Scribd
Upload
1 views14 pages

SDTM001 Chapter 3 Script

SDTM is a foundational standard in clinical data management that organizes collected data into domains with standard variables and controlled terminology. It is designed to hold various data types, including those from clinical trials and electronic health records, while maintaining the integrity of the collected data. ADaM serves a different purpose by supporting analysis of the collected data, complementing SDTM for regulatory review.

Uploaded by

kevinfeugo
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
1 views14 pages

SDTM001 Chapter 3 Script

SDTM is a foundational standard in clinical data management that organizes collected data into domains with standard variables and controlled terminology. It is designed to hold various data types, including those from clinical trials and electronic health records, while maintaining the integrity of the collected data. ADaM serves a different purpose by supporting analysis of the collected data, complementing SDTM for regulatory review.

Uploaded by

kevinfeugo
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 14

• <no script>

1
In the previous slides, we’ve examined SDTM as a single structure. Now, let’s look a
little bit closer at SDTM in both its modeling and implementation.

<ANIM1> SDTM is considered to be one of the CDISC “foundational” standards. SDTM


has a foundational model on which we build different implementations. Each of the
other standards, such as ADaM and CDASH, has its own model which is found in the
separate documents.

<ANIM2> For example, there are implementation guides based on the models for
human clinical trials, for nonclinical trials and other implementation topics such as
Medical Devices.

<ANIM3> Therapeutic areas user guides, or TAs, are focused on specific conditions
like H.I.V. or Diabetes. These TA user guides are built on the foundational models and
contain SDTM-related implementations as well as CDASH and ADaM implementations
focused on the therapeutic area of interest.

<ANIM4> Controlled terminologies are used throughout for a consistent


representation of data. For example, when we have a lab test called Cholesterol, we

2
will always use the same value to represent that test name.

2
SDTM is designed to hold anything you collect, regardless of the method you used to
collect it. This includes CRF and EDC data, but it also includes data received from a
core lab or an ePRO vendor.
SDTM takes all of this collected data and organizes into standard variables with
standard data types and other standard metadata.

3
SDTM organizes data into logical groupings called “Domains”. Each domain
contains data for one “topic”, such as Adverse events or Demographics, or Lab
Data.

Definitions of each SDTMIG modeled domain can be found in the SDTM


controlled terminology and in the SDTM Implementation Guide.

4
There are standard variable names, and fixed set of variables that are allowed
in each domain. All of these variables have standard names, labels, data
types and other attributes.

5
The data are represented using controlled terminologies so that the same data
concepts are represented consistently in the data repository. This includes
standard naming conventions for many common tests and measures, such as
lab values and vitals signs. Controlled terminology is a very important part of
building a consistent data repository.

6
Throughout the SDTM we find standard Assumptions, or rules, for
implementation. These rules make our data tabulations consistent across
studies and sponsors. Assumptions are found in each domain and in section 4
of the SDTM IG. Following the same rules for implementation is another
important part of building a standard data repository.

7
Although no sponsor defined, non-standard variables can be used in SDTM “parent”
domains, there is a standard structure for organizing non-standard variables and
linking them to the parent domain.

8
Again, SDTM is designed to hold anything we collect, regardless of how it was
collected. This includes data collected using paper case report forms updated by
queries, data collected through electronic Data Capture (EDC), and data received
through electronic transfers (lab, ECG) from a core lab, and even data retrieved
directly from an electronic health record (eHR).
Because SDTM should represent your data “as collected”, there are some General
Rules:
<ANIM1> Missing values should remain missing
Partial dates should remain partial, as collected
No Imputations of any values should be done in SDTM
and Special characters should remain on your lab data

<ANIM2> Keep in mind that SDTM does not specify what data to collect; the Sponsor
decides this based on science and regulation

9
<ANIM1> SDTM was developed to hold any data that are collected.

<ANIM2> CDASH, unlike SDTM, directly addresses data collection.

<ANIM3> CDASH is harmonized with SDTM, and the CDASHIG provides maps to CDISC
SDTM data variables and also provides maps from CDASH CRF data fields.

10
As mentioned on the previous slide, SDTM is meant to show data as collected. If
imputation rules or other rules for filling in missing data need to be applied before
you do an analysis,

<ANIM1> these imputations should be placed in another type of dataset called


ADaM. ADaM is the Analysis Data Model, and the basic purpose of ADaM is to show
how the collected data were used in the analysis.

<ANIM2> ADaM datasets should support the data presented in a Clinical Study
Report, and both SDTM and ADaM are useful to a reviewer.

<ANIM3> Please note that when SDTM was first published, it was the only submission
data standard and it includes some analysis-type variables in addition to the collected
data.

11
Even though some of the values might be the same across SDTM and ADaM
their basic purposes are different and they are both needed and useful to a
Reviewer.

12
Even though some of the values might be the same across SDTM and ADaM,
their basic purposes are different and they are both needed and useful to a
Reviewer.

13

You might also like