Longitudinal Projects and Events

Longitudinal projects are used when data are collected from the same participant or unit at multiple scheduled timepoints. REDCap implements this structure using events. An event represents a visit, phase, or timepoint, such as Screening, Enrollment, Day 7, Day 28, Month 6, or Close-out. Instruments can be assigned to events so that the same form can be completed at multiple timepoints. The advantage of events is that they organize repeated data without requiring separate variable names for every timepoint. For example, a vital signs form can be used at enrollment, day 7, and day 28. REDCap stores the event information alongside the record, making it possible to distinguish the same variable collected at different visits. This structure supports cleaner exports and more flexible analysis than creating variables such as `weight_day7`, `weight_day28`, and `weight_day90` for every repeated measurement, although some analysis workflows may later reshape data into wide format if needed. Designing events requires careful interpretation of the visit schedule. The data manager should identify all scheduled visits and determine which instruments belong to each event. Screening forms may appear only at screening. Enrollment forms may appear only at enrollment. Follow-up forms may appear at several visits. Adverse event forms may appear at many events or may be better designed as repeating instruments if the number of events is unpredictable.

Longitudinal design must also consider missed visits and unscheduled events. If the protocol allows unscheduled sick visits, the database may need an unscheduled event or a repeating instrument. If the study follows participants monthly for a year, events must be named and ordered clearly. The event structure should be understandable to coordinators, monitors, statisticians, and data managers. Event naming should be consistent and meaningful. Names such as "event_1" and "event_2" are less useful than "screening_arm_1" or "day_28_arm_1," although REDCap may create technical event names internally. Labels shown to users should be clear. Good event design reduces confusion during data entry and makes exports easier to interpret. [Figure 3.5: Screenshot of REDCap longitudinal project event setup and instrument-event mapping]

Table 3.7: Example Instrument-Event Mapping