Tuesday, March 19, 2013

Now Here Is A Blog I Really Agree With - I Have Been Saying Something Like This For Ages.

This appeared a few days ago.

Question: FHIR and un-semantic interoperability

Posted on March 12, 2013 by Grahame Grieve
 I did not understand the blog post about un-semantic interoperability.  Can you elaborate?  Will FHIR provide any of this un-semantic interoperability?
Well, the original post on unsemantic interoperability is just pointing out that many people mis-understand the nature of what semantic interoperability is trying to achieve:
We’ve had semantic interoperability in healthcare since we started having healthcare. Since the beginning of healthcare (by whatever definition you can use), healthcare practitioners have exchanged data using spoken and written words, and the semantic meaning has been clear (well, as clear as it can be given that human knowledge is limited).
So whatever it is that we are doing, it’s not introducing semantic interoperability. In fact, what we are doing is introducing a new player into the mix: computers. And not, in actual fact, computers, but the notion that there is something to be gained by processing healthcare information by persons or devices who don’t properly understand it. So, in fact, what we are actually doing is seeking for unsemantic interoperability.
It’s a matter of perspective. Perhaps, one day, we’ll really be working on true semantic inteoperability. But right now, we can afford to chase a lesser goal, which is exchanging data that can be used usefully in some limited pre-ordained ways.
More here:
The second comment - as italicised - is a ripper.
Karim Keshavjee
Unfortunately in all the discussion about semantic and/or non-semantic interoperability, we keep forgetting the purpose of achieving them. If we really took time to understand the purpose of interoperability, we’d have achieved a lot more than we have to date.
Sometimes I think we’re pursuing a Holy Grail when a plain chalice would do fine, or even a cup.
Essentially what is being recognised here is that complexity is something that takes so long to address that we need to approach solutions incrementally and not in a big bang.
Electronic semantic interoperability is really as hard and complex as it gets and doing what we can to while we address the possible is really smart.
Here are a couple of blogs where I and others have explored this especially the complexity issue.
Here are a couple of  more detailed discussions of some of the issues.
See here:
and here:
and here:
and here:
and here:
The bottom line here is that we should have work what can be made to work while continuing to work on improvement and better coverage.


Grahame Grieve said...

thanks David. If we're going with the chalice metaphor, then FHIR is just a plain cup - as plain and functional as we can make it - something cheap that everyone can drink from without any special holiness on their part. Because interoperability isn't the point, it's only a means to an end.

But if people really do want a holy grail, they can build on top of FHIR, as much as they want.

Anonymous said...

So doesn't this mean we should forget about FHIR, SNOMED etc etc & just concentrate on HL7 V2 as it is?

Dr David More MB PhD FACHI said...

No, not at all. As I see it we make sure we are fully using what is known to work and then make incremental improvements from that secure platform.

It is the attempts to deploy approaches that are not proven and evidence based and are overly ambitious that get us into trouble.


Andrew McIntyre said...

The question of the huge success of V2 does need consideration however. FHIR uses some of the same principles as V2, but uses a format that is more friendly to eg web developers. We do need to see if we can build on V2 while we see where FHIR is going and using both together is not impossible. The failure of V3 and the move away from "Big SOAP" suggests that V2 may have been on the right track all along and its not going away anytime soon. Its time to look at what V2 can deliver us in a short time frame and watch FHIR as its promising, but is not a complete solution and not mature enough as yet to abandon whats already working. Integrating SNOMED and more semantics into V2 is a good direction to take now.

Anonymous said...

What do you mean by - The failure of V3?

Isn't that what NEHTA is promoting?

Michael Legg said...

While I am all for the pragmatic and see the fantastic work that Graeme is doing as very positive, in that it is probably the way forward, I agree with Andrew that there is much to be gained by standardising where we can using v2 and then moving later (the target shooters approach - gather the shots together then move to the bullseye). Graeme himself I think estimates 5 years before FIHR is ready for prime-time (I am sure he will correct me if I am wrong).

I wouldn't like the argument about whether it is semantic or unsemantic interoperability to be a defence for woolly thinking. The assumption that when humans communicate using natural language that there is always shared understanding of meaning is a wrong assumption. We have, for example, found significant improvement in both the primary work and its communication with structured reporting of cancer from laboratories before it has got near a computer.

Grahame Grieve said...

For the failure of v3, see http://www.healthintersections.com.au/?p=476 and http://www.healthintersections.com.au/?p=482.

CDA - which is what NEHTA are using - is the one part of v3 that is getting traction. NEHTA didn't choose it lightly. You can find the evaluation here: http://www.nehta.gov.au/component/docman/doc_download/252-standards-for-e-health-interoperability-v10. It stands up relatively well to the passage of time.

I'm directly involved in the PCEHR/CDA implementation process. I do not believe that v2 would be a better choice for most of what is being done with CDA - the advantage of being well known would be far exceeded by the cost of having to build fragile edifices on top of it.

As for when FHIR will be ready, this is not a simple question. Already, many implementers are choosing to start new projects with FHIR in spite of the fact that it is not yet balloted at all, because they find it's advantages over the alternatives compelling. I expect to see production middle-ware and
systems early next year.

Then there's a period of evaluation and development where standards adopters have to decide which way to jump. The most conservative adopters would be looking at something like a 5 year time frame, like Michael says. But many will jump earlier. I think it will depend on the domain, the existing vendors, trading partnerships etc.

In the meantime, I do think that the existing HL7 v2 implementations are just awful and we could certainly benefit in the short term if we invested in improving the quality and sophistication of the v2 interfaces. MSIA showed some interest in this a couple of years ago, but that work has gone nowhere.

Finally, Michael is quite right about un-semantic interoperability. I wrote the original piece as a contrarian opinion to the almost fetish-like addiction to "semantic interoperability" that manifests in several quarters (still, and not just HL7). It's a matter of perspective, and of not letting one leg of centipede grow so big that it can't move at all.

Anonymous said...

I just love reading these sorts of discussions. Great to get some insights and opinions from -Australia's experts.