Convert your FHIR JSON -> XML and back here. The CDA Book is sometimes listed for Kindle here and it is also SHIPPING from Amazon! See here for Errata.

Wednesday, May 16, 2012

HL7 mHealth Workgroup Meeting

The newly forming HL7 Mobile Health Workgroup met this afternoon in Vancouver to discuss the workgroup charter.  I went to the meeting today, and there's another one tomorrow.  I won't be able to attend tomorrow because I'm teaching about Templated CDA.

What follows are a few high points of the discussion.  There's probably a lot more discussion that the workgroup needs to have before they are able to develop a charter and start getting to work.

One of the challenges that the group faces is defining the scope of their activities.  The discussion led me to tweet about the distinction between healthcare devices that are mobile, and mobile devices that are used for healthcare.  From my own perspective, healthcare devices that are mobile is not where the challenge lies.  Instead, it is the mobile devices that are being used to support Healthcare.  To explain one problem, I whipped out my iPad, showed my blood pressure results in one app, and then my weight in another.  Those two chunks of data are so much more powerful together, but even though they reside on the same device, I have to do some serious hacking to make the data appear in one place.

We have hundreds of thousands of apps, but no mobile apps that can put the data together from these separate mobile apps in meaningful ways (except in proprietary clouds).  We have the ability to put all that data into the cloud, but I'm still stuck with the device maker's web interface to access and combine the data in meaningful ways.  There out to be an app for that, but the problem is that there are no standards that enable that capability.  And the cloud isn't helpful either if everyone has to have their own cloud, and the various clouds don't talk to each other.  After all, a cloud that cannot interact with other clouds is nothing more than a fancy way to put a silo in the sky.

Someone asked what the distinction is between mobile and distributed.  Again, from my perspective, mobile means it moves with me.  Distributed can be a part of mobile, especially when the device interacts with the cloud, but in my case, there's no cloud.  All the data is on my device.  But distributed can be "bigger iron" with more capabilities than the mobile devices have.

Someone pointed out that mobile devices are getting more capable, with better technology stacks, et cetera.  But, we cannot simply wait until they have the world's best technology stack before we address the need for standards.  If we do, someone else will have already solved the interoperability problem, and that is part of HL7's mission.  If we fail to solve it, we fail ourselves and our industry.  It was well put by one observer who said:  We need to foster a community of exchange and accessibility.

In order to define what the scope of the Mobile Health Workgroup is, we need to understand what the scope of mobile devices is.  We have devices with simple bluetooth, GPS and simple accellerometer capabilities, more complex apps that fit into a smart phone platform, or connect to it (e.g., to capture blood sugar results), and other apps that work in a tablet platform.  Each of these devices has certain capabilities.  Ideally, we'd find a taxonomy of mobile device types and related capabilities that could help us understand the space.

One of the issues that the workgroup discussed was the issue of security.  Mobile devices have a much different risk profile than other systems.  They are easier (and more desirable) to steal, harder to protect, et cetera.  It's much easier to lock down and manage laptops and servers than it is to deal with mobile devices, as John Halamka points out here.  Someone suggested that the Mobile Health Workgroup should perform an assessment using John Moehrke's Risk Assessment white paper.  John and I will both point out that was a collaborative effort, and that a lot of the content came from a Canadian Ad Hoc Harley winner.  It is very nice to see John's efforts to promote that effort pay off.

Once we have found (or created if really necessary) the taxonomy of devices and capabilities, we can perform a risk assessment that will help inform future mHealth efforts.

There's still quite a bit of forming and storming that needs to occur.  I don't expect a charter by the end of this week, but certainly before the next Working Group Meeting.  This is another place I'll need to pay attention.


  1. I've heard of some EHR vendors developing mobile strategies of their own to be able to extend their applications on to devices like the ipad, android based tablets etc. How can HL7 govern the standards for data interchange between mobile devices and existing EHR applications when there are a slew of proprietary data formats already being developed? Or is this a different problem the group is trying to solve.

  2. What caught me off guard in that meeting was the discussion of cloud. Okay, mobile makes less sense without the context of cloud and vice versa; I understand that. But a mobile device, completely separated from a service provider, is still a platform containing numerous low-cost applications that should be able to communicate with each other using standardized events/intents/broadcasts using JSON or whatever data transport is supported. Regardless of the "cloud".

    But is there a business case for two small, unrelated mobile applications augmenting each other locally through (more or less unplanned) standardized communication of data? Or is the value really in the cloud not the device?