Define.xml

Hi !

Since I am pretty new to this define.xml / SDTM topic I do have a dozen of questions. In the define.xml I am listing the LBTESTCD (and other xxTESTCD) into the VLM section. For each LBTEST I assigned a LBTESTCD accordingly which is also listed in the CRF. The source data also contains the mostly correct LBTESTCD's (some of them I had to change though), so what exactly is my "origin" of LBTESTCD then? Assigned, eDT, CRF? Since the LBTESTCD is not directly mentioned in the original CRF, I would assume that CRF is wrong.

Thanks for your help in advance!

Hi!!

The new Trial Summary Domain has en expected variable TSVALCD. Is this supposed to be in the codelist, thus, linked to it? In the CDISC example define.xml 2.0.0 all the new variables do not show up yet.

Thanks in advance!

Hi,

I have a little problem with linking FATESTCD and FATEST in define.xml as VLM. The problem is that FATESTCD and FATEST are not a unique pair (sponsor accepted as it is...). So I have one value for FATESTCD and 5 values for FATEST. I created 5 rows for FATESTCD, but cannot assign the correct labels according to FATEST.

An example program is below (with only two rows instead of 5)

Hi all,

Does the ELEMENT and ETCD belong to the codelist of the define.xml? I put them into the codelist separately and after running the define.xml cdisc checker I got following error message:

"Value for ELEMENT not found in (ETCD) user-defined codelist"

Don't understand what the ELEMENT should be in ETCD.

Could someone help me out?

Shouldn't 3.1.3 for SDTM be a valid version number? 

I have a SAS macro that generates and runs a BAT file to validate define.xml.  It was working fine until a couple of days ago, when it produced an XLS file with the first sheet ("Issue Summary") populated correctly with config file, date, & column headers, but no messages.  Here's the file created by the macro:

@echo off

rem Date-time Friday, April 12, 2013 10:41 AM

rem OpenCDISC version [1.2.1]  Define configuration version [1.0]
rem Output XLS [c:\temp\validateDefineXML_2013_04_12_10_41.xls]

Will there be a check for Attribute Length="0" for variable definitions or VLM? It's a define.xml V1 specification requirement (ODM requirement Length attribute is a positiveInteger on which the define.xml V1 specification is based on).

Will there be a check for IsReferenceData? For example, trial design is usually referential and should be set to IsReferenceData='Yes', whereas most other domains are default IsReferenceData="No".

ODM v1.2.0

"If IsReferenceData is Yes, this type of item group can occur only within a ReferenceData element. If IsReferenceData is No, this type of item group can occur only within a ClinicalData element. The default for this attribute is No. ?

HI together,

my question is not really related to OpenCDISC validator, but maybe someone could answer it anyway. How do you interpret the ADaM Model Guideline 2.1. According to my understanding, the analysis-value-metadata which was previously a separate file has now to be integrated into the analysis-variable-metadata. This is no problem, the question is how can a define.xml be generated based on this new analysis-variable-metadata? How do you generate th define.xml for adams. Do you have different files as basis or only one?

Thanks for help,

If anyone has any idea why this is happening for one SDTM dataset and all SUPPQUALs ... maybe something to do with define.xml construction? It's not clear why I'm getting this error. In this example, it looks like STUDYID within domain PG is identified in a way consistent with what it is.

(Using Version 1.4)

(see attached pdf document which features:

OpenCDISC screencap

define.xml screencap

SAS Universal Viewer screencap

)