Clinical Data Hierarchies
For ODM Version 2, there are two forms of the clinical data hierarchy. The full hierarchy is typical of usage within data collection tools such as EDC systems. The dataset hierarchy reflects data tabulation formats that are typical of usage for data review and submission.
The ODM allows for both internal and external keys. Internal keys, known as OIDs, are used to designate entities within the model, and to allow cross-references between entities within and between ODM files.
Internal keys are not optimized for human use but are assumed to be unchanging.
External keys are any keys used by clinical personnel. They are typically represented in ODM metadata by an element Name attribute or a Description child element. They are intended to be useful for data review and interpretation.
The ODM allows for both internal and external keys. Internal keys, known as OIDs, are used to designate entities within the model, and to allow cross-references between entities within and between ODM files. Internal keys are not optimized for human use but are assumed to be unchanging. External keys are typically represented in ODM metadata by Name attributes or Description child elements. They are intended to be useful for data review and interpretation.
| Kind of Entity | Identifying Keys | Required for Subject data | Required for Dataset data |
| study | StudyOID | X | |
| subject | above plus SubjectKey | X | |
| study event | above plus StudyEventOID and StudyEventRepeatKey | X | |
| item group | above plus ItemGroupOID and ItemGroupRepeatKey | X | X |
| item | above plus ItemOID | X | X |
| annotation | keys for the annotated entity plus SeqNum | X | X |
A StudyOID uniquely identifies a study. A SubjectKey uniquely identifies a subject within a study. SubjectKeys are not intended to be used to identify a subject across studies.
A StudyEventOID uniquely identifies a StudyEventDef within a study. However there may be several study events of a particular type for a given subject. Thus, to fully identify a particular study event, we need the StudyOID, the SubjectKey, the StudyEventOID, and a StudyEventRepeatKey.
Collected data always belongs to an ItemGroup but either form of the hierarchy may be used within a Clinical Data element.
Each item group in clinical data can be identified by an ItemGroupOID which identifies its definition and an ItemGroupRepeatKey which identifies a data collection instance.
Within the ODM ReferenceData element, ItemGroupData does not relate to a study subject.
Each item belongs to an item group. Beginning with ODM Version 2.0, Item Groups can contain other item groups, Within an ItemGroupData instance, however, there may be only one instance of an ItemOID. That is, only the ItemOID is needed to uniquely identify an item within its item group.
For clinical data collected in support of submissions to regulatory authorities where CDISC Data Standards are required, the CDISC SDTM and SDTMIG standards detail the requirements for variables used to identify study subjects, study events and timing and other data elements used in processing the clinical data.