This is a ready-to-use clinical database lock checklist. It confirms every pre-lock condition is met, records the evidence and the functional sign-offs, and defines the hard-lock steps and the unlock rule, so the dataset the biostatistician analyzes is fixed, traceable, and identical to what the clinical study report and submission report. Replace every <<FILL: ...>> placeholder, set your document numbers, and route through your normal review. A filled specimen follows. Confirm each cited regulation against the current source before you rely on it.
Document control header
| Field | Entry |
|---|---|
| Document title | Database Lock Checklist, Study <<FILL: protocol number>> |
| Document number | <<FILL: CL-ID, e.g. DBL-ABC201>> |
| Version | <<FILL: version, e.g. 1.0>> |
| Governing DMP | <<FILL: DMP-ID>> |
| Lock type | <<FILL: interim / final>> |
| Planned lock date | <<FILL: date>> |
1. Pre-lock conditions
Every item must be Complete (or explicitly Not Applicable with rationale) before the hard lock. A single open critical item blocks the lock.
| # | Condition | Status | Owner | Evidence |
|---|---|---|---|---|
| 1 | All expected CRF pages entered (expected vs received = 0 missing) | <<FILL: Complete/Open/NA>> | <<FILL: CDM>> | <<FILL: missing-pages report + date>> |
| 2 | All critical-variable queries closed | <<FILL>> | <<FILL: CDM>> | <<FILL: open-query report = 0>> |
| 3 | Remaining non-critical queries dispositioned/justified | <<FILL>> | <<FILL: CDM>> | <<FILL>> |
| 4 | Medical coding complete and medically reviewed | <<FILL>> | <<FILL: Coder/Medical>> | <<FILL: coding sign-off>> |
| 5 | Central laboratory data reconciled | <<FILL>> | <<FILL: CDM>> | <<FILL: recon report, 0 unresolved>> |
| 6 | ePRO/eCOA and IRT/IWRS data reconciled | <<FILL>> | <<FILL: CDM>> | <<FILL>> |
| 7 | SAE reconciliation (clinical vs safety database) complete | <<FILL>> | <<FILL: CDM/PV>> | <<FILL: recon memo signed>> |
| 8 | Protocol deviations reviewed and classified | <<FILL>> | <<FILL: Clinical>> | <<FILL: deviation log version>> |
| 9 | Data review / listings review complete | <<FILL>> | <<FILL: CDM>> | <<FILL>> |
| 10 | SDTM/ADaM conformance run, findings resolved or explained in the reviewer’s guide | <<FILL>> | <<FILL: Stat Programming>> | <<FILL: conformance report>> |
| 11 | Dictionary and CT versions match submission metadata | <<FILL>> | <<FILL>> | <<FILL>> |
| 12 | Audit trail reviewed for the period, risk-based | <<FILL>> | <<FILL: CDM/QA>> | <<FILL>> |
2. Functional sign-offs
The lock is authorized only when the defined functions sign.
| Function | Name | Signature | Date |
|---|---|---|---|
| CDM Lead | <<FILL>> | ||
| Biostatistics | <<FILL>> | ||
| Medical / Clinical | <<FILL>> | ||
| Quality Assurance | <<FILL>> |
3. Hard-lock steps
- Confirm all section 1 items Complete or justified and all section 2 sign-offs obtained.
- Revoke edit permissions in the production EDC (verify no account retains write access).
- Take the locked snapshot/export of the database.
- Record the lock in the EDC audit trail and complete the lock memo (date, database version, who locked, checklist reference).
- Generate the final extract for SDTM/ADaM and statistical analysis.
4. Lock memo
| Field | Entry |
|---|---|
| Lock date/time | <<FILL>> |
| Database version / snapshot ID | <<FILL>> |
| Locked by | <<FILL>> |
| Checklist reference | <<FILL: this document + version>> |
| Statement | All pre-lock conditions met or justified; functional sign-offs obtained; write access removed; locked snapshot retained. |
5. Unlock and re-lock control
If an error is found after lock, the database is unlocked only under formal change control: a documented request, an impact assessment, a tightly scoped change to the specific records, re-lock, and a record of exactly what changed and why. Uncontrolled or undocumented unlocks are a serious finding. See change-control-validated-systems and deviation-management.
6. References
ICH E6(R2)/E6(R3), Good Clinical Practice. 21 CFR Part 11; EudraLex Volume 4, Annex 11 (audit trail, controlled access, accurate copies). ICH E2A/E2B (safety), for the SAE reconciliation basis.
Confirm the current version of each reference before issue.
7. Revision history
| Version | Date | Author | Summary of change |
|---|---|---|---|
<<FILL: 1.0>> | <<FILL: date>> | <<FILL: author>> | Initial issue. |
Filled specimen (excerpt)
The following shows the pre-lock conditions completed for an illustrative final lock. Details are illustrative.
Study: ABC-201, final lock. Planned lock: 15-Mar-2026.
| # | Condition | Status | Owner | Evidence |
|---|---|---|---|---|
| 1 | All CRF pages entered | Complete | CDM | Missing-pages report, 0 missing, 14-Mar |
| 2 | Critical-variable queries closed | Complete | CDM | Open-query report = 0, 14-Mar |
| 4 | Coding complete and reviewed | Complete | Coder/Medical | Coding sign-off, 13-Mar |
| 5 | Central lab reconciled | Complete | CDM | Recon report, 0 unresolved, 14-Mar |
| 7 | SAE reconciliation complete | Complete | CDM/PV | Recon memo signed, 14-Mar |
| 8 | Deviations classified | Complete | Clinical | Deviation log v3, 13-Mar |
| 10 | SDTM/ADaM conformance | Complete | Stat Prog | Conformance report; residual findings explained in SDRG |
Lock memo: locked 15-Mar-2026 14:20, snapshot DBL-ABC201-FINAL, locked by CDM Lead, write access removed and verified. Two weeks later a lab-unit error was found in three records; the database was unlocked under CR-2026-088, the three records corrected, and the database re-locked with a documented scope, rather than edited ad hoc.
Common inspection findings this checklist prevents
- The database locked with open critical queries or incomplete reconciliation.
- SAE reconciliation gaps: serious events in the clinical database not matching the safety database at lock.
- Write access not actually removed, so the “locked” database was still editable.
- No lock memo, so the locked version and its approvers cannot be identified.
- An uncontrolled post-lock edit with no change-control record, scope, or re-lock.
How to adapt this checklist
- Add or remove conditions to match your data sources (drop rows that do not apply, mark them NA with a reason rather than deleting silently).
- Set the functions that must sign in section 2 to your study’s governance.
- Verify write-access removal technically, not just on paper, at the hard lock.
- Reference your real change-control procedure for the database unlock and re-lock path.
- Confirm the references in section 6 against the current published version before issue.