Yesterday my colleague Dr. Bill Crounse wrote a great blog post about how --- perhaps --- we could accelerate progress in health data sharing by “starting over” with a patient-centric approach. This post won’t make much sense without reading that one first, so go ahead and do that now.
Dum dee dum, waiting… ok.
Bill’s story has received a number of reactions and comments, but one in particular from Doug Dietzman deserves attention, because it highlights a common but critical misunderstanding --- one that makes all the difference to why a personal, patient-controlled “HIE” is so much more likely to succeed in the long term.
Before I go any further, I should say that I’m actually a supporter of “B2B” health information exchange too. HIE organizations are doing important work, especially making technology and services available to “white space” providers and to patients themselves. As I often say, I am thankful for anybody and everybody that is truly trying to help.
The full text of Doug’s comment is at the bottom of Bill’s blog, but in short, he challenges how a personal HIE like HealthVault can support the business transactions of healthcare --- such as getting lab results back to the ordering physician, or alerting stakeholders when new information is available.
These things are *exactly* what HealthVault is built to do.
I wrote about this in detail about a year ago in the context of patient matching and identification. What you have to understand is that HealthVault is not just some database in the sky. It’s a neutral, universally accessible “hub” through which information can flow automatically based on patient consent.
Here’s just one example of this kind of workflow:
- Sean uses his family doctor’s patient portal to link his HealthVault record to the practice.
- Sean visits a specialist and requests a copy of his visit summary be sent to his HealthVault Direct address.
- The CCDA attached to that message is incorporated automatically into Sean’s HealthVault record.
- The family doctor’s patient portal receives a notification via HealthVault’s “eventing” interface that new information for Sean is available. It pulls that down into the EHR inbox and next morning, the family doctor sees the information and is up to speed on the case.
Note that in this model, at least three historical “HIE” problems go away:
- The two practices don’t need any data sharing agreement at all. The patient’s consent with each practice “brokers” policy issues between independent entities.
- The two practices didn’t need an EMPI or any patient matching technology. The patient creates links using the patient portal in the first case, and exchange of the Direct address in the second.
- There is a built-in mechanism for efficient notification between systems.
- Provenance models already in place support information integrity.
And of course the benefits compound. If the patient has linked their HealthVault record to other practices, they get the data too. If they use consumer-focused applications, they update too. First responder systems like NOKR get the latest information. Coordinating care even between countries? No problem --- with none of the technologies having a clue about the others. It’s pretty awesome.
Now, of course, nothing does everything and there are certainly transactions better suited to direct business-to-business connectivity (hooray that Direct does both!) … but for the core HIE problem statement --- getting accurate, up-to-date information in front of the right folks at the right time --- it’s a slam dunk.
And the good news is, slowly but surely, it is happening.
Image may be NSFW.Clik here to view.
