On Fri, 7 Feb 2003, Robert Stark wrote:
...
> From what I have studied here, forms maybe a whole lot simplier than
> archetypes, but archtypes are more powerful than forms which is
> another reason why this takes time.

Rob,

Just so that we don't bore the list by entirely repeating what was already
said, I like to reference a discussion between Thomas Beale and myself on
this exact same topic in Sept 2000. The subject line was "GEHR/archetypes
vs. OIO/forms" and started with this message:
http://www.mail-archive.com/[email protected]/msg02261.html
You can follow the link at the bottom of the page to read Thomas'
response.

You are correct that archetypes (per specification) is more powerful than
the current OIO forms, as implemented. However, there is no reason why OIO
(or any other Turing-machine-equivalent systems) cannot match the modeling
power of OpenEHR. The design challenge for EHR (and any software system at
this time) is not "power" - but ease-of-use and related "perception of
simplicity".

Forms appear simpler than archetypes - because we want it to appear simple
for the end users. :-) Hopefully, future OpenEHR implementation will also
make archetypes appear simple.

The implication of this "design goal" is that the OIO project approaches
documentation and implementation from the point-of-view of end users. I
believe this stems from difference in target audience for OpenEHR vs.
OIO. Accordingly, we avoid programmer lingo like "DV_DATE_TIME" and
"LOCATABLE" etc used extensively in OpenEHR. We implement OIO as a
software system that is immediately usable to illustrate the types of
tools that clinicians and researchers may find useful.

> I would like you to compare the two in terms of how forms describe
> data structures such as lists, table series, tree structures,
> datatypes, etc.

Complex structures like lists, table series, and tree structures are
modeled as multiple forms in the OIO system. Currently, these forms are
linked together through objects such as patients, workflows, reports,
folders, and schedules. Each patient may be described by multiple forms,
workflows, schedules, etc. Each workflow can contain multiple steps that
requires the completion of forms. Each report can contain information
extracted from multiple forms and patients. Folders can contain multiple
workflows, forms, schedules, etc. We are also beginning to implement
form-to-form links which may eventually include hierarchical tree
structure consisting of forms (at each node).

Please let me know if you like to discuss further complex structures and
strategies for modeling them. I think it will be quite informative to
examine how OpenEHR approaches modeling workflow (e.g. "DV_STATE" Class =
state machine) and how it relates to the rest of the OpenEHR in
comparison with OIO.

> Hope to hear from you soon

Thanks for raising these important issues! I look forward to discussing
them in greater detail. I think this type of discussion is the only way
for us to really understand OpenEHR and other systems like the OIO! No
amount of excellent documentation can fully replace human interaction.

Best regards,

Andrew
---
Andrew P. Ho, M.D.
OIO: Open Infrastructure for Outcomes
www.TxOutcome.Org


Reply via email to