Matthew Holt

Health IT policy: the fur is flying

Some fur is flying in the rarefied world of health IT policy geeks this morning. Health Affairs has three articles. The first from Markle’s Carol Diamond, writing with Here Comes Everybody author and Internet guru Clay Shirky, more or less says that obsessive attention to rigid standards is not helping and actually may be hindering the IT adoption process. And yes, in case you were wondering they do mean CCHIT and ONCHIT’s current policies and agenda which has been going for four years and which they’re accusing of “magical thinking.” Instead, we need new policies which target desired outcomes measured in improved patient care, instead of assuming that creating new technology standards will get us there. And by policies I think they mean money, and its redirection by current payers. After all, if putting in a RHIO costs hospitals operating revenue in reducing admissions and tests, why would they do it?

Not so fast, says Rob Kolodner (director of ONCHIT) who (writing
with Simon Cohn & Charles Friedman) says that on the four core
measures of adoption, governance, privacy and security, and
interoperability, real progress has been made in four years.
And they call it a foundation, even if no tipping point has yet been
reached. Of course, what they cannot say is what they’d do it if there
were any federal money available for the effort. Obama’s plan says that
there’ll be $10 billion a year for health IT. Right now ONCHIT’s budget
can barely buy coffee for Kolodner and his staff.

Finally, David Kibbe (yup, Health 2.0’s friend) and Curtis
McLaughlin (adjunct proffessor of business at UNC Chapel Hill) accuse
federal health IT policy of being the drunk looking under the lamp post
for the keys. Out in the dark are all the “unmentionable” Web-based tools and services that have transformed other industries, and there is clearly plenty of work going on it that sphere right now in health care.

But instead the Health IT standards are being set by representatives
of the big health IT vendors and the providers they work for. None of
them are particularly interested in interoperability or data
transference — as the potential efficiency of those concepts bring
reduces their potential revenue. After all the vendors make lots of
money building one-off interfaces between systems for their clients,
and the providers from doing unnecessary test and procedures.

They also perceive data fungibility as leading to direct competitive threats, and they are right to do so. If I can easily move my data to anywhere, why would I stick with opinions from my local providers and not move it to say, the Cleveland Clinic, Johns Hokpins or Partner’s online service? And if Google or Microsoft or an open source solution can provide 80 to 90 percent of the efficacy of a Cerner (and no they’re not there yet), Cerner is right to be worried.

Of course, as everyone knows and only Kolodner doesn’t say (because he can’t) there are only two solutions.

Either we reform the financing system (a la Enthoven or Holland) so that health care organizations compete on who can deliver population health most efficiently. Then they’ll start changing their processes and using the technology that other corporations are using to do that, or they’ll be chased out by competitors who do.

Or, we do what everyone else in the world has done, and accept that the current health system is what we have to work with, and that we aren’t prepared to see that type of market dislocation. Instead, we have the government fund the necessary changes in IT, and insist that all those in the system use it. By the way, this socialist diktak is Newt Gingrich’s preferred solution.

Neither of these solutions precludes the Health 2.0 solutions and technology advances that David advocates. In fact they both would make them more likely to be successful. But as ever (and as maybe Porter and Teisberg might one day discover) incentives come first.

And of course, I’d be remiss in my shameless marketing duties if I did not point out that Clay Shirky, Carol Diamond, Rob Kolodner and David Kibbe will all be at the Health 2.0 Conference to pick up on this conversation!

11 replies »

  1. Mattehew:
    This analysis is spot on. I know Dr. Kibbie, Rob Kolodner, and many others in the field are good people wanting to do the right thing and get the data flowing.
    We (well most of us) all want the same thing….a health care system that works and is a model of an efficient and kind, public/private partnership.
    I’d really love to see the market shake out a new interoperability specifications on its own. Let’s hope this is what happens with the CCR or the PCHRI format, but I’m not so sure about what will happen with the efforts lead by HSS/HITSP.
    Right now, with “government” standards, mainly based on HL7, are the status quo for most established IT vendors. As you pointed out, there is currently little incentive by the established players to make data fungible. I’m not sure change will happen there without a little brute force from the top down. Rob Kolodner has his plate full for sure….it must be a little like herding cats that belong to your neighboors.
    Just to contrast health care standards to identity/biometrics standards, in the the ladder, formats are controlled by the US government through NIST and not indirectly by non-profit organizations as is the case with HL7. Still all vendors, government, and other stakeholders have a chance to come together, develop, and agree to formats. Ultimately the standard is published and is made freely available by the U.S Government (NIST). There are no groups to join, fees to access the format, etc, etc. The biometrics world has achieved a considerable amount of interoperability compared to health care and still manages to involve all of the vendors and stakeholders with NIST at the helm.
    Case in point, NIST-ITL 2007 Biometric Standard:
    I’m not necessarily saying that NIST should be in charge of standards over HHS (That is another debate altogether), but I am asking your readers to consider that if there must be “government blessed” standards then shouldn’t those standards be published by a “government” agency? And shouldn’t any such work for the “public good” be a PUBLIC work, freely available to anyone or any group who cares? After all, either way tax dollars are being spent to create it these formats.
    The public/private health related format I’ve seen that most closely resembles this more open vision to date is the PCHRI format being used by DOSIA and others. Its all there on the Internet for the world in inspect. This is a very open approach that is somehow still novel in health care. I’d like to see this effort go somewhere too.
    I guess we’ll just keep supporting all the formats as best we can until all this shakes out in a few more years (at least).
    Looking forward to Health 2.0 in October!
    Alan Viars

  2. Matthew,
    Thanks for the excellent summary of the new Health Affairs papers on HIT. We’ve also just posted some commentaries on the Health Affairs Blog by Esther Dyson, Mark Leavitt, and Nancy Davenport-Ennis. Leavitt gives 1 to 5-star ratings on health IT’s progress in overcoming barriers to adoption.
    And Dyson takes on consumers’ fear of sharing health information with insurers: “This will all come to a head. Too much knowledge will ultimately make the insurance business untenable as it is currently structured: It is ignorance about predictable outcomes that enables most people to be insured. As that ignorance dissipates, our system will have to pay health care providers for improvements over the predicted outcomes — and will have to subsidize those whose predicted outcomes are worst. How we get there will be complex, and it will involve politics.”

  3. Matthew,
    Thanks for this post that brings light to these three papers and puts them all in context.
    The funny thing about this whole debate is that what is happening in healthcare with regards to IT adoption, use of standards, data exchange, etc. is no different than any other industry sector. Drawing reference to the last industry sector I was in, which happens to be have about a 20 year lead on the healthcare sector in IT adoption, manufacturing, it is a “Supply Chain” issue.
    Standards, be they de facto or formed by some committee are only aggressively adopted when there is enough pain in the system. I believe the healthcare sector is getting to that breaking point and it really does not matter what the feds do, adoption will occur when their is enough pain and subsequently money to be gained or saved by the key stakeholders. Like manufacturing, these stakeholders will adopt standards for their supply chain (in this case hospital(s), affiliated physician practices, clinics, labs, consumers) when their is a business case to do so. We are beginning to see such with the growth in HIEs.
    As I argued to an HHS executive recently, what is needed in this sector is not more trumpet blowing pronouncements for NHIN, but a more pragmatic, thoughtful and economic analysis of what the key levers are that will drive adoption. My hunch, it has very little to do with standards, certifications and the like and more about customer retention.

  4. Russell –
    There are no “Mother Theresas” in HIT yet, but there will be in 2 months.
    Matthew, this is right on point. Can’t wait to hear these guys in Frisco.

  5. I love the fresh debate about standards and the evolution of Health IT raised by the three Health Affairs articles. I agree wholeheartedly with the perspectives of Carol and Clay – and frequently reinforce these comments to customers, policy makers and audiences alike: value in the form patient care and business results can be improved by moving/reusing the data already in the system! – there is no need to wait for ‘standards’; health IT is a tool for better somethings (outcomes, safety, results, employee productivity, employee satisfaction) and not an end in itself; system design matters a lot; metadata is the answer to enabling exchange of info today to evolve to standard exchange tomorrow. These beliefs have informed the design principles of the software products we introduced in the marketplace – both HealthVault and Amalga.
    I believe consumers will demand ‘connected’ care and will increasingly make physician choices based on the ability and willingness of physicians to leverage communications/connected care to improve patient convenience and outcomes.
    In addition to consumers as a change agent – I remain hopeful that the buyers of large health IT systems will wake up and demand more from their vendors…not in terms of custom features… but in terms of a real commitment to interoperability and to unlocking the data that exists in systems already. Health IT buyers are critical stakeholders/components of the ecosystem – and need to demonstrate leadership in getting us to real solutions/value from HIT – and not let themselves be positioned as victims controlled by the vendors.

  6. Nice overview, the heart of which is encapsulated in the following:
    But instead the Health IT standards are being set by representatives of the big health IT vendors and the providers they work for. None of them are particularly interested in interoperability or data transference — as the potential efficiency of those concepts bring reduces their potential revenue. After all the vendors make lots of money building one-off interfaces between systems for their clients, and the providers from doing unnecessary test and procedures.
    I agree 100% with this, however it is not clear that it follows that we should be suspicious of the standards themselves. The fact that it isn’t in the institutional players’ narrow self-interest to make healthcare more efficient slows down the rate at which the standards are established and used by provider organizations. However, I still don’t see why we should be suspicious of the standards themselves as effective means to exchange data, once the will is there to use them.
    Creating the will to do so will could take 15 years of gradually increasing pressure from two market sources (consumerism demand for all online information and service, and purchaser demand for improved efficiency and value), or it could take 5 years with government mandates and incentives.

  7. Oh, yes. Thank God, there are so many future Mother Teresas in the health I.T. vendor sector, so self-less. Like the old Peking Youth Corps.

  8. As a small ‘best of breed’ provider of decsion support software, I have a keen interest in this issue. This post has about as clear an explanation of the current context as I’ve seen. I live and breath the friction between solutions that are reasonably priced, highly efficient and clearly better for patient care as well as clinical practice and the need for big health care IT vendors to maximize revenue opportunities and health system IT departments to ‘complete the mission’of rolling these systems out.
    It’s shameful that we waste so much time and capital (to the large vendors benefit)while less expensive, more efficient and more effective solutions- which the CLINICIANS in the health system agree are better- are left under utilized.

  9. From a very big picture perspective what seems clear is that there is no easy glide path for successful wide-spread, value-producing health IT development or adoption. I’ve written about the practical and financial problems of EMR and e-prescribing at, and from the comments I’ve received (both public and private), it is obvious that there are wide ranges of experiences and value produced depending upon the HIT system, and the user’s previous experience and expectations…. And I’m sure that the development, dialogue and controversy will continue.