Are you aware of any archetypes for thyroid related data?

Clinical (archive)
1

Dear all,
In the context of my PhD work at UCL, I'll be needing some archetypes
to model the clinical data which I'll be using. If by any chance. I
can get my hands on some existing archetypes which are likely to
contain the domain information that I'm working on, that'd help a lot.
At the moment I'm working on some thyroid data, and in worst case I'll
be creating a dummy archetype which would probably be a nightmare from
an experience modeller's point of view.
So if you know any existing archetypes (maybe in CKM?) which may be
relevant for the type of data I'm working on, it would be great to
hear about them. The following links point to descriptions of data
sets, which would provide an overview of the type of clinical data I'm
using at the moment.

http://archive.ics.uci.edu/ml/machine-learning-databases/thyroid-disease/allhypo.names
http://archive.ics.uci.edu/ml/machine-learning-databases/thyroid-disease/thyroid0387.names
http://archive.ics.uci.edu/ml/machine-learning-databases/thyroid-disease/new-thyroid.names
http://archive.ics.uci.edu/ml/machine-learning-databases/thyroid-disease/allrep.names

I'm likely to repeat this question based on other clinical data, but
for the moment I have not given up on using this data set.

Best Regards
Seref Arikan

2

Hi Seref,

What you have here is a classic example of questionnaire-based data
collection, which made sense in a paper-base world but is totally
out-of-whack if one is trying to retreive/pre-populate the data from
an EHR. It is exactly the issues we discussed about DV_BOOLEAN vs.
DV_CODED_TEXT

If you take the first example

on thyroxine: f, t.

EHR:: Query on ACTION.medication + Drug term

query on thyroxine: f, t.

on antithyroid medication: f, t.
EHR:: Query on ACTION.medication

sick: f, t.
Tricky !! Can only be recorded by the human investigator

pregnant: f, t.
Tricky !! Can only be recorded by the human investigator . Notoriously
tricky to keep track of a patient's pregnancy state

thyroid surgery: f, t.

Query on ACTION.procedure + Snomed term

I131 treatment: f, t.
Query on ACTION.procedure + Snomed term

query hypothyroid: f, t.
Query on EVALUATION.problem + Snomed term

query hyperthyroid: f, t.
Query on EVALUATION.problem + Snomed term

lithium: f, t.
Query on EVALUATION.problem + Snomed term

goitre: f, t.
Query on EVALUATION.problem + Snomed term

tumor: f, t.
Query on EVALUATION.problem + Snomed term

hypopituitary: f, t.
Query on EVALUATION.problem + Snomed term

psych: f, t.
Query on EVALUATION.problem + Snomed term

TSH measured: f, t.
TSH: continuous.
T3 measured: f, t.
T3: continuous.
TT4 measured: f, t.
TT4: continuous.
T4U measured: f, t.
T4U: continuous.
FTI measured: f, t.
FTI: continuous.

TBG measured: f, t.
TBG: continuous.

Queries on OBSERVATION.lab_test.v1 and ? LOINC/SNOMED terms

Realistically though you will probably have to re-create the Y/N questionaire!!

What is the overall requirement?

Ian

Dr Ian McNicoll
office +44 (0)1536 414994
fax +44 (0)1536 516317
mobile +44 (0)775 209 7859
skype ianmcnicoll
ian.mcnicoll@oceaninformatics.com

Clinical analyst, Ocean Informatics, UK
openEHR Clinical Knowledge Editor www.openehr.org/knowledge
Honorary Senior Research Associate, CHIME, UCL
BCS Primary Health Care www.phcsg.org

3

Thanks Ian,
The requirement for me is to have an idea about the archetype based
representation of the information contained in this dataset.
I can simply create a dummy archetype with simple datatypes, but that
would not be a representative of an archetype based system that would
be managing this information.
In short, I wanted to observe related archetypes in their natural
environment :slight_smile: Archetype based data is going to become input for my
research at one point, and I want to make that input as close to real
life as possible.

Best Regards
Seref

4

Most of these data points would be carried within any decently
maintained EHR e.g. drug names, significant medical history,
procedure, lab tests etc. and would be carried in regular clinical
statement archetypes or equivalents as suggested.

However some other questions can really only be answered by an active
investigator e.g pregnancy status and 'sick'.

Ian

Dr Ian McNicoll
office +44 (0)1536 414994
fax +44 (0)1536 516317
mobile +44 (0)775 209 7859
skype ianmcnicoll
ian.mcnicoll@oceaninformatics.com

Clinical analyst, Ocean Informatics, UK
openEHR Clinical Knowledge Editor www.openehr.org/knowledge
Honorary Senior Research Associate, CHIME, UCL
BCS Primary Health Care www.phcsg.org

5

Hello
Someone there any advice on options editor INTERSYSTEMS
It is as strong presence in the U.S. and in Sweden and within standard EHRCOM
Europe-
Siemens in France offers a solution to medical records for hospitals based on
the products of this company and we have choices to make soon

INTERSYSTEMS
· Products
o Caché
   High performance object database
o Ensemble
   Platform for connected applications
o DeepSee
  Embedded real-time business intelligence
o HealthShare
  Health information exchange platform
o TrakCare
  Web-based healthcare information system

Thank's best regard's
Dr LONJON from FRANCE