Tue, 5 Jun 2001 09:20:25 +0100

From a thread on the Fam-Med list on "Exit Strategy for EMRs"

There is quite a lot of sense in Kim Avant's (NextGen) thoughts, I'll
pick at a few of them though:-

Weakness of Escrow

Escrow of source code has worked to the benefit of a health
organisation exactly once that I know of in the UK.

Companies do not usually go out of business, rather, they fail, and
after a protracted period of unsatisfactory service (which doesn't
trigger escrow as usually written - now there is an idea...) they are
bought for £1 and other valuable considerations, or there is an
exchange of shares (which does not trigger the escrow) by a competitor
with a prduct of their own buying market share.  Dumb idea IMHO, they
acquire unhappy customers most of whom they then drive to do something
they don't want to do.

Those that are bought as good going concerns at about £3000 per using
practice may be bought by companies that don't already have a system
in which case life is less hard for the users - but if a company buys
one EMR vendor, they are likely to buy another, and then they have two
systems to develop, splitting their interests and disorganising the
support staff etc.

Who Exports                               midgley@mednetics.org
----------------
Remember that it won't be the _original company_ that exports and
translates the data if you move.  It will be the new people, possibly



    
using a specialist agent.

In the UK there is GP Data, which started as the developers from  a
system taken over by another firm, and seem to be fairly good.

I did a complete export from the old Clipper system we had, it is a
non-trivial task, and best done by people who really know what the
data in there means, I think, hence the time I took to do it.

ODBC and .DOC
----------------------
Up to a point.  I don't support the use of .doc for anything, Word
works adequately well with .rtf (Rich Text Format) as its standard
file format, and this allows interchange of highly formatted documents
without having to persuade everyone you deal with to buy the latest
and biggest version of Word.

Having a documented API to the database is a huge benefit to users, as
evidenced by Scotland's national GPASS system, and even more so by the
Premiere system that the Torex company in the UK is de-emphasising.
Users frustrated by the slowness, absence or incorrectness of
developments by th company have written their own additions.  In due
course there may be a core database replacement project among the
users.

Standards
-------------
SNOMED - Read is touted as a standard thesaurus for coding medical
stuff, I did some development in Read version 3.1 but I think it is
clear there is something wrong with the map of product to world.
Several years after it became available the bulk of GPs in the UK are
now seeing the point of it, even if all levels of the heatlh service
admin seem to immediately grasp the wrong end of the stick
    (we must make everyone use the same code
    for the same thing so we can search for that code...
    No, if people use different codes for what you consider
    to be the same thing, you may well be wrong about its
    homogeneity, and regardlss, what you do is search
    on all of those using the AND word in SQL...)

If other people are using ICD and IPCP and so on, then systems need to
handle incoming codes and phrases in those, or in any other code.  Not
impossible, but it goes against the use of code systems as
differentiating features and as ways to lock in users.

Open Source is the best way to go, but that was a very reasonable reply
from a vendor of a closed source system.

--
Midgley

John Faughan's papers on exit strategy Well worth reading.