Roles and access
Access is grant-based and scoped to a study, and optionally to a single site within it. A user has no standing authority anywhere until a grant gives it to them for a specific study.
The roles
Section titled “The roles”| Role | Can |
|---|---|
| Lab admin | Everything within a study, including granting roles. |
| Lab manager | Manage the study; verify and sign results; review the audit trail. |
| Technician | Accession and store samples; order tests; enter results. |
| Accessioner | Accession and store samples only. |
| Monitor | Review audit trails and custody records (external reviewer). |
| Read-only | Read access through study membership; no actions. |
Separation of duties
Section titled “Separation of duties”Two separations are enforced, not conventional:
- Enterer ≠ verifier. The four-eyes rule on results means the person who enters a result cannot verify it. This is checked by the system, not left to procedure — see orders and results.
- Admin ≠ lab authority. System administrators, who run the platform, hold no laboratory authority. Administering the software and signing off on lab work are deliberately separate powers; the same person may hold both, but through two distinct grants, and the audit trail reflects which one they were acting under.
This is why access checks are authority-scoped rather than a simple admin/non-admin split (requirement P11-04): the question is never “are you an admin?” but “do you hold this specific authority in this study?”
The same pattern governs the newer workflows. Terminal
disposal of a specimen is a supervisor-only authority,
separate from ordinary sample handling. The lab-wide
reagent inventory is authorized on an inventory.manage
authority held in any study, because reagents are not study-scoped. And the
QC review board is deliberately read-only — it introduces
no new writes and no new authority, so reviewing trends can never alter a
record.