-----Oorspronkelijk bericht-----
Van: Tim Benson [mailto:[EMAIL PROTECTED]]
Verzonden: Sunday, April 01, 2001 9:12 PM
Aan: [EMAIL PROTECTED]; Open Health list - Minoru
Onderwerp: RE: donkeys, was RE: Ars Longa IT Brevis
I agree that small local clinical projects tend to work and large ones
(directed by management) fail, but my conjecture is that this is a direct
consequence of system architecture. Few of us are clever enough to
architect systems which are able to please two masters with very different
agendas. The problem is not new.
The Good Book put it well "Render therefore unto Caesar the things which are
Caesar's; and unto God the things that are God's" Matthew XXII 21. For
Caesar read Regulators and Payors and other people with sticks (not to
mention swords or guns) - regulations change fast and follow fashion
mindlessly. Medicine deals with life itself, which only changes at the
speed of evolution (although stupidities like the irresponsible prescribing
of anti-biotics can give evolution a big kick).
The way forward is to insist that management information should be derived
from clinical information (used at patient-side) as a completely different
task, done later, and by an automated open source peer review algorithm.
The argument for this type of approach has been well made in the final
couple of chapters of Slee, Slee and Schmidt "The Endangered Medical Record"
Tringa Press, 2000.
Tim Benson
PS My question to Adrian's final question, which I think refers to
computerate clinicians, is to test whether they have the faintest notion of
information theory by asking them the difference between a coding scheme and
a classification. If they imagine them to be synonyms, beware.
> -----Original Message-----
> From: Dr Adrian Midgley [mailto:[EMAIL PROTECTED]]
> Sent: 1 April 2001 16:05
> To: Open Health list - Minoru
> Subject: re: donkeys, was RE: Ars Longa IT Brevis
>
>
> Chasing a dizzying stream of threadname changes, Tim Benson said,
> wisely and truly:
>
> >the drivers and the donkeys DO have a different agenda.
> >the NHS defined a hospital information system as:
> > In practice the management data was provided (not accurately) and
> >the clinicians thought they had been conned yet again.
> <snipped heavily>
>
> > the systems analysts did not understand that the management
> data (eg ICD10)
> >was useless for clinical management of patients.
> >..etc
>
> Midgley's observation is that small local projects aimed at wants
> expressed by clinicians tend to work, and their functions
> can grow over time, and their limitations are accepted by the
> users who were involved in selecting the compromises
> required.
>
> Midgley's theory is that if we consider the depth and breadth of
> knowledge and ability in IT matters and implementation (IE
> at the "show us some code" level, and in the usability
> recognition field, not at the chairing committees end, then we find that
> although there are not many clinicans who know more about all of
> IT than most IT managers and developers in a given
> health service, there are a few clinicians who know rather a lot
> about IT, and a lot of clinicians who know somewhat about
> some area, and are willing to study and take advice on applying
> their domain knowledge to getting a tool that they want
> going... and that the area under the curve - the product of
> expertise times numbers - means that the bulk of IT expertise is in
> the clinicians in any given health service or group of
> institutions in an area, and not in the IT professionals.
>
> However, in the NHS, the planning is done on the assumption that
> only the IT professionals can specify or work upon the
> solutions.
> Like Tim I simplify a little here, many general practices have a
> partner or who hacks scripts for their proprietary systems,
> albeit with no access to the real programming unless they feel
> like reverse engineering binaries.
>
> The challenge to the managerial end of culture in healthcare IT,
> as well as to clinicians, is to devise ways in which the
> broad but gappy and not uniformly deep expertise and experience
> in the clinicians can be used to develope in th smae
> direction as the whole org is going... and the depth and focus of
> the professional IT corps can be used to ensure that critical
> sections are got right, and that there is commuication where it
> is needed between various programs and bodies of data.
>
> An example of the latter is perhaps the need for effective
> clinical coding, of which ICD10 is not an example, and effective
> management coding, of which I gather a map of Read/SNOMED into
> ICD10 may be an effective example.
>
> In my local hospital I am told the IT dept and management are
> developing an EPR in house, while my consultant
> colleagues have been know to write such things as "B&Q (American
> equiv Wallmart?) have a better system for tracking
> cans of paint than we have for looking after patients."
>
> It seems to me that one of the possible enablers for such
> alignment of efforts would be if the source code was available for
> inspection by anyone who wished to do so, since they could then
> become familiar with the areas that affected them, and if
> they found or made improvements these could be aded in.
>
> One other factor which affects the area under the curve, adding
> another dimension to the plot, is that permanet staff
> clinicians are in post for 30 years, whcih is rarely the case for
> IT staff at either the programming or administrating ends of the
> spectrum. A little bit now and then in a system whcih you have
> seen and helped evolve over that long can add up to a
> great deal, whereas someone who blows in, gets up to speed, and
> blows out int he space of a mere handful of years may
> reaosnably think their role is to decide which company to buy an
> off the shelf solution from, and not see any great difference
> in the workload required to come up to speed with a new program
> from that required to add a function or two to an old one.
> And be right, for them selves.
>
> Enough for now.
> Except...
>
> Tim said:
>
> >I have a pet question for such analysts.
> >If they can explain what is the difference between what a nurse does
> >and a doctor does, then there is some hope.
> I tend to agree this is vital.
>
> What corresponding question would the panel devise to put to a
> clinician contmeplating making an addition to a module of
> a program which deals with their workflow or information needs
> several times a day?
>
> --
> Adrian Midgley
> Exeter
> http://www.swis.net/midgley/
>
>
**********************************************************************
This email and any files transmitted with it are confidential and
intended solely for the use of the individual or entity to whom they
are addressed. If you have received this email in error please notify
the system manager.
This footnote also confirms that this email message has been swept by
MIMEsweeper for the presence of computer viruses.
**********************************************************************
