Operationalizing Data Quality through Data Governance - Data Governance: Creating Value from Information Assets - page 78

and how can it help a clinical trial study team develop protocol, CRF, etc.,
in a pharmaceutical company?
To ensure clinical data quality and business process efficiency, a meta-
data-driven method is a necessary methodology in clinical data processing.
Using the metadata-driven method will allow all components to share infor-
mation about the data as it moves through its life cycle, thereby enabling
consistency, accountability, and true control of data. It also will increase the
capability to share and manage data within and across organizations and
reduce the impacts on the safety, effectiveness, and cost of healthcare by
having the right information at the right place at the right time. Using an
example of creating a study analysis data, we can see a tremendous waste
of time and resources when not implementing a metadata-driven process.
There are three processes in a clinical trial analysis data derivation: docu-
mentation, SAS programming, and QC (quality control) processes. Without
using metadata-driven methodology, variable name, variable label, vari-
able format, variable length, and variable type will be manually typed in
each process. Assuming there are 30 SDTM (study data tabulation model)
datasets and 20 ADaM (analysis dataset model) datasets for each study,
30 variables per data set, 5 attributes per variable, plus about 500 elements
of controlled terminology terms, the total data attributes can be as many
as 5,000 per study. Without metadata to use, each attribute is manually
typed three times (private communication). All information about clini-
cal trial development can be driven by a standard-based, metadata-driven
architecture—metadata repository. When completely implemented, a
well-designed metadata repository can help industry to reduce time and
cost, increase efficiency and quality, while remaining compliant and
aligned with evolving regulatory and industry initiative in the product life
cycle from clinical data collection, data processing, data analyzing, report-
ing data, and postmarketing research. Therefore, metadata and a metadata
repository can help us to reach interoperability while ensuring data quality.
The metadata repository is the origination point for semantic changes
and it manages metadata as an asset. Imagine every new development
project starting with the metadata repository to determine what data cur-
rently exists, which can be reused, who owns the data, how comprehen-
sive it is, what other processes affect it, and where it is currently reported.
A well-designed metadata repository is like a high-quality engine in a
vehicle; however, to specify a metadata repository requirement is a com-
plicated topic, it varies by the business goals and objectives. Based on