Build Me a Car, If You Can (But Will It Be a Boat Instead?) | Mark Hagland | Healthcare Blogs Skip to content Skip to navigation

Build Me a Car, If You Can (But Will It Be a Boat Instead?)

July 28, 2010
| Reprints

Attending the eHealth Initiative’s one-day forum on health information exchanges (HIEs) last week in Washington, D.C., seemed to validate all the perceptions I’d been gathering over the past year about the challenges and opportunities inherent in the HIE concept at this point in time. On the one hand, some real progress is being made, in places as far-flung as Vermont, Louisiana, and Arizona. On the other hand, the path has been, and continues to be, remarkably difficult, for a wide variety of reasons associated with funding, governance, affiliation, process, and other issues. And the technology is the least of it, though there absolutely do remain some tech issues as well.

Fundamentally, because every HIE is emerging out of a different individual situation, embedded in particularized geography, demography, industry relationships, and so on, every HIE is turning out rather differently, even given some very fundamental commonalities at certain deep levels. As a result, the situation is maddeningly complex, and doesn’t augur well for the rapid development of HIEs that federal authorities, policy leaders, and industry thought-leaders have been hoping for.

All this reminds me very much of reading about the early development of the automobile a little over a century ago. As hard to believe as it might seem, there was a very early phase in which it wasn’t even clear where the various basic parts of a car should go—the gas pedal, the brake, the clutch, the steering wheel, the driver’s seat, and so on. Not surprisingly, there were terrible road accidents as the very first, intrepid, drivers cranked up their engines and vroomed down the road (and not always on a universally agreed-upon side of the road, either!). It took a while, in fact, for a number of manufacturing, rules of the road, and other conventions to evolve forward.

The challenge with HIEs is that they are so very much needed, and that their development is proceeding forward in a time of intensified requirements and pressures. How long can we sustain this completely individualized model of HIE development? On the other hand, the alternative could well be the rigid development of exchanges that don’t serve the individual needs of their stakeholders or communities, and that would be even more frustrating. Just think back to the CHINs of the 1990s, or even the RHIOs of a few years ago—which were the same as the HIEs of now, except that the early and widespread failures of RHIOs led to their being renamed as HIEs…!

This is “so healthcare,” I have to say. But at this crucial juncture in HIE evolution, there seems to be little alternative right now. Hopefully, within the next two years or so, enough successful models will have evolved forward to the point where those who follow in their footsteps will be able to quickly choose from a relatively small-ish number of models to emulate. We certainly can’t afford ten years of tinkering before everyone agrees where the carburetor should be.



Thanks so much for your perspective. As you know, I have read two of your recent books (improvement, transformation and EMRs) and close to all of your articles in the last few years. Your national perspective on what you learned at eHI is especially well-informed and salient to readers.

Your example reminded me of something Thomas Friedman wrote in The World Is Flat. He said that, after electrical power and it's transmission became available, it took industry about fifty years to adapt to exploit the new realities. Prior to that, factories had belts that spanned the shop floors, conveying kinetic energy to workstations. As a result, the buildings tended to be long and flat, and the machinery was build to be compatible with the power delivery.

Within the decades that followed, we learned how to build and exploit relatively small electric motors to address the same workstations. This eliminated the geometric physical constraints of conveying belts. It made it possible to have small motors and larger motors, customized to the torque, speed and other mechanical differences in end-user needs.  Location became independent.  Complete re-engineering was possible and needed.

I'm pretty sure it wont take healthcare fifty years to retool to the new materials, engines, services, and delivery paradigms. I think we all know that 18 months would unprecedentedly short. And to your important point, we're not likely to get the desirable standardization, especially given the lack of policy standardization coming from the states.
If HIEs are going to facilitate better benefits/service management and risk pooling under our current healthcare reform mandates (individual, employer, states, federal mandates etc), the variation you outline will only delay that transformation, in political proportions.

Incidentally, some readers may recall that J D Kleinke described the solution to exactly this problem a decade ago in Oxymorons: The Myth of a U.S. Health Care System [Hardcover] (J. D. Kleinke). He explicitly called out the state-level variation problem and proposed what appeared to be a tenable solution.

Was this topic explicitly developed at eHI, Mark?

Thanks for your fantastic comment!!! First of all, thank you (I bow humbly) for all your compliments to me and for your following my writings so assiduouslywow. :-) Second, I'm very heartened by your expressing your thought that it won't take so terribly long for healthcare and healthcare IT to move forward in the HIE arena as it did for industry to learn to efficiently use electrical power! If it did end up taking so long, we'd all be in a heap of trouble...! Per the topic of state-level variation, the forum was organized as a series of panel discussions, and essentially, it was covered conversationally and was touched upon during each discussion.
Unfortunately, there's no simple answer or easy solution to the variation problem, and in fact, that problem goes back to my original post, I think, in that if we continue to produce such totally different HIEs, it will be hard for the latecomers to figure out what to do. What I'm guessing (and it's only a guess) is, that within three or four years, some, say, six or seven basic HIE models will have become clearer and more crystallized across the industry, and most people who are only now beginning to develop HIEs, will be able to pick from those somewhat clearly differentiated models. That's about as clear as anything seems right now, alas...!
Thanks again for your fantastic post just now, Joe! You're obviously paying attention!!!