You are viewing an old version of this page. View the current version.

Compare with Current View Page History

« Previous Version 4 Next »

  1. Although the general structure of the Trial Visits (TV) dataset is "One Record per Planned Visit per Arm," for many clinical studies—particularly blinded clinical studies—the schedule of visits is the same for all arms, and the structure of the TV dataset will be "One Record per Planned Visit." If the schedule of visits is the same for all arms, ARMCD should be left blank for all records in the TV dataset. For studies with study visits that are different for different arms, ARMCD and ARM should be populated for all records. If some visits are the same for all arms, and some visits differ by arm, then ARMCD and ARM should be populated for all records, to ensure clarity, even though this will mean creating near-duplicate records for visits that are the same for all arms.

  2. A visit may start in one element and end in another. This means that a visit may start in one epoch and end in another. For example, if one of the activities planned for a visit is the administration of the first dose of study drug, the visit might start in the screen epoch and end in a treatment epoch.
  3. TVSTRL describes the scheduling of the visit and should reflect the wording in the protocol. In many studies, all visits are scheduled relative to the study's day 1 (RFSTDTC). In such studies, it is useful to include VISITDY, which is, in effect, a special case representation of TVSTRL.
  4. Note that there is a subtle difference between the following 2 examples. In the first case, if visit 3 were delayed for some reason, visit 4 would be unaffected. In the second case, a delay to visit 3 would result in visit 4 being delayed as well.
    1. Case 1: Visit 3 starts 2 weeks after RFSTDTC. Visit 4 starts 4 weeks after RFSTDTC.
    2. Case 2: Visit 3 starts 2 weeks after RFSTDTC. Visit 4 starts 2 weeks after visit 3.
  5. Many protocols do not give any information about visit ends because visits are assumed to end on the same day they start. In such a case, TVENRL may be left blank to indicate that the visit ends on the same day it starts. Care should be taken to assure that this is appropriate; common practice may be to record data collected over more than 1 day as occurring within a single visit. Screening visits may be particularly prone to collection of data over multiple days. The examples for this domain show how TVENRL could be populated.
  6. The values of VISITNUM in the TV dataset are the valid values of VISITNUM for planned visits. Any values of VISITNUM that appear in subject-level datasets that are not in the TV dataset are assumed to correspond to unplanned visits. There will be a one-to-one relationship between values of VISIT and VISITNUM in TV. If a subject-level dataset includes both VISITNUM and VISIT, then records that include values of VISITNUM that appear in the TV dataset should also include the corresponding values of VISIT from the TV dataset.


  • No labels