plan.

What gets you up in the morning? What are you particularly excited and optimistic about?

I am as optimistic now as I've ever been. I hoped that personalized medicine would catch on more quickly than it did, but I believe that it really is gaining traction.

There are more than a few things I'm looking forward to.

For example, I'm very pleased that over the last year the Veterans Health Administration (VHA) declared that every veteran will be able to receive proactive, personalized and compassionate care. I'm working with them now to develop personalized health planning throughout the VHA.

Also promising is the impact of changes in healthcare reimbursements to reward favorable outcomes rather than number of procedures. Reimbursement for outcomes will drive healthcare to be proactive, patient-centered and coordinated. All this will spur adoption of personalized healthcare.

We have seen tremendous advances in technologies--whole-genome sequencing, the concepts of aggregating large amounts of clinical information to measure outcomes, and the ability to analyze big data with innovative bioinformatics. We have the convergence of technologies that are able to deliver very useful, predictive clinical information, much more than we had 10 years ago. Meanwhile, healthcare providers are realizing that they need patients to be more involved in their care--more personalized care, more proactive care and more coordinated care.

We thus now have the convergence of the technology that's able to deliver personalized care and a market that's ready to adopt it. Having both of these pieces aligned is new and will accelerate the field.

I really do believe that we're not far from a tipping point where personalized healthcare is the general approach. If you interview me 5 years from now and you say, "Wow, when did all this happen? Personalized medicine has truly taken hold."

What are the most promising advancements in personalized medicine that you've seen in the past 6 months? What is keeping you excited about the future of healthcare?

The Veterans Health Administration has committed to proactive, personalized care. I think that's a really big deal in validating the concept of personalized healthcare.

The state of Maryland just changed their reimbursement system to reward good outcomes rather than number of procedures. I've been in contact with the leaders at the University of Maryland School of Medicine, and they are very committed to developing better models of healthcare.

I've been touting, for the last decade or so, that academic medical centers need to be involved in creating better models of healthcare, and I think this has really caught the attention--I wouldn't say that it's because of me, but because leaders of academic medicine are beginning to realize they are responsible for innovation in healthcare delivery. There have been a number of articles in JAMA and The New England Journal indicating that the status quo of backward-looking healthcare delivery is not going to get us where we need to go.

For example, Duke University, under the leadership of Dr. Victor Dzau, has just developed a major health innovation institute to develop better models of care.

All of these things that have occurred in the last 6 months make me think that there is a groundswell of realization on the part of the major providers and payers that we need to have a more proactive, personalized approach to healthcare delivery.

If you could create a multimillion-dollar center from scratch today, whether at a top university or at a CRO, how would you do it?

I would focus on developing working and learning clinical models for the delivery of personalized healthcare. This would include creation of programs to treat disease or to improve health and would apply the best available technologies capable of personalized prediction and align all personalized predictive and tracking tools as one force to evaluate the individual's health risks and therapeutic needs over a period of time.

It is important to realize that a patient's involvement in their care is absolutely essential for good outcomes in most cases. Getting people involved in their healthcare is an important challenge and would be studied to improve how this is done.

Finally, we'd focus on the acquisition of data around clinical decision points to aggregate clinical data from many individuals to analyze and identify the factors that were formative in outcomes (whether they be good or bad outcomes). We'd develop the databases needed to do this, along with the ability to collect data over time and the bioinformatics tools to develop validated predictive models.

I think with the assemblage of these components, our institution could become a learning model for proactive, personalized healthcare.

So what I would do (and what I've done at Duke) is this: Set up a healthcare delivery model so that it becomes not only a learning tool but also a functional delivery model that allows you to determine the best ways to improve the practice of personalized healthcare.

There would be an "end to end" capability to constantly introduce and employ the best predictive tools, the best analytic tools to collect and analyze information, and the best capabilities to get individuals engaged in their own care. In all, we'd have a learning working model for continuously improving personalized healthcare.

There is fantastic R&D going on at top universities, but how do we bring that research downstream to the clinician? What's the best way to bring personalized medicine to market, to the patient's bedside?

Some people have talked about "the valley of death"--that is, the gap you have to cross to move outstanding, exciting research discoveries that have clinical applications from the academic institutions into the clinical marketplace.

I think the most important way to do this is for academic institutions to acknowledge that part of their mission is to transmit their research discoveries into practical application.

Making the declaration that as an academic institution, we have an obligation to create better models of healthcare is an important start. And I think more and more institutions are doing that.

When I became chancellor at Duke quite a number of years ago, we declared our overarching mission to be the creation of new and better models of healthcare.

In terms of how you actually do it, I offer the following suggestions:

Create an effective office of technology transfer that interfaces as a welcoming entry point for industry to come in and truly understand what is going on throughout the academic institution and makes it easy to license or acquire technology.

Additionally, make it easy for faculty to interface with investors so that startups can occur. Have an open door--with all the appropriate controls--for venture capital to access the institution, so that they understand what's going on that might be of commercial value.

Create an environment where faculty appreciate the importance of commercializing their intellectual property within all appropriate academic guidelines.

While I was chancellor for health affairs at Duke, I strongly believed that academic medical centers should have an entrepreneurial mindset. Seeing a need for developing innovative multicenter clinical trials, Duke established the Duke Clinical Research Institute, which quickly became the largest academic clinical research organization in the world. It helped spur the appreciation of clinical research as a valued activity at academic institutions.

Another vehicle for enhancing technology transfer is accelerators. Institutions such as Duke are evaluating areas in which they have tremendous capability of improving healthcare delivery but not the ability to move them into commercial application.

Duke is bringing together venture capital investors, government investment capabilities, other potentially interested entities, including philanthropists, to create a funding source to help accelerate the creation of sufficient capability to develop maximum value.

These are examples, but I think the starting point needs to be the leadership of academic institutions declaring that moving academic capabilities into the public good is part of our responsibility. We have to balance this very closely with our academic mission and be mindful of conflict of interest and the protection of our patients. Regardless, improving healthcare delivery is so important, and I believe this should become an institutional priority.

Ten Years into Personalized Medicine: What We’ve Learned and What’s Next

Originally posted on Healthcommentary.org on January 17, 2014

Ten years ago, the sequencing of the entire human genome, along with the development of aggregate “omics” technologies began giving rise to a fundamentally new capability for the practice of medicine – the ability to predict and track disease risks on a personalized basis, to understand diseases mechanistically, and to target therapy to treat an individual’s specific disease.

Based on the potential impact of these technologies, I predicted in my 2002 Chairman’s address to the Association of American Medical Colleges, that medicine would be transformed from being disease-focused and reactive to being proactive and personalized. I believed that the “one-size-fits-all” approach to disease care was outmoded and would soon be replaced by one that would prevent and treat disease on a personalized basis.

Since its inception ten years ago, many of those predictions have been realized; personalized medicine has begun to have major impacts on components of medical practice and has engendered health care industries estimated to grow to $450 billion by 2015 with $42 billion related to drugs, devices, and diagnostics. Molecular diagnostics have gained traction in cardiology, rheumatology, transplantation, endocrinology, and, in particular, oncology. Targeted therapies have revolutionized cancer therapy and whole genome sequencing is providing insights into baseline health risks and understanding of some diseases. However, while the use of personalized medicine tools to treat disease is gaining traction, the transformation of medical practice to being proactive, strategic, and personalized; i.e., personalized health care, has been slow to develop.

There is, however, evidence that this movement is gaining momentum and with increases in health care consumerism, a more predictable regulatory environment, and changes in medical reimbursement to reward better outcomes, the adoption of personalized health care is inevitable. Personalized medicine will go beyond the use of technologies to individualize disease care to finally transforming the approach to care itself by enabling individuals and care providers to foster proactive, personalized care. In turn, the clinical adoption of proactive, personalized care will broaden the need for personalized medicine technologies thus spurring the growth of this industry.

Appreciating the value of a ten-year review of personalized medicine, the Personalized Medicine World Conference will host, and I will moderate, a panel discussion titled, “Ten Years into Personalized Medicine: What We’ve Learned & What’s Next” on January 27, 2014 in Mountain View, Calif., with luminaries including Kim Popovits of Genomic Health, Randy Scott from InVitae, Brook Byers of KPCB, and Jay Flatley from Illumina, opining on what they initially anticipated, what they’ve learned, and what’s coming next.

Personalized Health Care in 2013: A Status Report on the Impact of Genomics

The November/December 2013 issue of the NCMJ describes the impact that genomics has had on the practice of medicine in the decade since the full sequencing of the human genome was completed in 2003. Specifically, it reports on how genomics is affecting health care delivery, describes the concept of personalized health care, and discusses the role that genomics plays in such care. The commentaries and sidebars that follow highlight the opportunities and challenges of bringing genomics into clinical practice. Reading these articles will hopefully give clinicians and others a better understanding of the benefits and limitations of genomic technologies. Emerging capabilities, resulting in part from genomic research, are providing an opportunity to move health care from a reactive, disease-focused model to one that is personalized, predictive, proactive, precise, and patient-centered. Genomics and related technologies have already changed many approaches to care, particularly in the field of oncology, and I believe they will help to transform our overall approach to the delivery of health care. With the rapidly accumulating capabilities being developed and the focus on patient-centered and personalized care, I expect that the practice of medicine will become proactive and personalized within the next decade.