gnumed-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Gnumed-devel] Paper on OpenSDE (structured data entry) just publish


From: Richard Terry
Subject: Re: [Gnumed-devel] Paper on OpenSDE (structured data entry) just published
Date: Wed, 19 Jul 2006 08:45:10 +1000
User-agent: KMail/1.9

I don't think this is a terribly bright piece of software and would be 
particularly useless in clinical general  practice. Perhaps ok for 
re-searchers not living in the real world with time to burn. 

Once again it is written/thought out by people who seem to believe that  by 
having trees/lots of checkboxes etc, you can cover every aspect of medical 
records.  

It's not intelligent.

Real people (US) have MOST of the questions in our brain. We quickly write 
down what we want ot know in front of us. Where we fall down of course is 
WHAT WE FORGET TO ASK. This is where an intelligent system would function. 
The doctor inputs his stuff, and it parses the result.

IE What would be intelligent was an automated method of data entry (such as 
the gift I gave gnuMed which hasn't been used) combined with an intelligent 
decision support.

This may interest the thoughtful - you may remember that I took place in the 
HUDGP software evaluation last year over a number of months.

 I didn't particularly like the method they used (Companies presented the 
software according to a script which they had to fulfil - to in theory 
demonstrate a basic set of functionality that they (the hudgp IT Dept) 
considered should be present in software. The scoring order wasProfile 
(Intrahealth), Best Pratice (Frank Pyfinch revisited), MDW and Medtech. 
However of interest no product could really be recommended as all had 
differing features.

Anway, the last week I've installed Win2000 on an 8Gig virtual machine on my 
Arch LInux Box (using Parallels - which I paid for - www.parallels.com - 
worth every cent of the %US$59 I paid for it).

I've installed Profile/Medtech (which BTW arrogantly overwrites any existing 
interbase installation)Best Practice and MDW. I've opened them all up and 
once and tried doing the same things in each of them to duplicate the task to 
see which I'd use if forced do do so.

So far my impression have been ( though I hate to admit it as everyone on the 
list knows my passionate dislike for it) that if forced into a corner 
tomorrow to use a program just to get my job done in a reasonable time with 
reasonable help - I'd use - wait for it, MDW!!!. Stupid that it is   - it 
does the job - most GP things are there somewhere, albeit in a clunky clumsy 
manner. Better Practice - 'an evolution', somehow doesn't hang together for 
me - harder to navigate around - slower to use, less features. Seems slow to 
access the database as well. Medtech - a combination of some good thoughts 
with really cretinous features - I suspect that at the end of the day it is 
tailorable to a much greater degree than would appear on the surface however. 
Malcolm Ireland who uses this day seems to find it quite managable and i 
respect his opinion - though he admits that form writing if bad and past 
history is bad.

Profile is interesting, and I'm still drawn to it above all the others.

It is the only program where the developers have put any real thought into 
things. They undoubtedly have the best progress note presentation/entry, 
semi-intelligent parsing of the input stream. Their script writing while 
failing miserably to pick up allergies or to present warnings re 
interactions, is the quickest and most intelligent of any of the commercial 
programs. Their method of allocating coding is really clumsy, though 
thorough, their past history is passable, their form generation is abysmally 
bad.  Overall it feels solid in its construction, is tailorable to the nth 
degree. It is the only program I feel could competently run the whole of 
Newcastle as a single huge amalgamated general pracitce with a central data 
repositoary. 

As a company despite their low prescence in Australia, they seem aware of 
their shortcomings and keen to do something about them. Their track record 
anecdotely in Australia is not good in terms of support (However I their 1800 
number and it connected through to the help desk in New Zealand where they 
were friendly). Only one practice in Newcastle uses their product and I spent 
several hours one night at home with the GP - despite him using it for 
several years I noticed he couldn't easily navigate around of know where 
things were - and he regards himself as an IT guru!!

Regards

Richard

On Wednesday 19 July 2006 07:24, Tim Churches wrote:
> Syan Tan wrote:
> > where can I get one ?
> >
> > Looks like a nice realization of clinical textbook stuff .
>
> The project home page has links to SourceForge downloads. Looks like the
> demo is Windows-only but it is the ideas which matter. I am not
> suggesting that their ideas are necessarily good or bad (I haven't even
> read the paper), just that here is a group of people who have thought
> deeply about the same issues which seem to bedevil GNUmed, and have
> produced some slick working prototypes to test and refine their ideas in
> practice.
>
> http://www2.eur.nl/fgg/mi/OpenSDE/
>
> Tim C
>
> > *On Tue Jul 18 6:56 , Tim Churches sent:
> >     Haven't read it myself but might be of interest to GNUmed developers
> > - see http://www.biomedcentral.com/1472-6947/6/29/abstract
> >
> >     Tim C
>
> _______________________________________________
> Gnumed-devel mailing list
> address@hidden
> http://lists.gnu.org/mailman/listinfo/gnumed-devel




reply via email to

[Prev in Thread] Current Thread [Next in Thread]