National Projects and Standards

It’s something you can see all around the world: governments sponsoring large national healthcare projects of one form or another (EHRs, prescription systems, HIEs, etc), and the bodies running these projects getting very involved with international healthcare standards bodies (HL7, IHTSDO, IHE, etc) (yes, I know IHE isn’t a standards body. but everyone knows what I mean). I’m referring to ONC, Infoway, Connecting for Health, NEHTA, etc (btw, declaration: I’ve worked for nearly all of these – or still do – in their standards programs).

There’s a difference between the goals of the national project, and the value proposition of using standards, and this difference can create considerable tension.

Projects

These national programs are generally constituted by elected politicians who commit large sums of money to big goals that are difficult to achieve, and quite risky politically. In fact, these projects only happen because there’s such a huge pressure on the national programs in terms of getting more for less, and these projects appear to offer the prospect of delivering that – and they *will*, if they succeed. But these projects are difficult at every level – hard to make change, technically demanding, and at the limit of our knowledge of informatics, and how to deliver computing support in really well integrated ways to a wetware (very wet) dominated process.

So there’s real risks, and because of election cycles, short time lines run by risk averse sponsors.These projects have to succeed, and have to stick to their timelines. (Which does make me wonder, where do they make these timelines up from?)

Standards

The Standards process, on the other hand, doesn’t work like that. It’s a slow, consensus based process which emphasizes getting agreement to a common position, and voluntary participation from the community with gradual buy in. That’s it’s greatest strength. It’s not going to run out and transform a community prospectively. But gradually, incrementally, and surely, the presence of the standards transforms the community and empowers it. However you can’t rush the process – putting a timeline on it, or throwing money at the volunteers – that is a high risk option. The real workers, the ones who can make a difference, because they know how to write a standard that people can make work – they’re busy, making the standards work. If you put a timeline on it, quality goes straight out the window. If you throw money at the process, the ones who are better at winning contracts rather than writing standards will take over. It’s not that timelines or money can’t accelerate the process – but both are high risk options that need careful management and strong community buy in before they can be successful.

Note, I’m not arguing that the standards process should be strictly linear, but resetting the direction and the community is also a high risk option.

Tension

Why then, would you mix these two? Put like that, using standards for a national project sounds like a sure recipe for a disaster. So why do national project offices even try to use standards? Why not just be XML Cowboys, and do everything in a bespoke way?

Well:

  • using international standards means that international vendors won’t charge an arm and a leg to play ball (probably just an arm…)
  • using standards means that the local community can rely on some change control over the process, protecting them from the whims of the sponsors, and electoral change etc. This breaks down to $$$ too
  • using standards means that the national program can leverage the existing skill base of the community without destroying the community while doing so
  • using standards means that the project is constrained to behaving more along the lines of international consensus

But these goals are mostly long term outcomes. You can only really achieve them if the sponsors are prepared to give something up – their own custom requirements and most of all, tight timelines. Because standards means giving something up in one place to gain in another. If they aren’t prepared to trade these things away (somewhat, anyway) for these other goals, they won’t achieve these outcomes from the standards process. They won’t get the meaningful and deep community engagement they need, and they won’t produce specifications that align with other countries  – and they will often wonder why the cost-benefit of standards is so low. And their engagement with standards will be… fraught with confusion and misunderstandings.

But it can work (well, even) – if the projects and their sponsors can live within their limitations.

 

5 Comments

  1. Rene Spronk says:

    > They won’t get the meaningful and deep community engagement they need, and they won’t produce specifications that align with other countries

    You haven’t really made the case for the first half of that sentence. Why is that desirable/necessary ?

    Can’t really push the volunteer process. One can fully control the localization aspect. Localization rarely means dealing with NEW requirements that haven’t been encountered before at the international level. As such a project’s involvement with a standards organization is fairly limited.

    > But it can work (well, even) – if the projects and their sponsors can live within their limitations.

    Last 2 sections of the post (conclusion) may neet rephrasing. I’ve read it a couple of times now and I’m not entirely sure what you’re getting at. Yes sure: ‘short timelines and being top-down-directive don’t fit well with standardization communities’. That’s a statement of fact, and not new to most readers. Where is the sweet-spot in cooperating together?

    • Grahame Grieve says:

      my experience is that localization – in the context of national programs – is often dealing with new requirements. That’s true both in and outside of Australia.

      For the first half of the sentence, which bit is not established – that they won’t get it, or that they need it? Note sure what else to say.

      As for the sweet spot – I have no good ideas because it seems to me it’s up to the politicians.

  2. Lloyd McKenzie says:

    Another of the reasons for relying on SDOs is that at least some of the national projects are envisioned as having a short existence. They need someone to carry forward the work after they’re gone.

    In some cases there’s also a perception that standards mean interoperability, and global standards must mean global interoperability, which seems like a good thing. Though when you dig, the idea of interoperating across borders is rarely, if ever, on the table.

  3. Michael Osborne says:

    At HL7 in January 2011, one of the speakers was from US Marines, a doctor and Health Informatics Director. He spoke of the need for cross border communications of health summaries , particularly when there are so many multi-national conflicts…think Afganistan, Iraq…When an Australian is wounded and being medivacd out of a country, quite often medical information is written on the patient himself. I see this as a great use case for shared electronic health summaries – and this requires international standards.

  4. Thomas Beale says:

    Your points 2 & 3 seem questionable. The de jure standards bodies are largely maintenance-free zones – all you can rely on with ISO and CEN is a 3-year wait. There is no meaningful way to report errors. OMG is better both in timeline (18 month wait) and there is a token attempt at collecting issue reports. HL7… well at least it is continually active I guess (although still no way to report issues other than travelling to WGMs), but far too much of the output is speculative. From an R&D perspective I have no problem, but from a standards perspective, it doesn’t really work.

    Point 3 seems to imply that international standards somehow align with the inbuilt technical capabilities of local industry / specialists, but this is almost never the case – 90% of industry effort is concerned with maintaining completely private commercial product architectures or home-grown public infrastructure.

    My own long term conclusion is that the only thing that really matters is working de facto standards, with an organically grown community and implementation-based evidence of efficacy. These can be turned into de jure standards when the time is right. Speculative R&D inside SDOs has no future.

Leave a Reply

Your email address will not be published. Required fields are marked *

question razz sad evil exclaim smile redface biggrin surprised eek confused cool lol mad twisted rolleyes wink idea arrow neutral cry mrgreen

*

%d bloggers like this: