For several years, I’ve been citing dismal statistics on the growth of health information exchange networks. Perhaps, back then, I was too hard on them. After all, fledgling, starry-eyed HIE groups were facing tough odds, given how few physicians and hospitals were even wired enough to support their efforts.

Fast forward to today, and it seems little has changed. Though hospitals and medical practices are going online in large numbers, the HIE business model still seems to be shaky.  The latest evidence of this comes from a study from the Harvard Business School, which concluded that — surprise, surprise — that most HIEs still aren’t financially viable.

The study, which collected survey results from 165 HIE groups, concluded that just 75 of these organizations were actually up and running. Those 75 groups are probably working very hard, but still only reach 14 percent of U.S. hospitals and three percent of smaller medical practices. And get this: only three of the 75 groups offer a data exchange model which supports Meaningful Use standards. Wow.

Not only that, most of the HIEs studied don’t seem to have a sustainable business model. Two-thirds of the operating HIEs ended up in poor financial shape once they burned through initial hospital and physician funding, the study’s authors found.

Now, it’s worth noting that the study’s authors collected their data in late 2009 and early 2010, and heaven knows EMR penetration, interoperability and health data exchange are moving targets. If HIEs were just starting out now they might have had more momentum.

The unfortunate truth is, however, that HIEs have faced a nasty chicken-and-egg problem; if they wait for providers to get up to speed they’dllnever get rolling, but they’re having trouble making it without enough provider support.

At some point, the provider community’s going to have to decide how serious it is about data sharing, and whether leaders are willing to invest in this model over the long term.  Waffling, posturing and playing chicken (i.e. “let’s see if anyone else is willing to spend money on this”) obviously aren’t going to work.

When you read this statement from HIT vendor Orion Health, it sounds oh-so-simple: why not establish an entire county’s HIE network on a single connecting platform?  Given the country’s already high EMR adoption rate — about 80 percent of GPs had one, as of March 2010 — New Zealand’s already part-way there.  Just knit offices up together and you’re ready to go.

Orion, of course has its own technology in mind, naturally. But whatever vendor you use, they may be onto something. I’ll pause here to say that the following proposal could incite a riot at a HIMSS floor full of competing vendors, but hey, ideas are harmless, aren’t they?

What if CMS decided that it would pay incentives not just to meaningfully, sensitively, insightfully install EMRs, but to connect them to an overall HIE?  And to take the thought into more controversial territory, what if it had a vendor or two of choice which doctors and hospitals had to use if they wanted the dough?

As we all know, the value of EMR installations isn’t just in automating, error checking and (hopefully) streamlining workflow in practices. The data is infinitely more valuable when it can be aggregated, shared, cross-checked and mined for best practices.

What are the odds of that, however, if you have an outbreak of regional and state projects using technology from a multitude of vendors?  You can talk standards all you want, but true interoperability isn’t going to happen anytime soon this way.  National connectivity?  Well, give me a couple of decades and let’s see how far that’s gotten.

On the other hand,  if CMS signed contracts with HIE technology vendors, and demanded that they give preferred pricing to those work with them, you’d see a rash of connectivity unrivaled since the invention of the telephone.  Before you scream that this just isn’t fair, doesn’t this kind of thing happen every day in, say, military contracting?

I know, I know, this may not be practical. But you can’t argue that It’d be interesting to see how the HIE and EMR market gelled if CMS took a strong lead.

As anyone reading this must know, Japan now faces horrific logistical problems in the wake of the earthquake and tsunami that hit on March 11.

While I’m sure Japanese officials have more urgent issues on their hands, one that will arise when caring for the injured is that natural disasters wipe out paper records — including medical charts, of course.

Japan does have an HIE strategy under way, funded with billions in stimulus funds, but it hasn’t had time to mature. EMR coverage is patchy, though like here in the U.S., on the upswing.  Research interest is high; in fact, medical informatics specialists there have developed their own clinical data exchange format, the Medical Markup Language. All told, however, health data digitization and sharing is still in its early stages, or so it seems from the reading I’ve done.

Now, back to the present. In a time of national disaster, wouldn’t it  have been great for Japan to have a robust HIE network in place, one which not only made it easy to share data but provided for backups offshore that wouldn’t be washed away by a wall of ocean water?  In fact, wouldn’t any country or region ride out disasters better with health data sharing in place?

While I’m not suggesting data portability in times of emergency is the most important reason for building out HIEs/EMRs, it’s a good public health tool when clinicians have to work fast.   It wouldn’t hurt when pandemics strike, either — just imagine the good a sophisticated, mature national health network in could have done in tracking, tracing and treating H1N1 when the illness was at its peak.

Just one more thing to think about as we argue, worry about costs and in some cases, drag our feet over EMR launches.

I’ll admit beforehand that I’m a member of the Fred Trotter fan club. He’s a little bit psycho when it comes to open source licensing and the like, but that’s probably why I love him so much. When he truly believes in something he’s fully engaged in that cause.

So, of course I am completely interested in Fred Trotter’s blog post about CONNECT where he said the following:

The right conversation starts with this: we can assume that CONNECT -will- unify the health information transfer in the US. It will serve as the basis for the core NHIN and regional networks will have the option of implementing it. That means that CONNECT sets the bar for health exchange. Software must be as good as CONNECT to be considered for a local Health Information Exchange, otherwise, why not use CONNECT?

I think this is the second time that I’ve heard the name of the project CONNECT like this. I think that’s a sign that I better do some more looking into this project.