The General Observation Classes

The majority of observations collected during a study can be divided among 3 general observation classes: Interventions, Events, and Findings.

• The Interventions Observation Class represents investigational, therapeutic, and other treatments and products that are administered to or used by a subject (with some actual or expected physiological effect). This includes treatments and products specified by the study protocol (i.e., "exposure").
• The Events Observation Class represents planned protocol milestones such as randomization and study completion, and occurrences, conditions, or incidents independent of planned study evaluations occurring during the study (e.g., adverse events) or prior to the study (e.g., medical history).
• The Findings Observation Class represents observations resulting from planned evaluations to address specific tests or questions such as laboratory tests, electrocardiogram (ECG) testing, and questions listed on questionnaires. The Findings class also includes a subtype, Findings About, which is used to record findings related to observations in the Interventions or Events classes.

Datasets based on any of the general observation classes share a set of common identifier and timing variables (see Section 3.1.4, Identifiers for All Classes, and Section 3.1.5, Timing Variables for All Classes). As a general rule, any valid identifier or timing variable is permissible for use in any dataset based on a general observation class.

The SDTM is the foundation for many implementation guides. Implementation guides include particular domains based on the general observation classes. Domain specification tables describe which variables are included in the domain for a specific implementation guide. Not all variables described in the tables in this document (SDTM tables) are appropriate for all domains in all implementation guides. The SDTM includes information on variables which are not to be used in human clinical trials. Refer to the implementation guides for specific information on any identifier or timing variables that are not allowed in a particular observation class domain. See Section 1.2, Implementation Advice for this Model.

In the tables in this document, the presence of 2 hyphens before the variable name (e.g., --TRT) is used to indicate the required use of a prefix based on the 2-character domain code. The domain code is used as a variable prefix to minimize the risk of difficulty when merging or joining domains for reporting purposes. The variable with 2 hyphens may be referred to as a "root" variable.

Domain-specific variables, a concept introduced in SDTM v1.5, are for use in a limited number of designated domains based on general observation classes and will be identified in the appropriate implementation guide. The variable names include the specific domain prefix. Domain-specific variables are included in the table of general observation class variables; the Usage Restrictions column of the table indicates the domains in which the variable is allowed.

In addition to the 3 general observation classes, a study will generally include a set of other special-purpose datasets; see Section 3.2, Special-purpose Domains.