Using SAS® Clinical Data Integration to Roundtrip a Complete Study

Using SAS® Clinical Data Integration to Roundtrip a Complete Study
PharmaSUG 2015 – Paper SS10-SAS
SAS®
Using
Clinical Data Integration to Roundtrip a Complete Study
Study Metadata (Define-XML) and Study Data (Dataset-XML)
Ken Ellis, SAS Institute Inc., Cary, NC, USA
ABSTRACT
SAS Clinical Data Integration 2.6 now supports the complete CDISC Define-XML 2.0 specification for both import
from, and creation of define.xml files. Support for value-level metadata and supplemental documents, as well as
support for the new CDISC Dataset-XML specification have been added and give the user a powerful means by
which to import and create complete study definitions. Topics covered include:




import of CDISC Define-XML file metadata, including domains, value-level metadata, supplemental
documents, computational algorithms and code lists
import of accompanying domain data from CDISC Dataset-XML files
creation of a complete CDISC Define-XML file including all imported metadata (roundtrip)
creation of CDISC Dataset-XML files from the imported domain tables (roundtrip)
INTRODUCTION
The CDISC Define-XML 2.01 specification (define.xml) provides a machine readable description of datasets used for
submission to the FDA. The new CDISC Dataset-XML2 specification (dataset-xml) provides a means for describing
the underlying data associated with CDISC Define-XML 2.0 compliant dataset definitions.
SAS Clinical Data Integration 2.63, in combination with SAS Clinical Standards Toolkit 1.74, provides a rich set of
tools for reading and writing both define.xml and dataset-xml files. This paper will focus on those tools and the
processes that facilitate working with these files. A sample work flow will be developed that roundtrips a define.xml
file and its associated dataset-xml files. It is assumed that the user has some familiarity with SAS Clinical Data
Integration and the CDISC specifications referenced as in-depth discussions of each is beyond the scope of this
paper. (See the References section for more information.)
Using SAS® Clinical Data Integration to Roundtrip a Complete Study
THE TOOLS
SAS Clinical Data Integration provides the following tools for reading and writing define.xml and dataset.xml files. We
will look at them more in-depth as we encounter them in our workflow example:
1.
2.
Study Level
a. Study from Define...
Creates a basic study shell using the selected folder template. Values for the study name,
description and protocol title are derived from the define.xml file. The user can optionally associate
a data standard and library with the new study.
b.
Domain(s) from Define…
Creates empty SAS table shells for each SDTM domain selected for import. All table level and
column metadata are derived from the define.xml file.
c.
Analysis Dataset(s) from Define…
Creates empty SAS table shells for each ADaM table selected for import. All table level and column
metadata are derived from the define.xml file.
d.
Codelist Table from Define…
Creates a single SAS table for all codelists selected for import. The generated table can be
associated with a terminology package that is associated with the study.
Table Level
a. Value Level Metadata from Define…
Imports value level metadata found in the define.xml file into a pre-defined SAS data table that can
be associated with the study via the study properties.
b.
3.
Supplemental Documents from Define…
Imports supplemental document data found in the define.xml file into a pre-defined SAS data table
that can be associated with the study via the study properties.
Clinical Transforms
a. CDISC-Define Creation
Generates a new define.xml file from the selected inputs.
b.
CDISC-Define Validation
Validates that a define.xml file adheres to a schema.
c.
CDISC-Dataset-XML Creation
Creates new dataset-xml files for each selected table.
d.
CDISC-Dataset-XML Import
Creates a SAS data table representation of a dataset-xml file.
SETUP
There are several administrative tasks that must first be done before continuing. They are typically the responsibility
of a Clinical Administrator and are beyond the scope of this paper. The basic necessities are as follows:
1.
2.
3.
4.
5.
Import and activate a Data Standard
Create a suitable folder structure template for your study data
Create a Terminology Package for any codelist tables and associate it with the study
Create a Value Level Metadata table (empty) and associate it with the study
Create a Supplemental Documents table (empty) and associate it with the study.
2
Using SAS® Clinical Data Integration to Roundtrip a Complete Study
THE IMPORT PROCESS
We will develop a basic process that imports the data and metadata associated with a define.xml file and its
associated dataset-xml files. We will examine the parts of each where applicable so that you gain some familiarity
with not only the process, but what is being created. Note that the tools provided for importing a define.xml file are
very granular and thus allow the user to pick and choose the parts of the define.xml that he or she might want to
import. Finally, we will export (roundtrip) our study data back into a new define.xml file and associated dataset-xml
files and will validate that the generated define.xml conforms to a schema. Let’s get started…
1.
Generate a new SAS Clinical Data Integration Study
Use the New Study from Define… context menu to launch the New Study from Define wizard (Figure 1).
The wizard will first prompt for the location of a define.xml to use as an input source. Once selected, the
wizard will validate that the file is indeed a define.xml file and then import the study name, description and
protocol title values from the define.xml file’s GlobalVariables element (Figure 2). Subsequent pages of the
wizard allow the opportunity to change these default values, associate a folder template, and optionally
associate a data standard (recommended), a library and a terminology package with the study. Once
created, the new study will be displayed in the SAS Clinical Data Integration Folder tree.
<GlobalVariables>
<StudyName>CDISC01</StudyName>
<StudyDescription>
CDISC Test Study
</StudyDescription>
<ProtocolName>CDISC01</ProtocolName>
</GlobalVariables>
Figure 2. GlobalVariables Element
Figure 1. New Study from Define Wizard
2.
Import the SDTM Domain definitions
Next, import the SDTM domain definitions and create empty SAS data table shells for each domain
definition found in the define.xml file. All metadata associated at both the table and column level will be
imported and stored in the SAS Clinical Data Integration metadata repository. Note that in order to reference
permanent data tables on the file system that are associated with the metadata representations in SAS
Clinical Data Integration, you will need to associate a Base SAS Library definition as part of the stored table
metadata. It is also important to note that physical tables will not be created as part of the domain import
process.
Use the New Domain(s) from Define… context menu wizard to launch the New Domain(s) from Define
wizard (Figure 3). The wizard will prompt you for the location of a define.xml to use as an input source and a
location to store the table metadata definition within SAS Clinical Data Integration. Next you will be required
to select a data standard to associate with the table(s) you are about to import. Subsequent wizard pages
allow you to select any (or all) domain definitions found and to optionally associate a SAS library definition
for reference to permanent table storage. Upon completion of the wizard your domain table definitions will be
created in the SAS Clinical Data Integration metadata repository (Figure 3, 4).
3
Using SAS® Clinical Data Integration to Roundtrip a Complete Study
Figure 3. New Domain(s) from Define Wizard
Figure 4. Partial Listing of Domains
A quick look at the AE domain’s column properties (Figure 5) from within SAS Clinical Data Integration
shows that the ItemGroupDef element’s attributes were imported correctly (Figure 6).
Figure 5. Column Properties for the AE Domain
4
Using SAS® Clinical Data Integration to Roundtrip a Complete Study
<ItemGroupDef OID="IG.AE" Name="AE" Repeating="Yes" IsReferenceData="No" SASDatasetName="AE"
Domain="AE" Purpose="Tabulation" def:Class="EVENTS" def:Structure="One record per adverse event per subject"
def:ArchiveLocationID="LF.AE">
<Description>
<TranslatedText xml:lang="en">Adverse Events</TranslatedText>
</Description>
<ItemRef ItemOID="IT.AE.STUDYID" Mandatory="Yes" OrderNumber="1" KeySequence="1"/>
<ItemRef ItemOID="IT.AE.DOMAIN" Mandatory="Yes" OrderNumber="2"/>
<ItemRef ItemOID="IT.AE.USUBJID" Mandatory="Yes" OrderNumber="3" KeySequence="2" MethodOID="MT.AE.USUBJID"
<ItemRef ItemOID="IT.AE.AESEQ" Mandatory="Yes" OrderNumber="4" MethodOID="MT.AE.AESEQ"/>
<ItemRef ItemOID="IT.AE.AESPID" Mandatory="No" OrderNumber="5"/>
<ItemRef ItemOID="IT.AE.AETERM" Mandatory="Yes" OrderNumber="6"/>
<ItemRef ItemOID="IT.AE.AEMODIFY" Mandatory="No" OrderNumber="7"/>
<ItemRef ItemOID="IT.AE.AEDECOD" Mandatory="Yes" OrderNumber="8" KeySequence="3"/>
<ItemRef ItemOID="IT.AE.AEBODSYS" Mandatory="No" OrderNumber="9"/>
<ItemRef ItemOID="IT.AE.AESEV" Mandatory="No" OrderNumber="10"/>
<ItemRef ItemOID="IT.AE.AESER" Mandatory="No" OrderNumber="11"/>
<ItemRef ItemOID="IT.AE.AEACN" Mandatory="No" OrderNumber="12"/>
<ItemRef ItemOID="IT.AE.AEREL" Mandatory="No" OrderNumber="13"/>
<ItemRef ItemOID="IT.AE.AESTDTC" Mandatory="No" OrderNumber="14" KeySequence="4"/>
<ItemRef ItemOID="IT.AE.AEENDTC" Mandatory="No" OrderNumber="15"/>
<ItemRef ItemOID="IT.AE.AESTDY" Mandatory="No" OrderNumber="16" MethodOID="MT.AE.AESTDY"/>
<ItemRef ItemOID="IT.AE.AEENDY" Mandatory="No" OrderNumber="17" MethodOID="MT.AE.AEENDY"/>
<ItemRef ItemOID="IT.AE.AEENRF" Mandatory="No" OrderNumber="18"/>
<def:leaf ID="LF.AE" xlink:href="ae.xpt">
<def:title>ae.xpt</def:title>
</def:leaf>
</ItemGroupDef>
Figure 6. The ItemGroupDef Element for the AE Domain
3.
Import the Codelists
SAS Clinical Data Integration can manage two distinct types of codelist data; those imported directly from
SAS Clinical Standards Toolkit, and external codelist tables such as those imported from a define.xml or
other source. The process of associating a codelist table with a study is:
a. Create a Terminology Package
b. Associate the Terminology Package with the study
c. Associate the imported codelist table with the Terminology Package
These steps are not within the scope of this paper and it is assumed that the user is familiar with the
processes involved. It is also important to note SAS Clinical Data Integration does not manage any
metadata associated with an imported codelist table other than the SAS Base Library association to the
physical data table itself.
Use the New Codelist Table from Define… context menu wizard to launch the New Codelist Table from
Define wizard. The wizard will prompt you for the location of a define.xml to use as input and a location to
store the codelist table’s metadata definition within SAS Clinical Data Integration. The next wizard page will
display all of the codelists that are available to import. Select any or all depending on your need (Figure 7).
The final wizard page asks you to provide a codelist name, optional description and a required SAS Library
used to create the permanent physical SAS data table. Upon completion of the wizard the new codelist table
is created and registered with SAS Clinical Data Integration. Remember that you must also register the new
codelist table with a SAS Terminology Package for it to be associated with a study.
5
Using SAS® Clinical Data Integration to Roundtrip a Complete Study
Figure 7. Codelist Selection Wizard Page
Finally, a quick look at a sample of the generated codelist table (Figure 8) shows that the CodeList element
was imported correctly from the define.xml file (figure 9).
Figure 8. Partial Listing of the Generated Codelist Table
<CodeList OID="CL.ACN" Name="Action Taken with Study Treatment" DataType="text" SASFormatName="$ACN">
<EnumeratedItem CodedValue="DOSE NOT CHANGED" OrderNumber="1">
<Alias Context="nci:ExtCodeID" Name="C49504"/>
</EnumeratedItem>
<EnumeratedItem CodedValue="DOSE REDUCED" OrderNumber="2">
<Alias Context="nci:ExtCodeID" Name="C49505"/>
</EnumeratedItem>
<EnumeratedItem CodedValue="DRUG INTERRUPTED" OrderNumber="3">
<Alias Context="nci:ExtCodeID" Name="C49501"/>
<EnumeratedItem CodedValue="DRUG WITHDRAWN" OrderNumber="4">
<Alias Context="nci:ExtCodeID" Name="C49502"/>
</EnumeratedItem>
<Alias Context="nci:ExtCodeID" Name="C66767"/>
</CodeList>
Figure 9. The CodeList Element for “Action Taken with Study Treatment”
6
Using SAS® Clinical Data Integration to Roundtrip a Complete Study
4.
Import Value Level Metadata and Supplemental Documents
The processes for importing Value Level Metadata and Supplemental Documents from a define.xml file are
identical other than the wizards used. The process steps are:
a.
b.
c.
Create an empty Value Level Metadata or Supplemental Documents table
Associate the table with the study
Import data from a define.xml into the table
Note that SAS Clinical Data Integration does not manage any metadata associated with the table or table
columns other than the metadata link to the table itself. Users are free to edit the data within the tables as
long as the table column integrity remains intact.
A Clinical Administrator would first create an empty table by selecting Study tab from within the Study
Properties page, then selecting the appropriate Add… button for Value Level Metadata Table or
Supplemental Documents Table (Figure 10). For this example we will trace the path of creating a new empty
Value Level Metadata table.
Figure 10: The Study Properties “Study” Tab
7
Using SAS® Clinical Data Integration to Roundtrip a Complete Study
The Available Value Level Metadata Tables window is displayed and the user can select a preexisting
table to associate with the study, or create a new one by selecting the New… button (Figure 11).
Figure 11. The Available Value Level Metadata Tables Window
If New… is selected, the New Value Level Metadata Table wizard is launched. The user must select a
location within SAS Clinical Data Integration to store the metadata link to the physical table that will be
created on the user’s file system. Once chosen, the user is prompted for a table name, optional description
and SAS Library used to create the permanent physical SAS data table (Figure 12).
Figure 12. The Table and Library Specification Wizard Page
8
Using SAS® Clinical Data Integration to Roundtrip a Complete Study
Upon completion of the wizard, the new empty table is created and pre-selected for association with the
study. And, as was noted earlier, this same process is used to create and associate a Supplemental
Documents table with a study, and after creating and adding both tables the Study Properties, Study tab
would appear as in Figure 13.
Figure 13. The Study Properties Page
A quick look at the study’s folders shows that the tables have been created (Figure 14).
Figure 14. The Empty Tables have been created
9
Using SAS® Clinical Data Integration to Roundtrip a Complete Study
The last step of the process involves importing data into the empty SAS data tables created in the previous
steps. To do this, launch the Value Level Metadata from Define… or Supplemental Documents from
Define… context menu by right-clicking on the table icon. The appropriate wizard will be launched and once
again prompt you for the path to the define.xml file from which you wish to import the data. Upon completion
of the wizard, the data will be imported into the appropriate table (Figure 15).
Figure 15. Partial Listing of imported Value Level Metadata
5.
Import Define-XML Data into the Domain Table Shells
SAS Clinical Data Integration provides a pre-defined Clinical transform that allows you to import CDISC
Dataset-XML files. The transform relies on functionality provided by SAS Clinical Standards Toolkit to first
create the domain tables (SAS data sets) on the file system and to then load data into them. The domain
tables are “linked” to their counterparts within SAS Clinical Data Integration via the SAS Library that was
initially associated with the metadata representation of the domain when it was imported from the define.xml
file. This same define.xml file must be used as an input to the transformation when creating and loading the
dataset-xml files as metadata definitions within it are used to validate the incoming data.
First, create a new Document within SAS Clinical Data Integration that references the define.xml by
selecting the New->Document… context menu from within the study and pointing the document to the
define.xml file that was used to import the domain metadata definitions (Figures 16, 17).
Figure 16. Document Definition for the define.xml File
10
Figure 17. Document Folder
Using SAS® Clinical Data Integration to Roundtrip a Complete Study
Next, create a new Job and drop the CDISC-Dataset-XML Import transform onto the Job palette. The
transform takes one input; the Document (created above) that references the define.xml file that was initially
used to import the domain metadata definitions (Figure 18).
Figure 18. The CDISC-Dataset-XML Import Job Palette
Next, set the input source (where the dataset-xml files reside that you will be importing) and output
destination (folder where the physical SAS data tables will be created). To do so, double-click on the
transform to open its properties window and then select the Generation tab. Browse to, and select the folder
where the dataset-xml files reside on your file system that you wish to import. Next select the output library
(destination for the SAS data tables that will be created). Note that you should use the same library definition
that was used to import the domain metadata definitions in step 2. By doing so, you will establish the “link”
between the metadata definitions and the physical tables (Figure 19).
Figure 19. The Completed Transform’s Generation Tab
11
Using SAS® Clinical Data Integration to Roundtrip a Complete Study
Finally, select OK button on the transform’s properties window and then run the transform. When the Job
completes, there will be a SAS data table created for every dataset-xml file that was found in the input folder
(Figure 20).
Figure 20. Partial Listing of Generated Domain Data Tables
To verify the link between the domain metadata definitions created in step 2, and the physical tables created
via the CDISC-Dataset-XML Import transform, simply right click on any of the domain definitions inside of
SAS Clinical Data Integration that have a corresponding physical table and select the Open context menu
item. (Figures 21, 22).
Figure 21. Partial Listing of AE Domain
12
Using SAS® Clinical Data Integration to Roundtrip a Complete Study
<ItemGroupData ItemGroupOID="IG.AE" data:ItemGroupDataSeq="1"><ItemData ItemOID="IT.AE.STUDYID"
Value="CDISC01"/>
<ItemData ItemOID="IT.AE.DOMAIN" Value="AE"/>
<ItemData ItemOID="IT.AE.USUBJID" Value="CDISC01.100008"/>
<ItemData ItemOID="IT.AE.AESEQ" Value="1"/>
<ItemData ItemOID="IT.AE.AESPID" Value="1"/>
<ItemData ItemOID="IT.AE.AETERM" Value="AGITATED"/>
<ItemData ItemOID="IT.AE.AEMODIFY" Value="AGITATION"/>
<ItemData ItemOID="IT.AE.AEDECOD" Value="Agitation"/>
<ItemData ItemOID="IT.AE.AEBODSYS" Value="Psychiatric disorders"/>
<ItemData ItemOID="IT.AE.AESEV" Value="MILD"/>
<ItemData ItemOID="IT.AE.AESER" Value="N"/>
<ItemData ItemOID="IT.AE.AEACN" Value="DOSE NOT CHANGED"/>
<ItemData ItemOID="IT.AE.AEREL" Value="POSSIBLY RELATED"/>
<ItemData ItemOID="IT.AE.AESTDTC" Value="2003-05"/>
<ItemData ItemOID="IT.AE.AESTDY" Value="3"/>
<ItemData ItemOID="IT.AE.AEENRF" Value="AFTER"/>
</ItemGroupData>
Figure 22. The ItemGroupData Element for the AE Domain Corresponding to the First Row
THE EXPORT PROCESS
SAS Clinical Data Integration provides a set of pre-defined Clinical Transforms that can be used to create define.xml
and dataset-xml files, and a transform that can be used to validate a define.xml file against a standard schema. The
transforms rely on functionality provided by SAS Clinical Standards Toolkit. We will first create a define.xml file that
incorporates the domain, codelist, value level metadata and supplemental documents tables that we created in steps
2-4 of the Import Process. Next we will validate the newly created define.xml file against the standard schema for
CDISC-Define-XML Version 2.0. Finally, we will export the physical domain table data into corresponding dataset-xml
files, thus completing the roundtrip.
1.
Create Source and Destination Documents
The clinical transforms used in the following steps require Document objects that reference physical files on
your file system. The output document (Output Define.xml) will serve as an output for the CDISC-DefineXML Creation transform. This document should reference the file that you want to use as the output
destination for the define.xml file that will be created in step 2 below. A separate input document (Input
Define.xml) will serve as an input to the CDISC-Dataset-XML Creation transformation and should
reference the define.xml file that was used during the import process. To create a SAS Clinical Data
Integration Document, select the New->Document… context menu from within the study. (Figures 23, 24).
Figure 23. The Input define.xml file Document Definition
13
Figure 24. Document Folder
Using SAS® Clinical Data Integration to Roundtrip a Complete Study
2.
Export the Study Domains to a Define.xml File
To create a define.xml file you must first create a new Job and drop the CDISC-Define-Creation transform
onto the Job’s palette. The transform takes the Output Define.xml document created in step 1 above
(Figure 25).
Figure 25. The CDISC-Define Creation Job Palette
Double-click on the transform to open its properties window and then select the Tables tab. Next, in the
Data Standard Type pull-down menu, select the data standard that you initially associated with your study
and the domain table shells that you created earlier. The Study/Submission pull-down menu will populate
with all studies associated with the selected data standard. Select the study created in step 1 and the
Tables list will be populated with all domain definitions that are defined within the study for this data
standard. Select the domains that should be included in the new define.xml that will be created (Figure 26).
Figure 26. The Tables Tab with all Domains Selected for a Study
14
Using SAS® Clinical Data Integration to Roundtrip a Complete Study
Next select the Generation tab for the transform. This page allows you to set various options for the new
define.xml file you are about to create. SAS Clinical Data Integration supports creation of both define
versions 1 and 2 (note that the default is version 1), as well as creation of a Define.pdf (version 1 only) and
support for default or custom stylesheets. Select 2.0 in the Define version pull-down, and select the Use
default output stylesheet radio button (Figure 27).
Figure 27. The Generation Tab
Finally, select the OK button on the transform’s properties window and then run the transform. When the Job
completes, the new define.xml file will be created using the physical file referenced by the output document.
3.
Validate the Define.xml File
SAS Clinical Data Integration can also be used to validate any CDISC compliant define.xml file against its
standard schema. To validate a define.xml file, first create new Job and drop the CDISC-Define-Validation
transform onto the Job’s palette. The transform takes only one external input, a document object that points
to physical define.xml file to be validated. In our case, this would be the Output Define.xml document
created in step 1 above as it references the define file created in step 2 above. (Figure 28).
Figure 28. The CDISC-Define-Validation Job Palette
15
Using SAS® Clinical Data Integration to Roundtrip a Complete Study
Double-click on the transform to open its properties window and then select the Validation tab. Note that the
transformation can validate either define versions 1 or 2. In the Define version pull-down box select 2.0,
then close the properties window by selecting the OK button. Run the transform. When finished, open the
CRT-DDS Validation Results (Work Table) and verify that no errors were generated (Figure 29).
Figure 29. The Validation Results Table
4.
Export the Domains to Dataset-XML files
To export dataset-xml files you must first create a new Job and drop the CDISC-Dataset-XML Creation
transform onto the Job palette. The transform takes one input; a Document such as the one created in step
1 that references a define.xml. Note that the metadata from the referenced define.xml file must match that of
the domain table(s) that are to be written out as dataset-xml files. For the purposes of this paper, we could
use either the input or output define.xml documents since we did not alter any of the domain metadata. For
completeness, we will use the Output Define.xml document containing the current metadata definitions.
(Figure 30).
Figure 30. The CDISC-Dataset-XML Creation Job Palette
16
Using SAS® Clinical Data Integration to Roundtrip a Complete Study
Double-click on the transform to open its properties window and then select the Tables tab. In the Data
Standard Type pull-down menu, select the data standard that you initially associated with your study and
the domain table shells that you created earlier. The Study/Submission pull-down menu will populate with
all studies associated with the selected data standard. Select the study whose table data you wish to output
as dataset-xml files and the Tables list will populate. Finally, select the tables for which dataset-xml files will
be created. (Figure 31).
Figure 31. The Tables Tab with all Domains Selected for a Study
Next, select the Generation tab in the transform’s properties window. Select a physical Output folder
location for writing the dataset-xml files and then select the OK button to close the properties window. Run
the transform. When the transform has completed, the output folder will be populated with individual
dataset-xml files for each table selected (Figure 32). Finally, a quick visual comparison of the output datasetxml file’s contents with the content of the corresponding original dataset-xml file that was imported will show
that the correct data was written to the dataset-xml file (Figures 33, 24)
Figure 32: Partial listing of the dataset-xml Files Created
17
Using SAS® Clinical Data Integration to Roundtrip a Complete Study
<?xml version="1.0" encoding="UTF-8"?>
<!-- Produced from SAS data using SAS Clinical Standards Toolkit -->
<ODM
xmlns="http://www.cdisc.org/ns/odm/v1.3"
xmlns:data="http://www.cdisc.org/ns/Dataset-XML/v1.0"
ODMVersion="1.3.2"
FileType="Snapshot"
FileOID="STUDY1.AE"
PriorFileOID="DEFINE"
CreationDateTime="2014-07-15T17:08:26"
data:DatasetXMLVersion="1.0.0">
<ClinicalData StudyOID="STUDY1" MetaDataVersionOID="MDV.CDISC01.SDTMIG.3.1.2.SDTM.1.2">
<ItemGroupData ItemGroupOID="IG.AE" data:ItemGroupDataSeq="1">
<ItemData ItemOID="IT.AE.STUDYID" Value="CDISC01"/>
<ItemData ItemOID="IT.AE.DOMAIN" Value="AE"/>
<ItemData ItemOID="IT.AE.USUBJID" Value="CDISC01.100008"/>
<ItemData ItemOID="IT.AE.AESEQ" Value="1"/>
<ItemData ItemOID="IT.AE.AESPID" Value="1"/>
<ItemData ItemOID="IT.AE.AETERM" Value="AGITATED"/>
<ItemData ItemOID="IT.AE.AEMODIFY" Value="AGITATION"/>
<ItemData ItemOID="IT.AE.AEDECOD" Value="Agitation"/>
<ItemData ItemOID="IT.AE.AEBODSYS" Value="Psychiatric disorders"/>
<ItemData ItemOID="IT.AE.AESEV" Value="MILD"/>
<ItemData ItemOID="IT.AE.AESER" Value="N"/>
<ItemData ItemOID="IT.AE.AEACN" Value="DOSE NOT CHANGED"/>
<ItemData ItemOID="IT.AE.AEREL" Value="POSSIBLY RELATED"/>
<ItemData ItemOID="IT.AE.AESTDTC" Value="2003-05"/>
<ItemData ItemOID="IT.AE.AESTDY" Value="3"/>
<ItemData ItemOID="IT.AE.AEENRF" Value="AFTER"/>
</ItemGroupData>
Figure 33: First Row of Input Dataset-xml File Data for the AE Domain
<?xml version="1.0" encoding="UTF-8"?>
<!-- Produced from SAS data using the SAS Clinical Standards Toolkit. -->
<ODM
xmlns="http://www.cdisc.org/ns/odm/v1.3"
xmlns:data="http://www.cdisc.org/ns/Dataset-XML/v1.0"
ODMVersion="1.3.2"
FileType="Snapshot"
FileOID="STUDY1.AE"
PriorFileOID="DEFINE"
CreationDateTime="2015-04-06T15:44:13"
data:DatasetXMLVersion="1.0.0">
<ClinicalData StudyOID="STUDY1" MetaDataVersionOID="MDV.CDISC01">
<ItemGroupData ItemGroupOID="IG.AE" data:ItemGroupDataSeq="1">
<ItemData ItemOID="IT.AE.STUDYID" Value="CDISC01"/>
<ItemData ItemOID="IT.AE.DOMAIN" Value="AE"/>
<ItemData ItemOID="IT.AE.USUBJID" Value="CDISC01.100008"/>
<ItemData ItemOID="IT.AE.AESEQ" Value="1"/>
<ItemData ItemOID="IT.AE.AESPID" Value="1"/>
<ItemData ItemOID="IT.AE.AETERM" Value="AGITATED"/>
<ItemData ItemOID="IT.AE.AEMODIFY" Value="AGITATION"/>
<ItemData ItemOID="IT.AE.AEDECOD" Value="Agitation"/>
<ItemData ItemOID="IT.AE.AEBODSYS" Value="Psychiatric disorders"/>
<ItemData ItemOID="IT.AE.AESEV" Value="MILD"/>
<ItemData ItemOID="IT.AE.AESER" Value="N"/>
<ItemData ItemOID="IT.AE.AEACN" Value="DOSE NOT CHANGED"/>
<ItemData ItemOID="IT.AE.AEREL" Value="POSSIBLY RELATED"/>
<ItemData ItemOID="IT.AE.AESTDTC" Value="2003-05"/>
<ItemData ItemOID="IT.AE.AESTDY" Value="3"/>
<ItemData ItemOID="IT.AE.AEENRF" Value="AFTER"/>
</ItemGroupData>
Figure 33: First Row of Output Dataset-xml File Data for the AE Domain
18
Using SAS® Clinical Data Integration to Roundtrip a Complete Study
CONCLUSION
SAS Clinical Data Integration Version 2.6, coupled with the underlying functionality of SAS Clinical Standards Toolkit
Version 1.7 provides a powerful set of tools for managing both the import and export of clinical trials metadata and
data. These tools, in combination with others provided for metadata management within SAS Clinical Data
Integration, facilitate the easy exchange of information between a CRO and a study sponsor, and can be used to
generate key pieces of a submission to the FDA.
REFERENCES
1.
2.
3.
4.
CDISC Define-XML Specification, Version 2.0, March 5, 2013
(http://www.cdisc.org/define-xml)
CDISC Dataset-xml, Version 1.0
(http://www.cdisc.org/dataset-xml)
SAS Clinical Data Integration 2.6: User’s Guide
(http://support.sas.com/documentation)
SAS Clinical Standards Toolkit 1.7: User’s Guide
(http://support.sas.com/documentation)
CONTACT INFORMATION
Your comments and questions are valued and encouraged. Contact the author at:
Ken Ellis
SAS Institute Inc.
SAS Campus Drive
Cary, NC 27519, USA
SAS and all other SAS Institute Inc. product or service names are registered trademarks or trademarks of SAS
Institute Inc. in the USA and other countries. ® indicates USA registration.
Other brand and product names are trademarks of their respective companies.
19
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertisement