Consent, screening, an eligibility decision, enrollment, then a schedule of visits that keeps itself up to date with the protocol.
The flow
The handoffs
| # | Who | Does | Result |
|---|---|---|---|
| 1 | Coordinator or investigator | Records consent (date and ICF version) and screens the participant (Screen a participant and record eligibility, Record consent and re-consent) | A screening record with eligibility Pending; the screening visits become available |
| 2 | Coordinator | Captures the screening assessments (Enter visit data) | The investigator can decide |
| 3 | Principal Investigator only | Records the eligibility determination | Eligible: the participant can be enrolled. Ineligible: screen failed on the spot, with the reason and the criterion |
| 4 | Coordinator or investigator | Enrolls, setting the baseline date (Enroll a participant) | The full visit schedule appears |
| 5 | Coordinator or investigator | Runs each visit inside its window (Run a visit) | Data capture |
| 6 | Investigator | Completes the participant, or withdraws them, or declares them lost to follow-up (Change a participant's status) | A final status; the data is kept |
The participant's transitions, from the code:
| From | Can move to |
|---|---|
| screening | enrolled, screen failed |
| enrolled | active, randomization failed, withdrawn, lost to follow up |
| active | completed, withdrawn, lost to follow up |
| screen failed | (none: final) |
| randomization failed | (none: final) |
| completed | (none: final) |
| withdrawn | (none: final) |
| lost to follow up | (none: final) |
A participant moves through eight states with two gates: consent is recorded before screening begins, and the investigator confirms eligibility before enrollment. Five states are final. Every final state keeps all the data captured, and a reason is required to enter four of them.
Consented and being assessed against the inclusion and exclusion criteria. A participant cannot be recorded without a consent date and an ICF version. Eligibility is pending, eligible or ineligible, and only the investigator decides it; recording an ineligible determination screen-fails the participant in the same act.
The investigator found the participant ineligible, with a coded reason. What was captured at screening is kept.
Terminal — no transitions out.
Eligible and enrolled; the visit schedule is generated from the baseline date. In a randomized study this is the state before randomization, where the arm allocated elsewhere is typed in.
Enrolled but never randomized (randomized studies only), with a reason.
Terminal — no transitions out.
On study and being followed.
Finished the study. Completion lists any forms still unsigned and needs an acknowledgement. Data can no longer be entered; a value is corrected only by answering a query.
Terminal — no transitions out.
Left the study by choice or by the investigator's decision, with a coded reason and a record of whether a safety follow-up is owed. The principal investigator makes this move.
Terminal — no transitions out.
The site can no longer reach the participant. An exit of its own, not a withdrawal: it needs a reason, what was tried and when, and a date.
Terminal — no transitions out.
The gates
No consent, no participant. A screening record needs a consent date and an ICF version; the consent date cannot be in the future or after the screening date. This is not a warning: the record does not come into existence.
Only the investigator decides eligibility. No coordinator, no data manager, no administrator. Recording ineligible screen-fails the candidate in the same act, attributed to the investigator: it is one decision, not two steps.
Enrollment has its own checks. The site must be active, the study must accept enrollment (not on hold, closed to enrollment, completed, locked or archived), and the baseline date cannot be in the future or before the participant consented.
Amendments need new consent. When an amendment asks for re-consent, a participant's answer is recorded and a participant who consents follows the new version; one who declines stays on the earlier version (Record consent and re-consent).
How the visit schedule behaves
The schedule is not frozen when someone is enrolled. It follows the design the participant is on: screening visits are created at screening, counted from the consent date; the rest at enrollment, counted from the baseline date; a visit added by an amendment appears on the participants who move to it. A participant still in screening sees only the screening visits; an enrolled participant sees the whole schedule. A visit has a window; one that is out of window is flagged, and one entered before its window needs a recorded early access with a reason.
Final does not mean deleted
All the ways out (screen failed, randomization failed, completed, withdrawn, lost to follow-up) keep everything captured. Four of them require a coded reason, and a withdrawal records the state of the safety follow-up. A completed participant's record is corrected only by answering a query. Nothing about a participant is ever removed.
Where this goes wrong
| Symptom | Cause |
|---|---|
| "I can't create the participant" | The consent date or the ICF version is missing, or a date is in the future |
| "Enroll is refused" | No eligibility determination yet, the determination was ineligible, the site is not active, or the baseline date is out of range |
| "Only two visits are showing" | Still in screening: the rest appear at enrollment |
| "The site can't enroll at all" | The site is registered, not active (Open and staff a site), or the study is not accepting enrollment |
| "The coordinator has no Mark eligible button" | By design: only the investigator records eligibility |
Do it
Screen a participant and record eligibility · Enroll a participant · Run a visit · Change a participant's status
Compliance Q&A
Yes. The system refuses to create a participant without the consent date and version, and keeps every consent and refusal against the version asked, with who recorded it and when.
Limits: the system records that consent was obtained and when; the signed consent form itself is kept by the site, and a consent form file can be filed in the document register.
Evidence: server subjects (consent date and ICF version required to screenconsent before screeningre-consent recorded per version)simulation of 20 Oct 2026, step 6.15
Yes. Only an investigator role can hold the permission, and the determination is attributed and timed.
Evidence: permission subject.eligibility is investigator-only (role editor and server)simulation of 20 Oct 2026 (coordinator and Project Manager have no control)