RE: Upcoming DSE TC topics

Hi,
some more background on CDISC following up Eric's question 
Eric wrote: Is there a reference for the current ~20 [SDTM] domains already
supported? 

	Currently the set of dataset tabulation models cover 15 so called
domains for clinical observations, tying the different kinds of observations
to subjects in clinical studies, and around 8 domains for trial design. The
Reserved Domain Codes 1) from the Implementation Guideline document for SDTM
lists 40+ different domains. Also three more domains for different kinds of
findings are under review, i.e. MB Microbiology, PC & PP for PK
Concentration and Parameters, DA Drug Accountability, see
http://www.cdisc.org/models/sdtm/v1.1/index.html

Eric also outlines a case of using an RDF enabled "classification" model as
an alternative the "tabulation model". 

	Part of that shift is to go beyond exchanging human-readable table
variables, such as decoded classifications, and also to go beyond the
confined identification of table records as sequence number per dataset and
studies.  


I look forward to continue this case and other topics in the TC on Friday. 

Regards
Kerstin


1) Reserved SDTM Domain Codes:

AD Analysis Dataset  Special Purpose 
AE  Adverse Events  Events 
AU  Autopsy  Findings 
BM  Bone Mineral Density (BMD) Data  Findings 
BR  Biopsy   Findings 
CM  Concomitant Meds  Interventions 
CO Comments  Special Purpose 
DA Drug Accountability  Findings 
DC Disease Characteristics  Findings 
DM  Demographics  Special Purpose 
DS  Disposition  Events 
DV  Protocol Deviations  Events 
EE  EEG  Findings 
EG  ECG  Findings 
EX Exposure  Interventions 
FH  Family History  Events  
HU  Healthcare Resource Utilization   Findings  
IE  Inclusion / Exclusion  Findings 
IM   Imaging  Findings 
LB  Laboratory Data  Findings 
MB  Microbiology  Findings 
MH Medical History  Events 
ML  Meal Data  Interventions  
OM  Organ Measurements  Findings 
PC  PK Concentration  Findings 
PE  Physical Exam  Findings 
PP  PK Parameters  Findings 
QS  Questionnaires  Findings 
SC Subject Characteristics  Findings 
SE  Subject Elements  Study Design 
SG Surgery  Interventions 
SK  Skin Test  Findings 
SL  Sleep (Polysomnography) Data  Findings 
ST  Stress (Exercise) Test Data  Findings  
SU Substance Use  Interventions 
SV Subject Visits  Trial Design 
TA Trial Arms  Trial Design 
TE  Trial Elements  Trial Design 
TI  Trial Inclusion/Exclusion Criteria  Trial Design 
TS  Trial Summary  Trial Design 
TV Trial Visits  Trial Design 
VS  Vital Signs  Findings 

-----Original Message-----
From: Eric Neumann [mailto:eneumann@teranode.com]
Sent: 31 oktober 2006 17:02
To: Kerstin.L.Forsberg@astrazeneca.com
Cc: Bo.H.Andersson@astrazeneca.com; James McGurk; Ted Slater; Uwe
Trinks; public-hcls-dse@w3.org
Subject: Upcoming DSE TC topics


Thanks Kerstin,

I agree with your comments... Is there a reference for the current ~20
domains already supported? Would be good to know them in more detail... I'd
like to capture all the necessary resources and link them on the wiki site.

We also need to come up with 2 or 3 examples of what "recombinant clinical
data" looks like and can uniquely do. This will be very important defining
our role within the larger CD community. 

Based on you note, another question to consider is "Where specifically does
the tabulation model fall short when trying to address biomarkers?" I can
imagine several potential places, but we should identify these as a group.
For example, do we use something like a "classification structure" that (1)
maps a qualified set of measured (probe base) molecular analytes to a
subject in a study, as well as (2) maps them to the gene or metabolite
references behind each analyte ((3)plus a pathway), and (4) bundle the
analyte group with a statement to what disease or condition (e.g.,
hepatotoxicity, inflammatory response) they should be applied to. A
comparison of the current domain-observation model to the proposed
"classification" model seems to be in important next step!

BTW, "Biomarkers" should always be "measureable factors" that, in addition,
are usually "biochemical features or facets", but could also be "enhanced
images of cells or tissues" or even other forms of per subject measurables,
such as EEGs. Clearly there should be an easy and direct way to determine of
the biomarkers are of type "biochemical feature", and here is where I see
RDF helping us out tremendously!

best,
Eric

On Oct 29, 2006, at 2:47 PM, Kerstin.L.Forsberg@astrazeneca.com wrote:


	Hi Eric,
	hope we manage to find a new asap for a TC, see another email
regarding
	avalibility.

	Here some other comments:

	- Glad to see that you have met Wayne Kubick, the CDISC/SDS (SDTM)
team
	leader. 

	- You also mentioned BRIDG: Yes, I think we should discuss potential
links
	to the HL7 BRIDG work. (A very good overview of BRIDG as a webcast
can be
	found at bettermanagement.com,
	http://www.bettermanagement.com/seminars/seminar.aspx?l=13763). My
	spontaneous comments on it is that the intention of BRIDG is very
good: "To
	define implementation-independent domain semantics, To uncover the
myriad of
	semantic ambiguities present in the complex domain of clinical
research".
	However, I think they may have got lost in trying to do backward UML
	modelling starting from existing data exchange standards.


	- I also wanted to highlight something I read in the last TC in the
overall
	HCLS group http://www.w3.org/2006/10/12-HCLS-minutes.html: > CDISC
have no
	support for biomarkers

	I don't think that is a correct statement:

	CDISC provides "... a general framework for describing the
	organization of information collected during human and animal
studies. The
	model is built around the concept of observations, which consist of
discrete
	pieces of information collected during a study."  So far, the model
has been
	applied for 20+ different so called domains of observations like
Vital
	Signs, Microbiology, Pharmacokinetic, and more to come. However,
with the
	fundamental model SDTM provide of qualified and identified
observations any
	type of observations on subjects can be specified even though CDISC
has not
	yet provided a data exchange domain for it.

	However, so far CDISC have only considered implementations for
	tabulated datasets for exchange per clinical study (as SAS format,
and later
	on in XML using CDISC's general messaging format).

	In the Drug Safety and Efficacy task force we will put a semantic
	web perspective on the model and propose an open ended RDF
implementation to
	ensure that observations become recombinant cross clinical studies
applying
	ontologies for different types of observations.

	When I read definitions of the broad term "biomarkers" such as this
one:
	"A biomarker is some measurable factor that is specifically
	associated with a particular medical condition." or "A biochemical
feature
	or facet that can be used to measure the progress of disease or the
effects
	of treatment."

	I do interpret "measurable factor" and "biochemical feature or
facet" as
	"observations". Would be interesting to hear your point of view. 


	Hope to talk to you soon
	Kerstin


	-----Original Message-----
	From: public-hcls-dse-request@w3.org
	[mailto:public-hcls-dse-request@w3.org]On Behalf Of Eric Neumann
	Sent: 26 oktober 2006 23:50
	To: Kerstin.L.Forsberg@astrazeneca.com;
Bo.H.Andersson@astrazeneca.com;
	James McGurk; Ted Slater; Uwe Trinks
	Cc: public-hcls-dse@w3.org
	Subject: Tomorrow's DSE TC



	Folks,

	We are still on for tomorrow's 9-10am EDT call, though Kerstin has
informed
	my she can not make it. If enough need to excuse themselves we could
	reschedule, but I would suggest we try and move things forwards with
this
	next call. BTW, Has everyone subscribed to the new list:
	public-hcls-dse@w3.org ?

	I also met briefly with Wayne Kubick today (Lincoln Tech is part of
	Phase-Forward now), and gave him an overview of what we've discussed
and
	plan to do. He has offered to  get us more closely connected with
the BRIDG
	project (SW and model alignment), so we should also discuss this
option.

	Proposed Agenda:  Discussion around SDTM Model, it's current
functional
	specs,  and it's limitations...

	 - please see
	
http://esw.w3.org/topic/HCLSIG/Drug_Safety_and_Efficacy?action=AttachFile&do
	=get&target=CRT_DDSpecification1_0_0.pdf,

	
http://esw.w3.org/topic/HCLSIG/Drug_Safety_and_Efficacy?action=AttachFile&do
	=get&target=CDISC+SDTM+overview.doc, and
	
http://esw.w3.org/topic/HCLSIG/Drug_Safety_and_Efficacy?action=AttachFile&do
	=get&target=CDISCs+SDTM+basics.ppt

	cheers,
	Eric









Eric Neumann, PhD
co-chair, W3C Healthcare and Life Sciences,
and Senior Director Product Strategy
Teranode Corporation
411 1st Avenue South, Suite 700
Seattle, WA 98104
+1 (781)856-9132
www.teranode.com

Received on Thursday, 2 November 2006 09:09:19 UTC