The timing of collection of observations is generally planned, but observations may also be collected outside a planned schedule. Additionally, the actual timing of observations is collected. Guidance in this section describes the expected population of variables to represent the planned. unplanned, and actual timing of observations. When guidance is applicable only to TIG Nonclinical or Product Impact on Individual Health use cases, this is denoted in the Implementation column.

Planned Timing

NumVariable PopulationImplementation
1Visits (VISIT, VISITNUM, VISITDY)

Product Impact on Individual Health only:

There will be a one-to-one relationship between values of VISIT and VISITNUM.

  • Values of VISIT, VISITNUM, and VISITDY will be represented as defined in the Trial Visits (TV) dataset. 
  • Values of VISITNUM are:
    • Expected when observations in a domain have been collected more than once per subject (e.g., vital signs in domain VS when collected for subjects over multiple visits)
    • Used for sorting and will, wherever possible, match the planned chronological order of visits
  • For visits that last more than 1 calendar day, VISITDY will be the planned day of the start of the visit.

 For unplanned visits, applicant practices for populating visit variables can vary:

  • VISITNUM will generally be populated, as it may be expected in a domain and will be populated when time points are represented. 
    • The easiest method of populating VISITNUM for unplanned visits is to assign the same value (e.g., 99) to all unplanned visits, although this method provides no differentiation between the unplanned visits and does not provide chronological sorting.
    • Methods that provide a one-to-one relationship between visits and values of VISITNUM, that are consistent across domains, and that assign VISITNUM values that sort chronologically require more work and must be applied after all of a subject's unplanned visits are known.
  • VISIT can be left null or may be populated with a generic value (e.g., "Unscheduled") for all unplanned visits, or individual values may be assigned to different unplanned visits.
  • VISITDY will not be populated for unplanned visits; VISITDY is, by definition, the planned study day of visit. 
2Nominal Study Day (--NOMDY, NOMLBL)

Nonclinical only:

  • Variable --NOMDY will be used to group records collected over multiple days under a single nominal study day for reporting purposes.
  • Variable --NOMLBL is a label for a given value of --NOMDY as presented in the study report (e.g., "Week 4", "Day 28", "Terminal Sac").
  • The value of --NOMLBL does not have to be unique for a given value of --NOMDY.
  • Variables --NOMDY and --NOMLBL will be used together (i.e., --NOMLBL will not be used without --NOMDY).
3--USCHFL

Nonclinical only:

  • Nonclinical studies are conducted in the context of a planned study schedule.
  • The unscheduled flag (--USCHFL) variable is populated when the timing of a performed test or observation was not driven by the study schedule.
  • In such cases, the value of --USCHFL will be "Y". Otherwise, the value of --USCHFL will be null.
4Epochs

Values in EPOCH reflect study design and will be populated for observations that start during a subject's study participation. Values of EPOCH are generally derived and will not be imputed. Values of EPOCH will be derived using collected dates. However, when this is not possible, EPOCH may be assigned per data collection sources (e.g., CRF instructions and structure). If it is not possible to determine the epoch of an observation, then the value of EPOCH will be null.

When EPOCH is included in:

  • a Findings class domain, its value be based on the --DTC variable, since this is the date/time of the test or, for tests performed on specimens, the date/time of specimen collection.
    • An unlikely exception may be a finding based on an interval specimen collection that started in one epoch but ended in another. In such cases, --ENDTC may be a more appropriate for EPOCH.
  • an Interventions or Events class domain, EPOCH will be based on the --STDTC variable, since this is the start of the intervention or event. 
5Variables for timepoints

Time points will be represented using both --TPT and --TPTNUM. There will be a one-to-one relationship between values of --TPT and --TPTNUM.

Actual Timing

In general-observation class domains, the dates and times (when applicable) observations were collected are represented in variable --DTC. In Interventions and Events domains, variable --STDTC is used to represent the start dates and times (when applicable) of interventions and events. In Findings class domains, the collection date is often aligned with when the source of the finding was collected (e.g., a specimen for laboratory testing) and may not be aligned with when the finding for the observation was generated (e.g., the date the specimen was assayed, and results were generated). Given this, for consistency, variable --DTC is always used in Findings domains to represent the start date/time of an observation and variable --STDTC is not used. Variable --ENDTC is used in all general-observation class domains to represent the end date and time (when applicable) of the observation. 

The length of time an observation continues (i.e., the duration of the observation) is represented using variables: 

  • --DTC or --STDTC for the start of the observation and --ENDTC for the end of the observation when date/times are collected
  • --DUR when only the duration of the observation is collected and, start and end dates have not been collected

« Free Text from Case Report Forms (CRFs)

Representing References and Relative Timing »

  • No labels