DDS Scope

[swhume]

What is the full scope of the final DDS?

The current scope for DDS is to represent the operational metadata for the entire research data pipeline. RWD, raw data, ODM metadata and data, DTAs, SDTM, ADaM, and ARDs are in scope, along with the logic needed to drive and trace transformations. Explicit semantics are referenced in the model. The model provides provenance tracking.

There is value in an operational model that drives automation throughout the pipeline. There are benefits to keeping the semantic and structural standards models separate from the operational model, even though they will be referenced in it.

There are other CDISC models that overlap this scope that CDISC leadership will review, such as the AC model and ARS standard, to determine how they co-exist. NOTE: 360i phase 2 will not test or review all aspects of the DDS model prior to the end of the year.

The DDE 360i Team will publish a draft DDS model at the end of the year. A roadmap will be published with the model that explains what has been tested, which parts have not been reviewed, and outlines plans to review the untested parts. The current version of the DDS model includes many external references that have not been tested, such as FHIR, PROV, OMOP, SKOS, qb, dprod, and numerous others.

Should the DDE code become primarily a mechanism to test, refine, and demonstrate the DDS model, given that it's very difficult to finalize a model of this complexity without implementing it?

The DDE code seeks to achieve two objectives: (1) as part of 360i to serve as a reference implementation for certain named use cases, and (2) to test and demonstrate the DDS model. The DDE code and an accompanying roadmap will be released at the end of the year along with the DDS model.