Comments on: Healthcare Interop and the ARRA: Hope Happens

By: Nicholas Bessmer

Nicholas Bessmer — Tue, 14 Apr 2009 18:13:10 +0000

The recent Harvard Medical School article about EMR states that the chief barrier to the acceptance of automation is the cost of implementation. My partner and I are working in the private sector on a start up that delivers a low cost data quality solution for HL7 (interoperability. We have found very few private equity funds who want to have any involvement with regulatory or standards based health care software systems.

I am still unclear whether stimulus money is available to start ups working on EMR / EHR solutions. No one seems to have the definitive answer.

By: Wes Rishel

Wes Rishel — Thu, 26 Feb 2009 22:17:53 +0000

Thanks, Joan. I think we are all aware that promoting healthcare IT in an environment with perverse incentives is trying to push a rope. An interesting moot question is would we recommend stopping Health IT incentives until the fundamental health incentives are in alignment?

The question is moot in the sense that we don’t get to say. It is not so moot if a hospital or vendor is determining a business strategy. If one really believes that improper incentives will doom EHR incentives and interoperability to failure, one should bet against it. That would mean gambling on CMS becoming the provider of health IT as it takes on a greater percentage of the population or a market that comes to be dominated by one or two very large vendors.

By: Joan Duke

Joan Duke — Thu, 26 Feb 2009 20:26:13 +0000

I agree with all Wes’s analysis. I am also very much aligned with Glenn Keet who wrote that “the really broken thing in U.S. health care is the current ‘pay for procedure’ reimbursement system”. It worries me that our reimbursement system does not yet provide the incentives for coordination of care that is facilitated by the exchange of clinical data. As long as our health care services are reimbursed on an incident/procedure basis (unrelated to outcome) rather than on an episode basis, there is not substantial financial reward for the expense or the changes in behavior need to promote the improvements that can be gained by access to information across the many settings of care.

By: Rob Bush

Rob Bush — Tue, 17 Feb 2009 21:40:53 +0000

Who is steering the ship?

Someone is working on a plan – right now – who has the blessing and support of the administration. But who knows who that is? As you clearly point out, the best plan will come from accumulating the knowledge of the people in the trenches who are working today to make interoperability happen.

How can this industry close the loop you describe by getting involved with the planning? I want to volunteer!

Rob Bush
Orchard Software

By: Patrice Kuppe

Patrice Kuppe — Tue, 17 Feb 2009 18:32:18 +0000

Wes,
Here we are over 10 years since we passed HIPAA “Administrative Simplification” and we are struggling with interoperability for those transactions. We have a new law in MN that requires providers to bill ALL PAYERS electronically (what a concept)! My company is advanced but we are finding we can’t achieve this mandate because we don’t have interoperabiltiy – as in connections. The workers comp, property and casulty, and auto insurers aren’t all linked up to clearinghouses. The ones I use don’t connect to theirs, or they don’t even have one. We have missed a great opportunity in the reduction of administrative costs in health care because we do not yet have seemless connectivity. We have the standards – we need a pipe.

By: Kenneth A. Kleinberg

Kenneth A. Kleinberg — Tue, 17 Feb 2009 18:14:56 +0000

If you had asked folks a couple of decades ago if they believed that effectively all the worlds computers and companies would be connected in one giant more-or-less interoperable network (the Internet), they would have laughed. Could it happen in healhcare? I’ve heard many say “not in my lifetime”. I tend to go along with what I heard another person recently say: “It can’t not happen.”. We learned a long time ago with OOP that getting folks to agree on one “perfect” data model” is unrealistic. Providing a lower common denominator of what information can be exchanged via the most needed and valuable use cases is the key. With some jumpstart to get more organziations exchanging information, the benefits of receiving will start to exceed the benefits of giving, and adoption will take off – probably faster than most of us could predict.

By: Glenn Keet

Glenn Keet — Tue, 17 Feb 2009 17:37:31 +0000

Wes, I believe your analysis is largely accurate. And I fully agree with the notion that the feedback loop is missing with regards to the standards bodies. Who ever really believed, though, that ARRA would be directly responsible for achieving interoperable health care? In my mind the free market is and has been working towards that, and I think that ARRA will only hasten it by making that free market bigger for a short period.

In my twenty year tenure of integrating health care records systems I have seen health care data interoperability progress from Rube Goldberg contraptions, on to wholly unreproducible science projects, and later on to fairly reproducible exchanges between unrelated commercial products that begin to make the health care data “liquid” (to steal the recent Booz Allen term). Yes it has a ways to go, but the market is already pushing vendors towards this.

So I suppose I am an optimist (and a Darwinist) when it comes to interoperability. There will be products that are interoperable with others in real and useful ways in my fiscal lifetime, while many other products that are not will disappear from the marketplace.

Now, the really broken thing in U.S. health care is the current ‘pay for procedure’ reimbursement system, and the challenge is to replace that without necessarily moving to a single payer model. ARRA doesn’t take any steps towards that, and that might be prudent since it is meant to be short term stimulus.

By: Jan Root

Jan Root — Tue, 17 Feb 2009 17:05:06 +0000

Wes,

As always, very good food for thought. Interoperability is definitely an iterative process. Not only does practice make perfect (or at least better) but inevitably we’re always playing catch-up as the industry comes up with new ‘silver bullets’ for how to lower health care costs (DRGs, P4P, etc).

However, I think it is key to remember that interoperability does not occur in a vacuum. It makes a lot of sense to us techy-types but it obviously does not to many others. For it to be truly sucessful, It must occur in a context where there is an economic incentive for the industry as a whole to not only adopt standards but require them. As you say, HIPAA has been a disaster in most of the country. However in Utah, it’s been successful. Why the difference?

In Utah a significant portion of the health care community has bought into the idea of standards as a way to grow market share and to reduce costs (long story). So, somehow we, as a country (or maybe region by region?) need to figure out how to move the industry from the ‘my way is the best (only?) way” to “interoperability (standards)” is the only way.
http://api.recaptcha.net/image?c=02Od1LsgJwDvASwjaNPfgzDBO4fRepo8p_181uz3gA5k-n4FUAgiMZybhYVRj-ala0vkT8Xl-r8e6uWfr-TTdjgX-ekzYlEjLbq8XJDr8Q47JBL44wDUA0MNleud-joC5AOVwb4wuTgrJAv-IQSHRSkpX9vTBBRHIp_iWWyHXq1dRnN8rIipikXehrINH6E4FelCsQQrk-fmr7lx9g6JpDQqejkh3v76oep6BStJJ3g5qNhsZUsmfSrwZy5gXXLrcmxbh5B_92EaYPmlGzVm3dLe8a-5oB
That’s a different discussion!

Thanks again for the interesting thoughts!

By: Andy Wiesenthal

Andy Wiesenthal — Tue, 17 Feb 2009 16:40:07 +0000

Wes–you have characterized the issue accurately, and the organic nature of NHIN development that you describe is likely to be much less expensive and much more likely to succeed over time. It will not succeed instantly, but we will get some effective data exchange fairly soon. That should be allowed to grow, and the rest of the country should be able to take advantage of successes and failures (what not to do). I would hesitate to point to the NHS as a successful venture, because they have created some important problems for themselves and do not have a successful deployment to point to or even on the horizon. Canda HealthInfoway and NEHTA in Australia look a little more promising at the moment.

By: John Mattison

John Mattison — Tue, 17 Feb 2009 16:37:33 +0000

Sorry to agree with you Wes, but I think your synthesis is spot on.
I would however extend one of your observations though, and that is that ” the goal will always be to integrate systems that are in different points in their life cycle and have different information models” . I agree that we will always be integrating systems in different points in their life cycle, but I haven’t given up hope on an incremental convergence on similar information models, hopefully based on the HL7 RIM. The current work of Peter Hendler, Gunther Shadow, and Rene Spronk and the RIMBAA group in HL7 holds promise for demonstrating if not inspiring such a migration. Clearly it is unrealistic to expect any vendor to change their information model except as an opportunistic response to a need to replatform an application. But when a vendor recognizes the need to replatform, it is often an opportune time to update their information model, and converging on a RIM-like information model could offer many attractive virtues, including lower cost of integration by potential purchasers. If this optimism is justified, there might even be a classical “tipping point”, at which having a RIM-based information model could be a critical criterion in system selection. Only time will tell, but given how painful and expensive it is today to integrate systems with different information models, one can only hope.