.jpg){kind=link}
- Published
- Updated
Time to an event cannot be interpreted without a clear starting point. Randomisation, first administration and diagnosis dates establish different time origins. The selected origin should match the process the trial intends to compare between groups.
Define the question and assumptions
Define sources for origin and event dates, their precision and treatment of same-day events. Selecting a post-allocation origin can change the compared participant set. Review the protocol question, included people and elapsed-time definition together; do not move dates merely to make calculated durations positive or easier to display.
Hypothetical study example
In a hypothetical study, some participants first receive an intervention several days after randomisation. Starting analysis at administration can treat events in that interval differently. The team explains the interval within its scientific question and evaluates participant timelines under one consistent prespecified rule.
Design review checks
- Specify the origin source and whether the starting day enters duration calculations.
- Investigate events occurring on the origin date or apparently before it.
- Plan assumptions and sensitivity assessments for incomplete event or origin dates.
Record the decision and its limits
Acceptance examples cover delayed administration, same-day events and missing times. Derivations must be reproducible. If different time origins address different questions, present results with that distinction; do not retrospectively justify an origin because it produces a more favourable treatment comparison in the available dataset.
Frequently asked questions
- Is first administration always the correct origin?
- No. The intended comparison and included population determine the appropriate starting point.
- Is a negative duration automatically a data error?
- Not always. Review date sources and the event definition before deciding.