This is a ready-to-use log for periodic reconciliation between the clinical systems that each hold a piece of the same subject-level event: EDC, IRT/RTSM, and CTMS. It covers drug accountability, enrollment and randomization, the query and correction trail, and emergency unblinding. It does not cover safety case reconciliation between EDC and the pharmacovigilance safety database; use the pharmacovigilance case reconciliation log for that. Replace every <<FILL: ...>> placeholder, run it on a fixed calendar, and retain the sign-off even when nothing is found. This content is educational reference, not legal or regulatory advice.
Field definitions and instructions
Retain each reconciliation record for <<FILL: retention period, per your records retention schedule; see the clinical system retirement guidance for the applicable regulatory floor>>.
Reconciliation record
A. Drug accountability: IRT dispensing vs. EDC dosing vs. site accountability log
B. Enrollment and randomization: IRT randomization log vs. EDC subject status vs. CTMS enrollment tracker
C. Query and correction trail: EDC query resolution vs. underlying source or monitoring visit report
D. Emergency unblinding events: IRT unblinding log vs. statistician / safety notification record
Acceptance criteria
- Every value compared is pulled directly from its source system for the period, never from a memory, dashboard summary, or prior reconciliation carried forward.
- Every mismatch is investigated to a documented root cause; “close enough” or a count-only comparison without record-level matching is not an acceptable closure.
- Zero-mismatch reconciliations are still documented and signed; the absence of a record is itself a finding.
- A full reconciliation across all four sections is completed and closed, with zero open and unexplained discrepancies, before database lock.
- The reconciliation runs on the defined calendar (recommended: monthly during active enrollment, plus before every interim analysis and before database lock); a missed or late reconciliation is logged as a deviation.
Signoff
References
ICH E6 Good Clinical Practice, sponsor oversight and record-keeping expectations.
21 CFR 312.62, investigator recordkeeping and record retention.
Company data management plan and reconciliation specification for the study.
Confirm the current version of each reference before issue.
Filled specimen
A completed drug accountability reconciliation excerpt for an example ongoing oncology study. Illustrative only.
Neither discrepancy here was a patient-safety event, but both were caught within two weeks of occurrence, while the source records and site memory were still fresh, rather than surfacing for the first time at database lock or during an inspection.
Common inspection findings this log prevents
- No documented reconciliation between EDC, IRT, and CTMS for an ongoing study; the first comparison happens at database lock, when the source records are stale.
- A discrepancy is noted informally in an email thread but never logged, investigated to root cause, or closed with a disposition.
- Reconciliation limited to totals (a count that matches) rather than subject-level matching, which can hide one subject missing and one subject double-counted.
- Zero-discrepancy periods have no record at all, so the reconciliation cannot be shown to have happened on schedule.
How to adapt this log
- Set the study, systems, and reconciliation type in the header.
- Add or remove sections to match which of the four reconciliation types apply to your program; not every study runs emergency unblinding events, for example.
- Run this on a fixed calendar with a named owner, and retain zero-discrepancy records the same as records with findings.
- For safety case reconciliation between EDC and the pharmacovigilance safety database, use the pharmacovigilance case reconciliation log instead; keep the two logs cross-referenced in your data management plan so nothing falls between them.
