Those of you paying attention for the past few days might have noticed on the one hand a sense of optimism and unity as Barrack H. Obama, somewhat somberly, began his presidency.
Meanwhile, over the past few weeks the fur has been flying among the electrons on THCB while some very knowledgeable and opinionated health care wonks and geeks have been battling it out about what exactly we should be doing in terms of federal health care IT spending.
Given that even among you smart THCB readers this may be all a little perplexing, I’m going to try to try to make what I hope are some elucidating comments to put this argument in context. I’m doing this partly because I’m perplexed too, but also because I think that there is some hope for a middle road.
First the basics: As sometime THCB contributor & uber-CIO John Halamka makes clear in this excellent post about The Greatest Healthcare IT Generation, some $20 billion of the soon to be passed “spend it as fast as you can” stimulus package is going to be targeted towards health care IT. Now, that’s by no means the biggest part of the $800 billion or so package, and it’s not even the biggest part of the health care spending in the bill. Nearly $87 billion or so is going to support Medicaid, although that will mostly will be replacing cuts being forced on states.
Let’s be clear, the stimulus package’s main role is to stop the patient from bleeding out and will probably need to be joined by a bank restructuring the likes of which we’ve never seen. Health care is a sideshow, but $20 billion is still $20 billion, and given that the current health IT market is only between $20 and $30 billion annually, it’s a huge potential increase for the industry.
This spending is separate from any larger health care reform proposed by Obama, even though it appears that such a reform package is fairly likely to appear in Congress as soon as Daschle, Baucus and Kennedy get their ducks in a row. And of course, most of the issues that divide THCB’s various commentators on health IT are rooted in problems that only a major reform can solve. And realistically even those most optimistic about the prospects for reform don’t believe that the Obama/Daschle/Baucus plan will get at the core problems of the U.S. health care system any time soon.
The underlying problems: American health care has two linked and intertwined problems. First, (and it is first) due to America’s unique political history, most of the pain of dealing with increasing health costs is deposited on a poor (or soon to be poor) and relatively powerless minority of the population—the sick who are uninsured and underinsured. No one is really responsible for their financial well being, nor is any entity forced to make their cost of care equal to that of the rest of society. So the health care system has reacted rationally over the years by increasing what it charges the majority, and not worrying too much about that minority even as it grows. Every other rational country instead distributes that cost evenly by putting (more or less) everyone in the same financial pool, and puts someone (usually the government) in charge of the total cost of the system to society. The US is different, as it’s the only place where for the players in the system doing more means getting more absolutely. Everywhere else, the rest of society stops the health care system arbitrarily grabbing more resources.
The process whereby the health care system grabs more and more in the US comes mainly via an incredibly strong supplier community that has political control over the large share of spending that is government funded, and economic control over the main providers of private spending—employers. Of course it may not feel that way to the suppliers in the system but the numbers (share of GDP going to health care more than doubling over 30 years) don’t lie.
This leads to the second problem: The mechanism of that provider control is a payment system that encourages piece work, acute care, fixing rather than prevention, specialization over primary care, and—not least—big hospitals over community based clinics. Now as Atul Gawande points out in the New Yorker this week, it’s not as if every other country does exactly the opposite, but the scale by which we’ve tipped over here is unprecedented. And multitudes of articles on THCB over the years prove it out, and show that it’s very hard to change that status quo.
This starts to get us back to IT. The first significant use of IT in health care in the US was in large hospitals primarily aimed at accounting (and billing) for piece work. Eventually most large and many smaller hospitals began to extend their IT capabilities to automate other aspects of their activity, but even among the most sophisticated, the role of IT supported rather than transformed the way they delivered care. The goal of the hospital is after all for the hospital to thrive and prosper, not for the community to improve the care it gets at a lower cost. This has continued as IT investment has picked up, especially since 2003, and of course it’s been reflected in the stock price and profitability of the winners in health IT, notably Cerner and Epic.
Some other countries were introducing more clinically-based IT into their health care systems over the 1990s, and they tended to place it where their systems were focused—which tended to mean in primary care. The three Ns (Norway, Netherlands, New Zealand) along with the Danes & the Brits were close to 100% EMR use in office based primary care by the early 2000s. The UK was able to use this to actually track what its primary care docs were doing, and started paying them bonuses for doing the “right thing” in terms of prevention and evidence based medicine. But note that the IT was shoe-horned into a system which was by and large doing that already. It did not change the core way care was delivered. In fact in the UK the links between primary care (GPs) and specialty care (hospitals) are still by and large absent—despite a huge budgeted (and not fully spent) investment in Healthcare IT that exceeds what’s in the current US stimulus package for a country 1/5th the population! The astute observer will also notice that several countries that don’t have EMRs in universal use among physicians, notably France and Germany, also have pretty well regarded health care systems.
So it’s a pretty fair assumption that IT use in health care will reflect the system it’s put into, rather than transform it to something completely different.
So why all the aggro? Since the Obama win and the recession made the stimulus a certainty and health care IT’s inclusion in it a reality, THCB has been the venue for a series of articles in the form of Open Letters to the Obama Administration by David Kibbe & Brian Klepper, a defense of the proposed spending by John Halamka, an analysis of the perversion of administrative simplification under HIPAA by interested parties (clearinghouses) from Rick Peters, all culminating in a declaration that Health IT is in fact ready, just add $20 billion, by Mark Leavitt yesterday. This argument has by no means been confined to THCB, and in fact this latest discussion was started mostly in a series in Health Affairs, in which Carol Diamond and Clay Shirky accused the current health IT establishment of “magical thinking”, while somewhat surprisingly Microsoft—which sells a lot more in copies of Windows and Office to hospitals than health care tools to consumers or small practices—in the person of its health care leader Peter Neupert suggested that we should take a pause before hitting the “buy” key, and spend money on outcomes not on technology for its own sake. And enfant terrible Jonathan Bush is warning anyone who’ll listen that we’ll be locking in place outmoded technology.
Essentially the argument comes down to two things (Warning—gross generalizations ahead!). I liken this to the common truism that dogs attach themselves to people while cats attach themselves to places
Outcomes vs. technology: The Kibbe/Klepper/Peters/Neupert/Bush faction (the dogs)assumes that we need to change the incentives in the system, and then IT will naturally follow—and the current embryonic decision support systems will flourish quickly. But that current clinical systems aren’t good enough to invest in mostly because current results from EMR installations are very disappointing.
Although they discuss incentives (and most of the $20 billion will probably be aligned with some P4P measures), the cats’ (Leavitt/Halamka/Kolodner) view is closer to the thought that if you get the appropriate clinical technology (essentially in-patient and outpatient EMRs) into the hands of clinicians, then they’ll figure out what to do with it, and eventually the government can pay them according to how well they do it. (Halamka is a bit more open about this also being an IT public works scheme).
Patients vs. facilities: There’s also a more philosophical bias which harkens back to the difference between American and European health care systems, but not in the way you might think. The dog faction is in general primarily in favor of light-weight tools (and standards) that allow for innovation and service of the consumer patient by primary care teams (that’s the European part). The recent emergence of Web-based tools and patient communities that allow patients to apply self-service techniques and easy communication with teams of providers (yes, yes, that’s Health 2.0) are for them the keys to enabling better care. These tools are relatively cheap (and flexible) and mirror the SaaS trends in the rest of technology. (Think Gmail vs Outlook).
The cats’ view is closer to the opinion that the real work in American health care happens in big hospital systems, and that the key is to get everyone connected to their core clinical systems. Hence the concern with standardizing on products and private networks (RHIOs) rather than allowing a mass of anarchic applications out on the Internet which are more likely to “mash-up” together.
What about the biases? There have been plenty of accusations of bias and self-interest on these pages, and it’s worth quickly looking at it. For example, Jonathan Bush believes he’s riding a wave towards a SaaS future, whereas most of his competitors are still MUMPS-based client sever technologies—he fears the $20bn will be used to subsidize SUVs in a world where AthenaHealth is selling the Prius. Mark Leavitt, while on loan to the Chairman of CCHIT is the former CEO of MedicaLogic (now the core of GE’s Centricity EMR product) and a former HIMSS official. Most people in health care think of HIMSS as a vendor promotion association. Halamka of course is part of a very large AMC and the chair of the HITSP standards organization which tends to favor larger systems, whereas Kibbe is well know for his work promoting the lighter weight CCR standard and his involvement in the American Academy of Family Physicians—long the orphans of our specialty dominated medical culture. I also suspect that Kibbe’s recent criticisms of the EMR haven’t exactly helped his chance of receiving consulting dollars from the major HIT vendors, but he is also involved in consulting with smaller companies that stand to benefit from the emergence of new models of care. And of course Brian Klepper and I are involved in Health 2.0 (together in a consulting venture and me as co-founder of a conference).
However, I perhaps naively choose to believe that everyone involved is motivated more by their cultural takes on what’s important for health care, than their desire to put a hand-out for the $20 billion. I think that these same positions would have been taken by the same people 6 months ago when the bailout was yet a twinkle in Obama’s eye.
So in change, can we find unity? So despite the fierce words and the grave differences of opinion, can cats and dogs live together in harmony? I think that in the spirit of Obama’s first days in power we can find some unity. At the same time we need to be realistic about what comes next.
First, we need to get rid of the notion that IT is going to transform our health care system. To transform healthcare we need to get real health reform right. Obama is not doing IT any favors by imposing that pressure on its use—the dogs are correct here.
Second, we are not immediately going to move away from an acute care, large system-focused health care environment. Any change towards overall more consumer-focused health management will continue to be slower than some of us might like. So getting those acute care facilities that are not at BIDMC or InterMountain levels of technology readiness in better shape can’t hurt.
Third, despite all the harsh words, as the bill now stands, most of the $20 billion will reward actions with some form of P4P. If the folks at HHS & CMS who will be responsible for crafting the incentive structures hear a more unified voice from the assorted commentators, geeks, cats & dogs arguing over here, we’ll likely get more incentives rewarding better process and less direction on particular technologies—which will be a good thing.
Fourth, there’s an underlying concern among the dogs that in the cat view (which is more clearly the “establishment view”) the patient is lost. Yet I still see Leavitt and the others from the establishment keen to bridge the gap to patients with connectivity tools and to give them control over their data in PHRs. And I don’t believe anyone is suggesting that we halt the rapid development in the ability of patients to manage themselves, and communicate with each other and clinicians, which is at the heart of Health 2.0.
So I have hope that, while the health IT part of the bailout will not be the ultimate cure for our health care system, it will help in many ways all over the system. I also hope that while there won’t be complete consensus both cats and dogs will agree that it should fit some core basic principles so that in years to come we’ll judge that this money was wisely spent.
Of course I’ll be reserving judgment at least till the bill makes it out of committee, and probably until we see exactly how CMS wants to spend the money!
Categories: Matthew Holt