On Wed, 07 Jan 2004 20:48, Thomas Beale wrote:yes, well...!
I wouldn't know for most of these but it seems reasonable. My only
comment is that this classification is fine for a sort of maturity index
of software; things like openEHR have a lot of work in the specification
space, shared (pioneered) by OMG HDTF, CEN 13606, HL7 and others, which
You know that I am all for standards, and you are certainly right in many ways. However, the sad reality is that standards in our domain don't work. Very sad.
But in a world where you can't even get the some rogue but very influential countries to agree to most sensible standards with straightforward benefits such as adherence to the metric system, what hope is there for health care standards?
Corbamed is sensible and fairly complete in the sense that you can implement it here and now and do something useful with it. I know OpenEMED has implemented some, but honestly: hands up how many world wide installations are there of any Corbamed system? So much for Corbamed as a standard.probably true - despite the efforts of Dave and many others who worked on it. But let's not forget some very things it did, which still have great value today: it provided a very sensible separation of concerns of the domain, in which I still cannot find any error; it provided a lot of good design work which is being absorbed into more recent works, and it provided PIDS which is widely used both as a design inspiration and a concrete specification.
And HL7? It is still a sad joke. Despite it's (probably entirely unnecessary) complexity it still doesnn't fulfill any expectations. I haven't seen yet any two not-inhouse systems that can talk to each other HL7 without need for a home baken translation level. And even then things can go wrong (see the pathology download tragical comedy in Australia).HL7v3 is complex, no doubt about it. We will just have to see.
And CEN standards? I'd love to see them work in our domain, but please point me to any significant installations using them.There are quite a few installations using the current 13606, I doubt if a single one of them could move an EHR Extract to another site not running the same software. But - current revision work stands to change that. I have met many implementors, and I think they will get behind the comprehensive revision. Again we will have to see....
Standards that work nowadays and make everybodies life easier have arisen out of somebody actually "doing something", and the process of becoming a de-facto standard has been helped by either sheer commercial market domination or complete openness. Complex domain specific "standards" developed on the white board and then imposed onto humanity have not often worked well AFAIK.Agree totally; the mantra of the openEHR work is 'implementation, implementation, implementation' (professor David Ingram at University College London came up with this;-) - it's taking some time, sure, but the mentality of the project is more or less: create a workable version of best of breed of standards and other specifications, while feeding back the results into the standards bodies. Those bodies that listen will eventually have good standards, if there is enough weight of implementation experience.
- thomas beale
