Stat counter

View My Stats

Sunday, January 29, 2012

The changing goals of healthcare

The IOM has published a wealth of interesting papers. My attention this week was drawn this week to a paper entitled "Knowing What Works in Health Care: A Roadmap for the Nation" which can be found at It speaks to the need for a transparent and believable process which is used to assess what delivers value to patients and how to package that information where it is accessible and understandable to health care providers and to patients and their families. It is a complex problem which will require collation, synthesis, and interpretation of vast amounts of information.

It brought to mind how the landscape of the health care delivery world has changed. That might seem like I am stating the obvious but I believe that the changes which are most fundamental one the ones that are least appreciated. When Jesus was dining with the sinners and tax collectors, he was asked by the Pharisees why he would engage in such behavior. His response was"
“It is not the healthy who need a doctor, but the sick. I have not come to call the righteous, but sinners.” (Mark 2:17)
This is a far cry from the present world where doctors offices are filled with people who are the worried well. This underscores how our perspective on the purpose of health care has changed. While Jesus' remarks may have been recorded almost 2000 years ago, the role of physicians really did not change much until mid way through the last century. Physicians took care of people who were sick and generally acutely sick or injured. There was plenty to do and the outcomes were often not desirable ones. Well people did not seek out care from physicians because there was no reason to believe that physicians could deliver anything of value to a well person and perhaps even those with chronic and non-life threatening conditions. The question now is whether times have actually changed.

For those who needed the attention of a physician, for the most part the interventions were for acute conditions and the time frame required to assess success or failure was measured in hours to days. The goals of treatment were very simple. Did the patient live or die. No one was concerned about 30 day re-admission rates. Until relatively recently (meaning the last 50 years), few people were admitted to hospitals for anything (Jimmy Carter was the first President of the US actually born in a hospital). The revolution unleashed by drug development, especially the development of antibiotics (see "Demons Under the Microscope" for a great read on this completely changed not only how physicians practiced, but fundamentally changed what the goals of our jobs would end up being.

There are still many physicians who deal with life or death situations. However, increasingly our services are delivered to patients who are no longer sick in the same sense as those viewed as ill in previous generations.  Perhaps even more fundamental is the change in the time line for the assessment of success or failure of interventions. We no longer exclusively attempt to focus on acutely saving lives. Measurement of outcomes in acute care settings is still demanding and can have pitfalls, but it is inherently easier than measuring more difficult to define outcomes which may occur years in the future. In my opinion, the IOM paper was written because of the need to know what works in the realm outside of acute life or death scenarios. We need to be able to figure out which interventions are best for patients when we look out months, years, or decades. This is exceptionally hard and in certain circumstances perhaps not even possible no matter what technology we deploy.

Our confidence in the health care industry is to a substantial degree still living off the momentum created by the almost magical accomplishments in acute care setting that have occurred in the past century. It reminds me a an Arthur C. Clark observation.
Any sufficiently advanced technology is indistinguishable from magic. Arthur C. Clarke, "Profiles of The Future", 1961  
There is no question that we are still are amazingly good at dealing with acute care situations where we can deploy technology which has the semblance of magic. It is questionable whether accomplishments in this realm should translate into confidence in medicine's ability to intervene and make meanigful impact in a more distant time frame.

Sunday, January 22, 2012

Innovation and Medicare - Fundamentally Conflicted

The CBO just issued a report on Medicare Demonstration Projects: (

The idea behind these projects is to deliver better care for less money. It is a noble and essential goal since health care spending stands to undermine our entire economy within the next 10-20 years (and perhaps sooner). The results can be characterized as mixed at best with roughly equivalent numbers of pilots costing more money as opposed to less money. Granted the time frames had short time horizons and we have to take as a given that most innovative approaches will fail. 

My problem with the whole idea of Medicare Innovation projects is that the idea of top down driven innovation is really a non-starter and that meaningful innovation of Medicare the third party payment system it is embedded in are ones that will result in something entirely different. That will not happen if the innovations are embedded in the Medicare/Third party payment system we now use. They will co-opt any real change. 

I believe that Clayton Christensen's book the Innovators Prescription has an analysis of this problem which is spot on. He identifies various strategies which companies can use to address disruptive innovation in their industry. The disruptions almost always are related to some new product or service which comes in at the low end of the market and ultimately moves up market to disrupt the market leader. 

One example he uses was IBM, which was a leader in the mainframe computer industry. During the early 1980's, technological innovation allowed for the development of desktop computers which had a much larger market than the mainframe business. The first desktops were no match for the computing power of the mainframes.  Multiple mainframe and mini-computer companies saw the change coming but none except IBM actually adapted to their impact on the market. Companies like DEC and Wang went out of business but IBM did not. Why?

IBM leadership had the insight to understand that their attempts to grow a desktop computer business would require creation of a separate entity (IBM Entry Systems Division in Boca Raton, Florida - 1981). They understood that otherwise their innovations would be co-opted by the very business model (mainframe and mini-computer) which successful deployment of the desktop computer will displace.

From my perspective, successful innovation in health care payment and delivery means undermining third party payment systems, including Medicare. Given how Medicare is organized and administered, that will not happen. Markets allow for tremendous flexibility and do so by allowing participants to tweak from the bottom up. Medicare demonstration projects are a top down endeavor. The ideas may come from a variety of parties but they must percolate through the Medicare bureaucracy before they can be test or implemented. As it stands now, Medicare and other third party payers will always co-opt any attempts to change the system in any meaningful way and are now in positions to insure they are always in a position to co-opt any attempt using their control of the payment system.

The most constraining part of the Medicare and third party payment system is executed through the lack of the ability of health care providers and delivery systems to experiment with re-bundling of services. Medicare constrains providers through legal constraints. You either participate in Medicare with all its restrictions or you opt out. Similarly, other third party payers use the Medicare template to force providers to be either all in or not in at all.

I see the concierge movement is simply an effort to repackage and bundle services. In my estimation, it is exactly the right thing to do. It is viewed by critics as allowing the camel's  nose in the tent, the first step in disrupting a payment system which should be used to guarantee health care access to all. I also see it as a first step in disrupting a payment system which needs disruption. Once physicians and other providers of health care services are empowered to bundle services differently from what is allowed by Medicare and private insurance, innovation in those realms will explode. The public will benefit the most.

Finally, our perspective on the nature of desirable innovation needs to change. Hearkening back to Christensen's work again, the largest impact comes from innovations that enter at the low end of markets. Generally, these innovations have impact by delivering a product which is inferior to what was previously available but they do so at prices hugely less. While a mainframe computer may have costed millions of dollars or a mini-computer hundreds of thousands, the desktops cost $5-10K. They could do much less than their more expensive alternatives, but they could do infinitely more than nothing.

True innovation in health care which can save money to the degree that will be required to prevent financial calamity will require deployment of innovations which can cut costs, not in small and inconsistent  increments, but in huge chunks. The trade offs required will be modest compromise on the deliverables.  We put computers in a huge percentage of homes in 2012 not by promising million dollar mainframes but by initially deploying crappy desktop machines.

Furthermore, we succeed in deploying sophisticated technology by initially targeting the affluent few after which markets relentlessly drive costs down to make them available to the many at a fraction of the costs. High end whistles and bells in cars such as ABS braking systems and high end electronics were first available only in the top end luxury cars. They are now standard equipment at a fraction of the initial cost. Why has this not happened in health care? Non-market based, administratively controlled payment systems have served as a brake on this type of innovation. Medicare demonstration projects will not disrupt these systems and without that type of disruption, no meaningful change can occur.

I have a simple plan for innovation. Allow us to get a waiver from Medicare which allows us to continue our participation in Medicare while we experiment with alternative payment systems. We need no grant monies, just additional autonomy and the ability to bundle and price our services in novel ways.

Wednesday, January 18, 2012

He who sets the prices controls healthcare

Monday, January 16, 2012

Reflections on my old home town

Steve Malanaga wrote an Op-Ed piece in the WSJ which took me back to my childhood. I grew up in Buffalo, New York. Oddly enough my recollections were almost Shangri La like. It was a wonderful place to grow up, particularly if you had nothing to compare it to from the perspective of climate. I thought everyone played pick up basketball outdoors at -20F wearing a parka and mittens. There was also a certain appeal of women whose natural curves were augmented by down filler.

However years have past (many) and the shine has worn off. While I grew up there, Buffalo was clearly off its peak but there were many other less affluent and less thriving communities in the US. That has changed and if it were not for places like Detroit, Buffalo could take the prize as the most fallen from economic grace.

I had not ventured back to visit for many years until 2001 when I was invited to give a seminar at the University by a colleague who had moved to Buffalo to run a training program. I thought it would it would be great to wander around my old haunts and visit the few old family friends who still remained. It turned out to be an odd trip at many levels.

The visit was planned for around September 18, 2001. It almost did not happen because of 9/11. However, I took one of the first flights after airline travel was restored. On the way to the airport, I listened to a lecture from the Teaching Company. I am a TC freak. The lecture series was on the history of the United States and the specific lecture was focused on the US at the turn of the last century..1900. It centered on what was then arguably the richest city in the country. You guessed it... Buffalo, NY. The lecturer spoke of the confluence of transportation systems, the steel mills belching smoke, and the vistas of grain elevators.

I recall flying into Buffalo after listening to this lecture and our approach to the airport actually took us over the sites of those previous thriving industrial sites. How things had changed. What I saw were the rusted hulks of those majestic enterprises and the land was well into reclaiming them. The good news is Lake was much cleaner than I recall. No longer is the small boat harbour water stained orange from the slag from the Bethlehem Steel plant. The bad news is, Buffalo has joined the ranks of a number of more pristine but very poor places on this earth. Both the pollution and the jobs left.

Which brings me back to Steve Malanga's Op-Ed piece. What takes a place like Buffalo from richest to poorest in slightly more than 100 years? Is that a rapid transition or a gradual one? Is it surprising or predictable? Are transitions like this consistently preventable or inevitable? Will we be looking back at Silicon Valley in 100 years and have witnessed the same thing? Will Buffalo undergo a revival and become a destination location?

What makes some places rich and other places not so much so? The answer is wealth as as one of my favorite authors P.J. O'Rourke has written in his hilariously funny books "Eat the Rich" we tend to have little understanding of this process. Creating wealth and the entities that create wealth is not like baking a cake. Even under the best of circumstances we can and should expect failures. In addition, we should expect radical change. Entities which support the enduring generation of wealth are entities that are prepared to reinvent themselves, even if it means their reinvented selves look little like their own selves.

Places like Buffalo die because they tried to hold on to what they were and failed to empower those who might create a new and different Buffalo. Large infusion of political dollars do what they do best; preserve their political bases. They do not create wealth because that tends to upset the political status quo. What will bring back Buffalo? I don't know and I would venture to guess that no single person knows. However the decisions in the political realm which can help can best be described as permissive. At best they can allow them to happen. In the present state the default is to prevent them.

Sunday, January 15, 2012

You'll shoot your eye out!

A favorite ploy of health care economists who are market deniers is to channel through the spirit of Kenneth Arrow, whose sentinal work on health care economics in 1963 still shapes the debate today. In Uwe Rheinhardt's piece today in the NYTs, there is such an example:
In a recent interview with Conor Clarke in The Atlantic, Professor Arrow was asked how much of his 1963 paper “is still an accurate representation of the problems the health market faces.”
He responded:
I think the basic analysis hasn’t changed. There are wars over the details, but the basic analysis is accepted. Some specifics have changed. If you look closely at my argument there is a sociological structure. There is a kind of sociological thesis. The market won’t work – it doesn’t work well in the health context. But something else supplements the market, and the thing I put stress on in the paper are the elements that put a non-economic influence on the market: professional commitments to provide a service, to engage in services that aren’t self-serving. Standards of caring decided by non-economic actors. And one problem we have now is an erosion of professional standards. In a way there is more emphasis on markets and self-aggrandizement in the context of health care, and that has led to some of the problems we have today.
The emphasis is mine. Whenever market based solutions to health care economy problems arise, the response is predictable. A firm and confident declaration is made that the market does not work in health care. End of story. Debate is closed. Justification or empiric support is not required. A Nobel Laureate has spoken.   It reminds me of one of my favorite movies, A Christmas Story where the main character Ralphie desires a Red Ryder bee bee gun for Christmas. When he expresses his desire, he consistently gets the same response:

Ralphie: I want an official Red Ryder, carbine action, two-hundred shot range model air rifle!  Mrs. Parker: No, you'll shoot your eye out.
He gets his courage up to ask a departmental store Santa and he gets the same response. Why would you want'll shoot your eye out!!! No argument. No response. Ralphie's desires are trumped by a statement for which he has no response. He cannot ask whether this claim is supported by any sort of empiric evidence. He is not at liberty to ask whether the risk of eye injury is different in Red Ryder air gun users. It is the gospel truth and should not be questioned.

So tell me why markets won't work in health care? I don't want to be shown that market solutions will not be perfect. I take that as a given. I do not want to be shown anecdotes of imperfect outcomes. Bad outcomes will happen even with even good systems and should not be used as a basis to scrap market approaches. The center piece of Arrow's argument was the existence of information asymetries which he comments on further..

Such has changed since the early 1960s. In particular, the unimaginable advances in information technology have revolutionized many sectors of the economy. In health care, this electronic revolution has made it possible for patients to be much better informed about the efficacy of alternative medical treatments. That, by itself, should have reduced the problem of information asymmetry.

On the other hand, as medical science and practice advance rapidly, the information gap between physicians and their patients increases. Many transactions in the market for health care therefore still proceed on the basis of trust in the expertise and integrity of physicians and other health workers, rather than on the countervailing power of equally well-informed buyers and sellers, each looking out only for their own self-interest.
Is this really any different from other segments of the economy? I do not understand how to fix the brakes on my care and I suspect that a faulty brake job more likely puts me in harms way than most interactions with my physician. I have to trust that my mechanic installed the shoes correctly. I have to trust that the  mechanic who maintains the engines on the planes I fly is not cutting to may corners. I need to trust the cooks who prepare my food are not adulterating their fare. Our entire economy is based upon trust and skepticism. You need both. 

Again there are times where those who are ill at inherently at an insurmountable disadvantage but not always. In the same vein automobile repairs do not always have to happen within the context of being broken down in a desolate place with a tight time table. However, if your car breaks down in the boonies where you know no one you are more likely to be fleeced by someone who can and will take advantage of you. That does not mean we should eliminate the market to deal with automotive repairs.

The fundamental tenants of my belief in markets in health care are simple. First, it is basically incontrovertible that markets are the best tools yet developed to optimally allocate scarce resources. Second, it is impossible to make any consistent distinction between what is inside and outside of the health care economy at the margins.  Almost any good or service which enhances human life could arguably fall within the health care realm. Between health care cost inflation and ongoing redefinition of what entails health care, the health care economy will basically absorb the rest of the economy. If the health care economy is divorced from market allocation schemes, it means that the world of the future will be a throwback to the pre-market allocated world. Unless there is some other resource allocation mechanism which miraculously develops to replace market mechanisms in the near future, we will commit our descendants to wealth destroying race to the bottom.

It is not as if it needs to be all or nothing immediately. For each health care entity, give us the flexibility to move parts of our business to outside the third party system. As it stands now, you are either all in or all out. That is no way to structure a system that needs innovation. No one wants to make that big bet. Which portions of the health care business that can move to market and consumer driven models will be decided by many little bets. Many will lose. Some will win. A few will win big and winning big I mean the consumer will end up getting much more for much less. That is what markets do. Yes, someone may shoot their eye out. I am willing to take that chance.

Shedding light on misaligned incentives

Sometimes I read a story which appears to be unconnected to health care which sheds light on the health care payment quandary. Today is such a day, In today's WSJ Kate Linebaugh wrote a piece on the challenges facing municipalities which are trying to deploy light emitting diode (LED) streetlights. While LED streetlights are still more expensive than high pressure sodium lights, the cost of such streetlights has reportedly dropped by half in the past three years. Furthermore, the frequency of replacement is substantially less and municipalities are generally willing to absorb the up front costs in order to garner ongoing lower energy costs.

The economics of this share certain similarities to health care in that there are three parties involved.  Here there is the supplier of the service (electric utility), the direct payer (municipal government), and the actual recipient of the service (the public receiving street lighting). Here you have a technology which is disruptive in that it is less expensive in  terms of power needs. However, it often can only be deployed by the very entity which relies on selling power to remain viable (electrical utilities). 
From the perspective of many light-owning utilities, LED streetlights are too expensive and present a host of uncertainties—from light quality to how they would handle violent storms, extreme heat and cold and vandalism. LED lighting today "is nowhere near cost effective," Xcel Energy's Mr. Romero says.
Plus, utilities that have the capacity to power cities during the day have little incentive to try to reduce use at night, when power usage is low. "If you are an investor-owned utility and your profit is based on your revenue, what possible motivation do you have to conserve off-peak energy?" said Dan Howe, assistant city manager in Raleigh. 
Why should they adopt a technology which undercuts their ability to make money even if it benefits the other two parties involved. Utilities might encourage customers to cut power use during peak demand times but why adopt a new technology which save power during off hours? Utilities claim that the risks associated with adoption of this new technology rests heavily on them. They are not inclined to take such a risk when it is coupled to a less robust revenue stream. 

Similarly, medicine has triangles like this one. Provider organizations behave like regulated utilities. Payers are like municipalities.  Individual people are recipients of services. Less expensive interventions may be available but will not be widely adopted if they have lower profit margins. Why push to use a cheap chest x-ray when you have a higher margin CAT scanner available for use? In addition, there is risk associated with newer and less expensive approaches to care. More expensive is almost always associated with increased sophistication and sensitivity, whether additional patient benefit can be demonstrated or not. What if the CXR fails to reveal something the CAT might reveal. Liability in medicine is open ended. This is similar to the arguments put forth by the electrical utilities:
"When a utility makes a decision, for instance about lighting, it ends up becoming a permanent decision because once you put a light up in the air, you have to maintain that light for the rest of its life," said Rick Larsen, Progress's director of market and energy services. "There are a lot of risks." 
There are initiatives which change how these decisions are made. Municipalities who own their own light fixtures are heavily incentivized to adopt the LED technology. They rapidly see return on their investment in terms of lowered power and maintenance costs. They are all too willing to take on the costs and risks of installment when they also see the savings. In some sense we have seen changes in payment which change how health systems use resources. The DRG bundled payment system for hospital payment did succeed in moving to shorter hospital stays and drove for cost controls for hospitals to a point. When hospitals were placed at financial risk and could reap the financial benefits from limiting resource use, they became very effective at this while simultaneously demonstrating better outcomes.
Can we (or should we) coerce industries to adopt technologies which undermine their business model? The problem with this is no matter what you try to accomplish, if it places entities at financial risk, it will not work.
"It all comes down to money," said Gabriel Romero, a spokesman for Xcel in Colorado. "It doesn't save us any money. It saves them money. We pay for the installation when they receive all the cost savings."
If the financial incentives are not aligned, even the most obvious desirable outcomes will not happen. Furthermore, consistently aligning incentives with complex triangular financial architectures is essentially impossible. This is why simple and adaptable (read limited regulation) markets work so well. Parties who have goods or services to sell can offer them on their terms. If their terms are unattractive to those who might be around to purchase them, the sellers will either adapt or disappear. If the buyers are unreasonable in their demands, they will go without goods or services they might benefit from. In each case the impact of bad decisions tends to fall upon those who make the bad decisions. There are limited gaming opportunities. It is much easier to get the incentives aligned because there are simply fewer incentives to align.

Healthcare in the US (and perhaps around the world) will not get incentives in better alignment until the transaction architecture evolves into a simpler one. The ACO movement and its concept of gain sharing is a perfect example of this flawed approach. Hidden behind the copious verbiage in the documents defining its missions and structure is a simple focus... to save money. The gain sharing is a sharing of gains between providers and insurers. Whether it is at the patient's expense is arguable and will boil down to perception rather than fact. It will always appear to be at the patient's expense.

The problem goes away with a simpler architecture. While the triangular architecture of health care paid for by their parties will always be with us for realms of health care expense associated with the catastrophic and unpredictable, there will be a benefit from the conversion of the more mundane portion of the health care economy to move to the simpler market model. This is not the current state where the incentives are almost always to provide the most expensive (and high margin) service possible, whether it adds additional value to patients or not. Patients generally have limited awareness of the differences in cost, but because of the marketing efforts are attracted to the higher end services. Average people can obtain these services because third parties insulate from the cost. Average people can gain above average resources to pay for goods and services in only the health care economy. We do this by borrowing from the future and we can do this for only so long.

Ultimately, those who offer goods and services in the health care economy need to figure out how to provide them at prices attractive to average people with average resources. As opposed to rewarding those who find more and more expensive ways to deliver something, the incentives need to reward providers of services who can  do this at every decreasing costs and deliver more. This is how markets work. I can buy a flat screen TV now at a fraction of the cost three years ago and it will be bigger and sharper. That will never happen in health care if we push for a structure where insurers and providers are in cahoots to split the proceeds from stinting on patient care.