Editor’s Note: In terms of “bending the cost curve,” health-care providers in La Crosse, WI., have clearly demonstrated the ability to deliver high-qualty care for comparatively low costs. La Crosse was one of ten communities featured at a July 21 conference in Washington, D.C. titled “How Do They Do That?  Low-Cost, High-Quality Health Care in America.” The conference was organized by four nationally noted health care improvement experts: Don Berwick, Elliott Fisher, Atul Gawande, and Mark McClellan.

But that is only part of what has grabbed national headlines for this community that borders on the Mississippi River in Northwest Wisconsin.  La Crosse has become embroiled in a national controversy over end-of-life planning that has swirled around the health-care reform debate.  La Crosse’s largest provider of care—the Gundersen-Lutheran Health System—is a pioneer in ensuring that the care elderly patients receive in their final months complies with their wishes.  About 95% of the elderly citizens of La Crosse have signed advance directives on end-of-life care.

This year, Gundersen-Lutheran played a leading role in persuading House Democrats to propose that Medicare compensate  physicians for advising patients on end-of-life care. Republicans have charged that such a provision would lead to the rationing of care for elderly patients.  In an extended interview with John Iglehart and Chris Fleming of Health Affairs, Jeff Thompson, a neonatalogist who is chief executive officer of the Gundersen-Lutheran Health System, vigorously defends the value of advance directives, discusses why he believes competition among providers in La Crosse leads to wasteful spending and suggests the end game of health-care reform should make health-care systems accountable for the health of their communities, not just individual patients. 

JOHN IGLEHART:  La Crosse, WI., a city of 52,000 residents, was one of ten communities studied recently by four distinguished physicians who are interested in how health-care providers in cities large and small are able to deliver high quality care at low costs.  Among these ten communities, La Crosse had the best ranking based on a variety of measures linked to quality and costs.  In brief, how was La Crosse able to achieve this top ranking?

JEFF THOMPSON:  There are 3 system-level characteristics of Gundersen Lutheran Health System, our not-for-profit organization, that make a difference.  The first is a culture that focuses on the wellbeing of the community, not just the financial health of our system.  Secondly, Gundersen Lutheran is a broad, integrated system that provides a wide range of services from inpatient and outpatient hospital care to home care and many other services in between.  The third important characteristic is a strong tradition of physician leadership and engagement.  Physicians are CEOs, they are equal partners on our board of directors, and they are the medical vice presidents, and department chair who have paired administrative partners.  Doctors are integrated throughout our system.  The medical staff feels like it’s their organization and their community.
 Other important characteristics include a dedication to process change and learning from others, to hiring physicians, nurses, administrative staff and others who believe in our mission.  At the patient level, important characteristics include broad availability of the electronic medical record.  I know there’s some debate on exactly how much the electronic health record saves, but I can certainly tell you that there isn’t a doctor on our medical staff who would want to work any place that has less of a [complete electronic] medical record [system ]. 

We have a culture of trying to do the right thing for the wellbeing of the patients and the community.  This translates into a very high use of generic drugs, short lengths of inpatient stays as clinically appropriate, an investment in a care coordination program that cares for about 1,000 of our sickest patients and actually keeps these fee-for-service patients out of the hospital.  We spend several million dollars on care coordination that takes away several million from our revenue side but we see it as adhering to best practice.

Additionally, we receive a lot of credit for the advance directives that about 95% of elderly people sign voluntarily in consultation with their families and physician. Our staff works directly with a family to make certain that patients receive exactly the care they want favor as their health declines—either aggressive care in a hospital setting, or palliative home care and hospice if that is their wish.

Gundersen-Lutheran’s Pioneering Work In End-Of-Life Care

IGLEHART:  End-of-life care, as you know, has become embroiled in the health care debate, with Alaska’s former Republican Governor, Sarah Palin, asserting that Democrats would create “death panels” to determine who among the disabled and elderly should live or die.  What’s your take on these issues as they relate to health-care reform?

THOMPSON:  I think it is a great example of politics and the rhetoric around health care reform leaving the patient out of the picture.  The people who criticize end-of-life planning are trying to bring down someone else’s proposal, and they aren’t really talking to the patient.

You talk to our seniors here and they are very supportive of discussing their advance directives.  We don’t go to them in the ICU (intensive care unit) when they’re intubated and talk about this.  We discuss these questions when they’re feeling really quite well, and are surrounded by family members, and it isn’t some government official talking, they discuss these issues with folks from their church, or a doctor or trained nurse when they’re healthy.  They just want to plan ahead.

You’re not going to surprise old people that they’re going to die sometime.  They know this.  They talk about it with their friends all the time.  

IGLEHART:  How do families respond to the signing of advance directives by one of their elderly members?

THOMPSON:   Families really appreciate knowing exactly what their parent or loved one wants.  It adds great clarity to some otherwise very difficult times for families.  I feel badly that it turned into this firestorm around the country, because, in fact, seniors would receive the care they want much more often if this was broadly adopted across the country.

IGLEHART: One of the members of the La Crosse team, Kirk Stoa, the chief financial officer of a large grocery store chain, commented at the Washington gathering July 21 that La Crosse presented a “conundrum” for their business.  He explained that, among the communities where his company operates food stores, La Crosse has among the highest health-care costs despite the many awards its providers receive for providing good care.  How do you explain this seeming discrepancy? 

THOMPSON:  Despite our many efficiencies, the health insurance premiums in La Crosse are above national average for several reasons.  For one, the population in our region is older and poorer than most urban areas and, two, government programs significantly under-reimburse for services.  For example, Wisconsin’s Medicaid program has not increased its payments to providers in 13 years.  Its Our Medicaid payment rates rank about 47th lowest among the 50 states.  To make up the difference some of the costs of providing care to Medicaid patients — care I might add that is equal to privately insured patients — are shifted to private carriers.  Nationally, this is a well-known pattern that helps the solvency of hospitals that care for a substantial number of Medicaid patients.

In addition, as the Dartmouth Atlas data documents, because of the great variation in clinical practice patterns across the country, billions of Wisconsin taxpayer dollars are essentially shifted to other states because doctors there provide many more Medicare services to beneficiaries than is the case in Wisconsin. This pattern, too, leads to some cost-shifting to private insurers.

Competition’s Role In Duplicating Services And Increasing Costs 

IGLEHART: In your remarks at the Washington meeting, you said that competition between the two large systems that dominate health care in La Crosse is harmful to the community’s best interests.  Further, you said that the Gundersen Lutheran system generated annual revenues on the order of $800 million and because of duplicative capacity which you described as waste, you thought that – I think your estimate was $100 – $150 million dollars, could be saved if this duplicative capacity could somehow be taken out of the system.  What are your views on the role of competition in the La Crosse market? 

THOMPSON:  At the Washington meeting, the team from Richmond, VA., was asked how they improved their efficiency rating over the years of the study.  Richmond’s answer was, the community contracted from 14 hospitals to three systems and, in the process, they eliminated 500 inpatient beds.  

IGLEHART:  So, in your mind, how does competition fit into this picture?

THOMPSON:  I don’t know all the ins and outs of what drives different economies, but I can tell you that at some level of population, I don’t believe there is any evidence that competition in small markets improves the well-being of patients or the efficiency of providers.  We don’t benchmark our system against the other system in town.  Our doctors know better what the national benchmarks are and strive to meet them for the care of breast cancer patients, for our bariatric program, our cardiac care unit and other services.

I can tell you what the health-care model could have looked like in La Crosse.  In 1995, we tried to set up a single community health system with a strong community board that would benchmark quality and cost against the best providers in the state.  We couldn’t pull off that effort.  Consequently, the other system was sold to the Mayo Clinic, so we ended up with two large systems in town.  The other system does 35 to 40 percent of the health care business while Gundersen Lutheran delivers 60 to 65 percent of the care. 

As for efficiency, a couple of examples: At the time of the emergence of the two systems, we had three linear accelerators [a tool used in cancer care].  Since then, the other system has added two more.  The community needs maybe three, maybe two and a half of these expensive machines.  We have three cardiac cath labs and the other system has added another one.  The community probably needs two and a half.

We’ve always had two neonatal intensive care units, which is my specialty, and between the two systems we have two tiny units that amount to a wasteful duplication of equipment, duplicate staff, duplicate transport teams, and duplicate people on call. 

The additional equipment and staff were added to provide those services that are most profitable. At the same time, their inpatient behavioral health division, inpatient psychiatry, has gone from 28 beds, I believe, down to 12.  And how many do we need?  Well, we need a whole lot more than that.  But it’s harder to make money in behavioral health.

There may be some markets large enough where having these competitive forces at work could drive down prices and improve efficiency, but La Crosse is not one of them. 

IGLEHART:  How would you address these capacity questions if you were a policy-maker?  At the gathering in Washington, D.C., several of the communities spoke of the impact Certificate of Need (CON) laws have in their states, saying they had proved effective in constraining capacity.  Do you support CON or other forms of greater public oversight of health care? 

THOMPSON:  I don’t support CON laws as a way to limit capacity. The data across the country would argue that these laws often add costs eventually.  I support community levels of control where citizens recognize community interests, not just the interests of health-care systems.  We’ve had great success with a board composed of physicians and other community leaders who have diverse backgrounds and work experience.  The focus of their concerns is on the mission of the organization, which includes the wellbeing of the community.

I think to truly bend the cost curve long-term, it’s going to take two really fundamental pieces: 1)We have to change how we deliver health care and replace the fee-for-service model with a more suitable set of incentives; and   2) We need to place greater focus on improving the health of populations, not just providing health care after people become ill.  There are models achieving progress on these fronts—decreasing unnecessary treatments and keeping people healthier.

Gundersen Lutheran received a lot of press through the Washington Post about how efficient we were. Other cities responded that everybody in La Crosse is middle class and white: “Gee, if I didn’t have any poor people to take care of, my outcomes would be good, too.”

The truth is that, compared to Philadelphia and Los Angeles — as well as Boston, New York, and Miami — our patients, our senior citizens are older, and poorer.  They smoke more, they drink more, and they have worse eating habits than their counterparts in any of those places.
IGLEHART:  You pointed out is that physicians are integrated into every aspect of your system, from serving on the board to a whole variety of other roles. Are your physicians for the most part employees of the hospital, or the system, or are they simply private physicians who have practicing privileges at your hospital?

THOMPSON:  Some doctors are just on the staff; some doctors are employed but not truly engaged in the system; and then there are the doctors who are engaged in the system — that is, the physicians who are on the board, active members of the major committees, part of the leadership teams of both operation and governance, and most importantly, believe that it’s part of their job to help make the system better.  I take umbrage at people making fun of the kids coming out of training, saying that they aren’t willing to work much.   We have many young people in the first five or ten years out of their training who are doing outstanding, courageous, hard, thoughtful things: medical staff peer review; running major league projects to reorganize the flow of patients in the hospital; taking on the medical home and disease management issues; and dealing with very difficult issues in the outpatient areas.

That’s a level of engagement that’s beyond just being an employee.  I would argue that that gives us a huge advantage.  You develop an environment where the medical staff look at the administrative group as partners and where we all have the same mission going forward, to take care of the patients

IGLEHART:  So in your mind, Jeff, you don’t differentiate between a physician who is an employee of your system versus a physician who has privileges in your hospital but is not an employee?  You see them both as being by and large strongly committed to the system?

THOMPSON:  Right, and to be completely honest here, John, 98 percent of all the work done in our inpatient and outpatient areas is done by physicians, who are Gundersen employees.

Moving Toward New Payment Models

IGLEHART:  In that context, I’m sure you have heard countless times from Washington policymakers and groups like the Medicare Payment Advisory Commission of their support for moving away from fee-for-service payments, and developing bundled payments for episodes of care.  Some of these proposals leave to individual hospital systems like yours the tough job of carving up the bundled payments between physicians, hospitals, and other players. What’s your view of moving toward a system of bundled payments or at least toward hybrids that might be part fee-for-service, part capitation, etc.?

THOMPSON:  I believe it’s exactly the right way to go.  I think paying for value and outcomes over a set amount of time makes a lot more sense than what we have today.  Maybe you couldn’t have done this 20 years ago because we didn’t have the data, but we really have the ability to do it now and we can measure true outcomes: Not just how many days a person is in the hospital for a hip replacement, but starting far enough back to make sure that the appropriate amount of imaging is done, and giving the patient a rehab score, and seeing how well they’re really functioning.

I will argue that something like that will do more to force improvement and coordination in health care organizations. Whether you’re a hospital, an individual physician, a clinic, or a system, episodes of care will force people to look at the long-term outcome of each patient and to ask what we have to expend to get the best outcomes.  It will incentivize a search for best practices. And when you put everybody in the same pool like we have, it makes for better partnerships.

I speak to many groups.  Recentlya group from Pinehurst, North Carolina, did a site visit at Gundersen Lutheran.  They sent 24 people up here because of their concern about the very thing that you just mentioned:  they believe payments are going that way.  All their physicians are independent practices, and they’re trying to figure out how to move to a model that can accommodate a bundled payment structure.

We had them meet not only a lot of our doctors, but our administrative people, too, because you need both the doctors and administrative staff who say yes, this is the model, we’ll work as partners. One of the reasons we’re successful now is because we’ve been able to attract a very strong group of administrative and medical leaders.

So we believe paying for value is the way that Gundersen will do well, but also the way that the system will be able to deliver better outcomes for patients in the long term.  To really go out on a limb, you could imagine a model that would push that even farther and pay to keep the community healthy as opposed to paying to treat sick people. You can migrate out from paying for a short amount of time around a hip or a coronary artery bypass graft (CABG) procedure to paying for caring for a diabetic over a length of time. From there you could keep moving out to whole populations of people that you’re responsible for keeping as healthy as possible over time.
IGLEHART:  The prospect of comprehensive health reform, whatever that might mean, seems to grow increasingly less likely as the weeks and months go on.  Where do you stand on some of the major issues, such as moving to universal coverage and creating a public option?

THOMPSON:  Well, first of all, despite a lot of people’s disappointment that things haven’t moved along faster and we don’t have a big reform bill, the good news is that we’ve had more discussion about a very serious and important issue than we’ve ever had before.  We’ve had lots of different ideas put on the table, and, of course, there’s been some flaming out on each side.  But we’ve had a lot of interest, a lot of hard work on it.  More people know more about what’s going on in health care than ever before across the country, and certainly in Washington.

So the fact that the whole level of understanding has been raised is very positive.  And you get a lot of hand-wringing that everything hasn’t been fixed yet.  But we’ve made progress, even if a bill hasn’t been passed yet.
 I think another thing that’s very positive is the insurance reform pieces – that is, can we get to coverage for everyone, and can we get insurance affordability?  Can we get larger across-state pools for small businesses to help them out? Some of those things have been talked about enough and almost everybody is agreeing to them, so that gives you hope that in at least a first-line bill that gets passed, those things will get addressed. That’s important, even if it isn’t the whole big change.

Now, the problem with the big reform is that it gets, of course, very complex.  It has to do with a lot of money, and Congress gets pretty nervous about big potential money shifts.  We’ve been as clear as possible in saying that we have to go for the long-term improvement of health care, and we have to face down this huge potential cost coming forward.  One of the most important ways to do that is payment reform, as we’ve discussed. And there’s a variety of ways to go about doing that.

Regarding the House bill, just saying we’re going to have a public option that pays Medicare rates isn’t  fundamental payment reform.  That’s just fee for service, and that model is part of the reason why the costs are rising so rapidly. 

Instead of setting up a bunch of new agencies, I think you have to ask, “What are the models that have really worked?” Gundersen is not the only model that’s working, but there’s a fair amount of evidence that our model is one that is working.  So why not develop incentives for other people to at least move towards that for now? Then we’ll improve on that and make it better?
CHRIS FLEMING:  Earlier you mentioned a number of factors that you thought enabled La Crosse to produce higher quality care at relatively lower costs, and some of them involved the culture of La Crosse that has developed over a number of years. Just now, you were talking about the importance of changing payment incentives.  What is your view is in terms of the relative importance of those two factors, of top-down versus bottom-up pushes in terms of getting to better care and less expensive care?  Do you think that changing the payment incentives will be enough to enable other communities to do the sorts of things that you have, or do you think it’s going to take them a number of years even if you do change the payment incentives to develop the same kind of culture that you now have in your community?

THOMPSON:  To get there at any pace, you’re going to have to change the payment system.  That will get people thinking about what they are all about, and how they get better integrated, and improve the wellbeing of the community.  This would help change the mindset of the doctors and nurses, administrators, all the important pieces of the system.  It’s not impossible to develop the mindset that exists in La Crosse in another region; there are wonderfully trained doctors and nurses, pharmacists, and administrative people in every part of the country. If you start changing the payment system, then you’ll get more people focused on the right kind of goals.

The Different Brands Of Physician Advocacy

IGLEHART:  One thing that has always struck me about physicians, and I suppose I’m particularly sensitive to it having written for a medical journal, the New England Journal of Medicine, for many years, is that physicians are generally leading citizens in the communities where they practice.  They are respected, they bring authority and credibility to their tasks as doctors. Yet when physicians have gathered in various collections, whether it’s the American Medical Association or countless other medical organizations, when they gather in Washington and try to harness their authority and respect, it just breaks down, and it mostly breaks down because the issues that they bring to Washington are largely economic and pertain to their incomes.  This has always struck me as odd and perhaps presents a challenge for organized medicine to figure out a different model of advocacy that would not only serve their own interests but those of the larger community and society. Do you also see a disconnect here?

THOMPSON:  That’s a great observation, John. Here’s my answer back to you in a question.  Why is it that 85 or 90 percent of pediatricians belong to the American Academy of Pediatrics, but probably less than 15 percent of practicing physicians belong to the American Medical Association?

IGLEHART:  Interesting question.  I have said in the past — I’m not sure it’s as true today as it was traditionally — but the pediatricians in Washington have been held, I think, in greater respect by congressional staff, members of Congress, etc., because they largely advocate on behalf of their patients: children.  They’re not oblivious to economic questions and their own economic interests, but that hasn’t been foremost.  I think the pediatricians have said, “If we advocate on behalf of those that we serve, the income things will come along.”

Now, the income end of things hasn’t come along as well as they might have liked, but I think that’s why the Academy is held in high regard, generally speaking.  However, I think the Academy has come under pressure in recent years from its members, whose incomes obviously are at the low end of the pay scale in terms of medicine.  But historically and traditionally that’s how the Academy has been viewed in Washington.

THOMPSON:  We agree 100 percent on this, and I used that model because I am a pediatrician and I have always been a member of the American Academy of Pediatrics for the very reason that you state, because I knew where their priorities were. The AMA by contrast has been so embarrassing at so many times, it’s why at my age and behind me, the percentage of people that have engaged and paid dues has been tiny.  Time after time the AMA has screamed about their finances and so they have lost their credibility in Washington and with the public.

More than once somebody has gotten in my face and said, “Well, then, why don’t you get in the middle of the AMA and fix it?” Maybe one of these days I will.  But right now I’m a little busy trying to work on this other stuff.

On health reform, I can tell you that we’ve had some success recently.  In almost no time flat, the people from Gundersen, Mayo, Marshfield, Inter-mountain, Everett Clinic, and Cleveland were able to put together a group of folks that responded to a number of issues in the bill. We worked with staffers from the major committees and the major House and Senate offices as a voice for some of the most successful systems in the country.

For example, the proposed House bill was going to pay everybody in the Upper Midwest inadequate Medicare rates, which would pretty much destroy all of us in the Upper Midwest.  We put together a group very quickly and we were, I believe, received very well.  We objected to the House proposal and presented this alternative: You can hold us accountable for outcomes.  We’ll compete with anybody on outcomes.  That’s where we need to go for the future. I know it’s a big pill to swallow for the parts of the country that haven’t invested in electronic records, and haven’t invested in other things.  But that’s where the country has to go if we’re going to stay competitive in health care.

FLEMING:  When you referred to the House bill earlier you said that just having a public plan would not work the kind of fundamental reforms in payment and in health care delivery that you felt were needed.  You mentioned that some Congressional staffers have been listening to your concerns. I know the Senate Finance Committee, for instance, has been talking about paying more attention to reform steps that would really affect the cost curve, and trying to do more than just add more agencies and more people covered, however admirable adding coverage would be. Do you feel that people are starting to pay more attention to the kind of fundamental reforms that you think are needed, or do you still feel like fundamental delivery system reform and fundamental payment reform is not getting enough attention in the reform debate?

THOMPSON:  Well, I’m not sure those are the two opposites.  I think fundamental reform is getting much more attention compared to a year ago, two years ago.  People in Congress and elsewhere are being courageous who never were before, they’re standing up and saying this is where we need to go…for example, saying that the public has some skin in the game as well, that they have to work on becoming more healthy.

These are very important things that need to be said, that need to be worked on down the road.  So I’m actually pretty encouraged.  To imagine that you could make that big of a change in a matter of a few months in something that is such a large part of our lives is asking an awful lot, and so it doesn’t surprise me that it’s going to take a little more time.

But I am encouraged. We have been to Washington, we have talked with the staffers, we’ve talked with the leading members in Congress, and they are very sincere, and they are taking on very hard issues. The big questions are being given very careful consideration, much more than a year or three years ago.

IGLEHART:  Jeff, I have one last question.  Since the Mayo system has entered your marketplace, what impact has that had?   The talk about the Mayo model in Washington puts it on the top of the mountain as a system, but I’d be curious to know what impact it has had on your system.

THOMPSON:  Well, I can tell you that Mayo has a huge considerable reputation, and they have done some wonderful things.  I have sent patients there and they have received great care.  All those things are true.

My concern focuses on the people who are looking around the country for a model.  People keep talking about the Mayo model, and immediately the model is dismissed because the other 90 percent of the country says, “Let me see, do I have a billion-plus dollars in an endowment?  Do people fly over here from Saudi Arabia to get care?  Do I have a national brand?”

So how do you give hope to the rest of the country?  You say okay, let’s take a smaller system, like Gundersen.  Let’s take a place that only has one tertiary hospital and doesn’t have this international branding. How can they compete?  Well, it turns out that we get just as many five star rankings on in-patient measures from HealthGrades as Mayo does, that we’ve been able to put up numbers on chronic disease that are better than Mayo, that we’ve been able to attract stunning physicians, administrators, pharmacists, and nurses into our system to compete directly with Mayo.

Do we work with Mayo on some things?  Sure, I already told you how we worked on health care reform with them.  We have our combined Gundersen and Mayo physician assistant program.  This year we’ll have the highest certifying exam scores of any PA program in the country.  We’ve worked on education, we’ve worked on some research together.  I’ve consulted with Denny Cortese on health reform.

Do I believe Mayo’s coming into La Crosse changed the trajectory of our work on quality or efficiency?  Not at all. In fact, as I mentioned, I think it slowed down the overall efficiency of the community because it created some unneeded duplication of equipment and services. We’re trying to improve quality constantly, and for the things that we do better than Mayo, we don’t use them as the benchmark and say, “Oh, we’re as good as Mayo on our diabetes measures, now we can quit.”  We say “No, the guys at ThedaCare are a little better than we are, or there’s some guy in Philadelphia whose outcomes on bariatric surgery are a little better than ours.  How can we get better?”

I think we’re a more plausible model for the rest of the country than Mayo is, and that could be why the group from Pinehurst, North Carolina, came up and visited us.  They can see their way forward to try to develop into a system like ours, in developing care, patient by patient.  But it’s pretty hard to imagine that they can come up with a five billion dollar endowment in the next five years.

If you are going to reform health care for the rest of the country, I would argue that the Mayo model doesn’t get you much traction.  The Gundersen model frankly isn’t a whole lot different than the Mayo model — we’ve been doing many of the same things for decades. But I would argue that the Gundersen model has the potential to become actionable in 85 percent of the places in this country.

IGLEHART:  Thanks so much.