Yesterday evening my family and I went to see the new Spiderman movie (my wife has somewhat of a crush on Andrew Garfield). It was pretty good, although I still do not get why magnetizing the web shooters protected them against Electro. Whatever.
What really caught my eye was the trailer for Edge of Tomorrow, an upcoming show where Tom Cruise has to re-live the same day of battle over and over again (kind of like Groundhog Day but with aliens and Emily Blunt). The tagline is LIVE-DIE-REPEAT, and suddenly sitting there I realized --- this captures exactly why I’m so lukewarm about FHIR! (Seriously, that happened.)
In case you’ve missed it, FHIR is the shiny new thing in health interoperability circles. A lightweight, modular, composable, built-for-REST set of content standards to describe healthcare stuff. It’s clean, and there’s a lot to like about it. Why, then, do I find it so uninteresting?
Because we’ve been here before. At least a few times. And it’s always the same story, because standards folks only have one tool in their toolbox: make standards. Here’s what happens:
We never get to spend enough time on the adoption part to make it stick before somebody creates a new shiny thing and we all run after it. I really thought that Meaningful Use would change this … we got such great buy-in around Direct and the C/CDA, it’s being implemented, every day more and more patients are really getting their data, the BB+ Push pilots are working…. But it’s slow, and we’ve certainly run into challenges.
So, here comes the new standard to save the day. But even if this one is a little better technically --- and I’d wager it is --- that won’t matter. It’ll have its own set of scale adoption issues. What’s the service URL for each provider? How do you register each app with each provider’s OAuth endpoint? Are there policy requirements before somebody allows your app “in”? Anybody hear about something called Covert Redirection? What about push events; are we all going to be polling each other for new data all the time? And on, and on, and on.
The really sad thing is that I have to take some blame for this with Direct too. I thought we could break the cycle by including ideas like distributed “circles of trust” --- a system whereby once you’ve installed Direct, you can expand your trading partners over time without any technical changes. This is working, but not fast enough … the cycle of frustration with things like the Federal Bridge and arguments over identity proofing risk sending us right back to the starting line.
I don’t have an answer. And you know, if FHIR gains traction you can count on HealthVault to implement it --- our model fails without interoperability, so we have no choice but to play along. I just can’t force myself to get excited about the standards any more. As my good friend Umesh always says … old wine in new bottles.
All that said, the movie looks like a lot of fun.
It is very easy to sit on the sidelines and critique, if you really want to make it better, you need to get out and do.
Keith, I agree 100%, but that's not really the point I'm trying to make in this post. My concern is that the standards repeat cycle is covering up the real challenge, which is getting through the adoption cycle with *any* standard. And there's a mea culpa there too ... I thought I saw in Direct technology changes that would fundamentally alter the dynamic, and I'm not sure that was right.
But that said --- if your implication is that I'm sitting on the sidelines and pointing fingers, I think that's laughable on its face. I would claim that there are very few (by no means not zero, but few) who have committed more of their personal and professional time and capital to concrete actions here since 2009.
I don't usually feel comfortable spewing this stuff, but I want to be clear that I think I have earned the right to comment, so just a few: my personal code still runs in the core of Direct; I dedicated (for free) my most valuable HealthVault engineering resource for a year as the primary creator of the open source .NET reference implementation; I ran the Direct Security & Trust workgroup; I spend at least three hours every week reaching out 1:1 to EHRs where our C/CDAs don't interop to fix them; I personally wrote code that runs in hospitals to help EHRs share C/CDAs with patients when their vendors blew them off; I pro bono operate the technology and manual workflows supporting the BB+ trust bundles; I have publically commented, testified and run workshops in detail on all of these, not supported by any staff to do so. And as a day job I run the team here at Microsoft that delivers HealthVault to millions of users and thousands of development partners.
And if that wasn't your implication, sorry for the rant. ;)
Hope you like the movie when you see it. See www.healthintersections.com.au for a response
For those following along here, I added some more thoughts over on Grahame's blog at www.healthintersections.com.au.
Your comments are insightful, as always!
Some of us hope that one of FHIR's most interesting use-cases will not be in the creation of new national interchange platforms, but rather in the enabling of "plug-in apps" for EHR vendors who want to shift towards being a "platform" rather than a pre-packaged "solution."
We (Cerner) have started working closely with the the SMART on FHIR effort being coordinated out of Boston Children's. As you know, the SMART Platform initially used a non-standard RDF-based data access model, but they have recently switched to a FHIR-based model.
At HIMSS, we partnered with three other "container" backend data services and demonstrated a half-dozen SMART apps, plugged directly into the clinical workflow. Each of the SMART apps used (DSTU) FHIR to access necessary data from the backend service. FHIR enabled a dramatic reduction in the friction of negotiating a mutual API, and the ease of embedding a web app meant that the total work effort was measured in weeks, not months.
The key to replicating this elsewhere will probably come down to getting the vendor community to agree to a set of core FHIR Profiles. We plan to start with some simple, publicly defined core profiles (problem, medications, labs, etc.) and go from there. Each vendor can establish their own app certification and trust criteria, at least initially.
There are some details here: smartplatforms.org/smart-on-fhir
Thanks David ... this is an interesting twist. I'm a big fan of the SMART work but have to admit that I've fallen behind and wasn't really understanding its connection to FHIR --- that is a nice combination and shame on me for not paying more attention!
Unfortunately, many folks aren't as nuanced as Cerner. I have had multiple people in positions of unfortunate authority (ONC, etc.) tell me confidently that FHIR will solve the problems that Direct is having in the context of Meaningful Use. This I simply don't buy and it makes me sad.
So maybe you've shown me a path to do something constructive here --- if we can better evangelize the right tool for the right job ... then we may come out ahead. I'll give that a shot.
"I have had multiple people in positions of unfortunate authority tell me confidently that FHIR will solve the problems that Direct is having in the context of Meaningful Use"
It's lovely that they have confidence in what we're doing, but as you have pointed out, it's not so simple. I'd say that using FHIR is a pre-condition to solving some of their problems (though the question of Direct seems rather orthogonal here - it's a different kind of thing), but we both agree that just changing from one standard to another and doing things the same way will lead to a pretty similar outcome.
But what might lead to different outcomes is if the ethos of FHIR, and the manifesto that we work under, are contagious... we can only hope.