But why couldnt the practice identify its own patients?
I dont see the need to take data out of the practice, analyse it, and
then feed it back. What is the purpose? It starts to sound like a system
of GP behaviour control - this is what concerns me.
Why not use the data for public health and research, rather than
behaviour management and cost control?
The value of a large dataset will be in data mining for "nonintuitive"
correlations - ie those that are not obvious.
It could potentially provide a post marketing surveillance system for
adverse drug reactions for example. No such system exists now - eg
connect hospital admission diagnoses to drug prescribing
R
Greg Twyford wrote:
Andrew Patterson wrote:
Once again, I think this discussion about de-identified data needs
to be
watched closely. in many respcets it can't be useful to GPs if it is
de-identified, including to source, and even if the GP is identified,
he'll want to know who the problem patients are. The Adelaide West
atlas
doesn't use de-identified data.
Greg, as I understand it, the data that the Atlas uses for its
funky graphs and maps is definately de-identified (from
a patient perspective).
Obviously the source is not de-identified - but that is
because each Atlas is done for a particular GP practice and
hence there is no doubt as to where the data came from!
Andrew,
So if the GP wanted to identify his non-compliant diabetics as a
group, the atlas wouldn't help? I must confess that I thought, based
on the published information about the atlas.
The following is an extract from information about it, which suggests
the sort of thing I've had in mind:
?Identifying diabetic patient cohorts (+/-co-morbidities)?
Exploring how a practice identifies and manages these patients
ie. Diagnosis coding, disease registers, outcome indicators
A GP can't search MD for patients with HbA1C > 7 and other relevant
measures. This is one thing we do with the identified data we hold in
Argus.
As I said, too, I feel that identifying data by GP and sending it to
Canberra won't be acceptable to many GPs. Adelaide Western hasn't
taken the data beyond themselves.
Greg
_______________________________________________
Gpcg_talk mailing list
[email protected]
http://ozdocit.org/cgi-bin/mailman/listinfo/gpcg_talk
