Glossary · Data and documentsUpdated September 28, 2026

What is database lock in a clinical trial?

Database lock is the point at which a trial database is closed to further changes, after data cleaning is complete, so the final analysis can be performed on a fixed dataset. In blinded trials, unblinding follows lock.

  • Database lock in plain language
  • How it works in practice
  • Related terms

Free sandbox · No credit card · 21 CFR Part 11 aligned

Pre-lock checklist · Study ABC-201

Complete

7 / 9

Open queries

0

Target lock

14 Oct

  • All expected data entered
  • All queries closed
  • SDV complete on critical fields
  • Medical coding complete
  • SAE reconciliation with safety database
  • External data reconciled
  • Investigator sign-off on casebooks
  • Protocol deviations classified
  • Lock approval signatures
Lock, then unblind

In short

  • Before lock, all data is entered, queries are closed, coding and SAE reconciliation are complete, external data (such as central labs) is reconciled, and investigators sign off.
  • A soft lock freezes data for final review while allowing controlled changes; a hard lock closes the database for analysis.
  • Interim locks support planned interim analyses, often on a subset of data.
  • Unlocking after lock is possible but must be justified, approved and documented, because it undermines confidence in the fixed dataset.
  • The time from last participant last visit to lock is a common data management performance measure.

Explained

Why lock matters

Lock is the line between collecting data and analysing it. Once the database is locked and, in a blinded trial, treatment codes are released, any later change to the data invites the question of whether it was influenced by knowing the results. That is why lock is preceded by a formal checklist and approvals, and why unlocking requires a documented reason.

Most of the work happens long before lock. Studies that resolve queries as they arise, verify critical data continuously and reconcile safety data monthly reach lock in days; studies that leave cleaning to the end can take months. Good edit checks and risk-based monitoring shorten the path.

In Capture

Freeze and lock at field, form or subject level

Data managers freeze or lock a single field, a form or a whole set of subjects, in bulk. Unfreezing and unlocking need a reason from a set list, and every action lands in the audit trail. The status overview shows what is complete, missing or overdue, so the path to lock is visible.

  • Bulk freeze and lock.
  • Reason required to unfreeze or unlock.
  • Query, SDV, freeze and lock status on every field.
Clinical data management software
Status overview · subjects by visit
SCRD1W2W4W8W12EOS
01-001
01-002
01-003
01-004
01-005
01-006
Complete Partial Missing Overdue Not done Screen-failed

FAQ

Common questions

Something not covered here? Ask us directly.

What happens at database lock?

The database is closed to changes after cleaning, coding and reconciliation are complete, so the final analysis runs on a fixed dataset.

What is the difference between soft lock and hard lock?

A soft lock freezes data for final review with controlled changes still possible; a hard lock closes the database for analysis.

Can a locked database be unlocked?

Yes, but only with a documented justification and approval, and the unlock and any changes are recorded in the audit trail.

How long does database lock take?

It depends on how much cleaning is left at the end. Continuous cleaning during the study can shorten the time from last visit to lock to days.

Get to lock faster

Continuous cleaning, bulk freeze and lock. Free sandbox.

Try Capture free