Chaordics Revisited? Stark HR6898 | [node:field-byline] | Healthcare Blogs Skip to content Skip to navigation

Chaordics Revisited? Stark HR6898

September 30, 2008
by Joe Bormel
| Reprints

The role of government regulation in U.S. healthcare and HCIT is complex and probably necessarily so.

The legislation introduced on September 15, 2008, by Representative Stark, Chairman of the House Ways and Means Committee’s Subcommittee on Health, and the subsequent responses are important for the leaders who read the HCI blog. Here is

a HIMSS entry-point link which outlines the high level issues and resources. I tried to find the single best, unbiased link for you that was short, authoritative, raised the issues clearly, and was smartly done. It also offers definitions and clarifications on the terms of the

open source issue.

I would be interested in your take on this complicated issue.

Mine, briefly is as follows. I use and value both OSS and proprietary software offerings. For specific purposes, one is considerably more cost effective than the other, and it goes both ways. The largest software and services vendors do the same, and have put interesting offerings for free distribution into the open source space, as well as proprietary or "closed" offerings. I like having the market to choose from.

As most articles covering this legistlation note, the VA software is a proven example of how well 'open source software' can work. The benefits to healthcare deliver are real and are well documented;

here's a great report that elaborates the great work of the Veterans' Health Care in a balanced way. Their work with bar code medication administration is industry leading and well recognized. The challenges on other side of the ledger, most of which are also challenges for commercial software vendors, are rarely if ever publiclly discussed. This mismatch makes me uncomfortable: the 'happy talk' about large, enterprise class open source initiatives, and the realities of software development across industries, when dozens of applications interoperate on a scalable platform.

There also appear to be strong parallels between R & D return-on-investment business issues between how pharma brings innovation to markets and how that's done with I.T. in the private sector. The HIMSS fact sheet notes that: "HIMSS’ concern is focused on the

federal government’s development of a system; not an opposition to open source software." This is no small or simple issue; we all remember Medicare Part D debates.

So, there are at least three, distinct issues for us to understand and perhaps discuss here:

1) Do we want a balanced playing field for commercial and open source offerings to compete?

2) Do we want the government developing software?

3) Do we want to undermine work under AHIC 2.0, which focuses on creating interoperable standards and certification, to allow the consumers of HCIT to select knowledge-ably from commercial and open source offerings, though open markets for both? (See final link on chaordics)

I started off stating that the role of government is complex. I would welcome comments that would add to everyone's clarity on any issue discussed here. Is it as simple as defining

where and how we structure HCIT competition?



The VA's commitment to re-architect the VISTA software to use Service Oriented Architecture is just one of many ways they have shown leadership in Healthcare software development. Ask any Veteran (myself included) about how efficient the VA information system is at capturing your personal health information. They value the need to integrate not only across a regional enterprise, but nation wide. They can run circles across any RHIO in terms of data exchange and MPI. I would have residents and physicians that would work rotations at the local VA regional hospital and come back to my Academic Medical Center just glowing about VISTA and asking why we can't get something like that.
If you take the top 5 Health Information Systems vendors, you will see a market in which some vendors pick the client, not the other way around. Organizations are sometimes bound to a vendor because the cost associate to change is astronomical in terms of project costs, support training and organizational impact. I know of only one vendor that capitalizes on the VISTA open source and offers it to the private sector along with a support model. I think the top 5 vendors needs to be aware that when VISTA completes its SOA migration, commercial vendors will look like the PC guy on the MAC commercials.

Dr. Joe,
Good posting. Nice to see you're staying objective with a balanced presentation of such a controversial subject. I'll take a crack at replying to your three issue questions.

1. Of course we should want a balanced playing field, at least in terms of the government not skewing support for one category of offering over another. It is no more the government's place to endorse a specific technology than it is for the feds to endorse one product or company over another. And in fact, government regulations prevent such product or company endorsements with few exceptions. The same should hold true for technology.

2. My first reaction is to say that I don't want government developing software. That's a job for the private sector. However, let's face facts.

The vendors have had more than a decade of warning that they need to develop interoperable systems. And many claim to have done just that, until you read the fine print in their disclaimers which read something like, "but only with our stuff."

To date, most vendors of large, complex HCIT, particularly the clinicals, have produced proprietary systems that lock-in customers and protect long-term cash flow.

Therefore, I say let VA move forward. If the agency does come up with a viable solution, the code will be available through FOIA as is the VistA system. It's too bad that not more than one vendor adopted VistA for commercial use. If more had done so, we may not be having this discussion. But if VA does take the bull by the horns, at least a number of private sector developers will share in the profits from bringing a system to fruition.

3. With a proper balance of participation in HCIT by the private and public sectors, I don't see us undermining work under AHIC 2.0. I do not think it's appropriate for me to go into a rant here, but the private sector has had plenty of time to create interop standards, etc., and shirked its responsibility.

So now the vendors are facing a good, hard kick when they could have heeded the rather subtle warnings that were issued years ago. But all that aside, healthcare providers should be able to systems from multiple offerings without being subjected to, at their level, pressure from the federal government. Per one of your previous postings, one size does not fit all.

Before closing, I would like to bring up two additional points. First, we should all understand why HIMSS does not oppose development of open source software, only government's development of the same. How could HIMSS take any onther position? This organization represents the vendors! Only the vendors. Let's not kid ourselves here.

Second, Rep. Stark, in his 25-year tenure in Congress, has not exactly built a reputation for following through in the various battles he brings to the table. On any number of issues over the years, it appears to me that he's more than willing to get into the ring for a few rounds, make a few headlines, and then disappear from the fray. That said, his propsed bill, as likely as not, will simply disappear in committee or be changed in such a way as to not resemble the original draft. That would make a good portion of this disucssion moot.


Following on Pete's comment above, and my comment on HL7 V3 above, there's this:

GCN: HUD embraces SOA

The connection per Pete's comment is that the VA embraces SOA as well. 

Although SOA will dramatically improve the ability to create much richer, machine-to-machine interoperability for services, the fundamental challenges we're discussing remain.  They are, paying for the development and software lifecycle, modeling the underlying transactions (right level of granularity, performance, interoperability, scaling, etc ... all recognized challenges in the SOA world), and fitting into the broader context where we all live.

Pete's right.  SOA is an important evolution.  It's not a fast panacea, nor an inexpensive one to the bigger HCIT challenges we've been discussing.

Jack, Wow! Thanks for the sobering commentary.

I agree with your themes, and I do appreciate the insights about Stark, HIMSS, and the vendors. They're exactly why blogs are such a great communication vehicle. They're two-directional conversations, and, between people who've generally earned the right to have an opinion!

Your point that the private sector, a.k.a. proprietary vendors, have had their chance and "shirked it..." feels a little harsh. The definition of needs in the industry has changed. Ten years ago, systems to manage departments, like pharmacy, were the generally understood need. Inter-operating with the lab department was seen as good but not a duty. The market wasn't looking for, in this example, a pharmacy system that could correctly reason over microbiology laboratory results to get the antibiotic choice correct (and keep it correct, in real time, as new information became available).

"Closed-loop" medication management has been a work-in-progress in most of the industry for 15 years or less. Sure, there are a few notable exceptions.

Only the most visionary are starting to understand that, in the same way, "Closed-loop" [unified] communication and collaboration underlie closed-loop medication management, the paradigm goes much further. And, of course, workflow management demands that degree of tightness.  For example, how an medication order evolves to a dispense and administration event is an important property of closed-loop medication management.  People are starting to realize that the full life cycle of a result/task/message coming to or leaving an inbox can be just as critical, and unifying the communications between individuals requires specific focus.  This is undeveloped or under-developed in many commercial systems, as attested to by the panoply of little vendors, sitting around big vendors, to get specific workflows to work adequately.

In short, any software developer, public or private, has to deal with rapidly evolving and rising expectations. Doing that more slowly than we'd like is a common problem that shouldn't be equated to shirking responsibilities. That's just too simplistic.

That said, your parallel point that these private sector folks have invented standards, rather than discovering and harmonizing to the kinds of standards that underlie IHE and the CCHIT certification is a separate point.

There, too, there's a simpler explanation that's more concordant with the facts (cf deliberately heightening barriers-to-entry for competitors). There are simply too few people in the world who adequately understand our best hope for semantic interoperability, that is, HL7v3 Development Methodology. That's the privately shared opinion of the most insightful and knowledgeable people who share their opinion on this topic. If this wasn't the case, the in-suite behavior of any major vendor would be considerably stronger than we know it is.  (See Wes Rishel's Gartner syndicated content, G00140951, for explicit elaboration of this issue.)

Thanks again for your thoughtful contribution to this topic.

Why the title: Chaordics Revisited (I received emails asking for clarification.)

It seemed to me that, at the heart of HR6898, and the point that Jack raises above, are the issues of  how should we, as a society, organize to create and distribute the software we'd like to create value in HCIT.  Software to coordinate care, manage orders, results, documents, tasks on the grainy level, and outcomes, performance, quality and costs at the macro level.  As we discussed above, there's almost certainly a need for commercial, open-source, and effectively hybrid models that will be needed.  (I'm sure lots of folks are anxious around 'hybird' - an example is the Apple OS-X operating system, which has a traditional software license, and contains BSD Unix, the Apache web server, and tons of other OSS components, often wrapped by the acronym, LAMP; similar hybrids exist in IBM's WebSphere; the list goes on.)

Chaordic is a term that gets at how to structure these hybrids, of simultaneously cooperative and competitive elements.  It's a great story and is a deep, philosophic contributer to the VA's leadership thinking.  (Thanks to Rob Kolodner for introducing many of us, including me, to this body of work, 15 years ago and earlier.)

By Chaordic, I [quoting Dee W. Hock] mean any self-organizing, adaptive, nonlinear complex system, whether physical, biological, or social, the behavior of which exhibits characteristics of both order and chaos or loosely translated to business terminology, cooperation and competition.

... Why, ... , are organizations, whether governmental, commercial, educational or social, increasingly unable to manage their affairs? Why are individuals increasingly alienated from the organizations of which they are part? Why are commerce and society increasingly in disarray?

Chaordic Principles:

1.It must be equitably owned by all participants. No member should have intrinsic preferential position. All advantage must result from individual ability and initiative.

2.Power and function must be distributive to the maximum degree. No function should be performed by any part of the whole that could reasonably be done by any more peripheral part, and no power vested in any part that might reasonably be exercised by any lesser part.

3.Governance must be distributive. No individual, institution, and no combination of either or both should be able to dominate deliberations or control decisions.

4.It must be infinitely malleable yet extremely durable. It should be capable of constant, self-generated, modification of form or function without sacrificing its essential nature or embodied principle.

5.It must embrace diversity and change. It must attract people and institutions comfortable with such conditions and provide an environment in which they could flourish.

Anyone want or need references?  Does this make sense?

Thanks Pete. Not to mention their HealthEvet PHR.

This past summer, June 2008, MicroSoft and Wharton published an interview, audio podcast and transcript, here (

"Tom Hanrahan, Director of Linux Interoperability at Microsoft, speaks with Wharton Legal Studies and Business Ethics Professor Kevin Werbach about the value of open source development in today's marketplace."

In it, Hanrahan makes it crystal clear how and why it makes sense for MicroSoft to make sure the Windows Platform creates a great experience for OSS users, and how commercial and OSS fit together in an ecosystem. 

Thanks for your reply. In retrospect, I was too hard on the vendors, as stated in your second paragraph. Point well taken.

Also, your comment, "There are simply too few people in the world who adequately understand our best hope for semantic interoperability, that is, HL7v3 Development Methodology," certainly describes my overall understanding of the subject.

By the way, Pete Stark has been in Congress for 35, not 25 years as I stated. It was either a typo or a senior moment, but I just can't remember!