Course Content
Data Entry, Validation, and Access Control
This chapter explores the operational aspects of data collection within REDCap, including data entry workflows, validation procedures, user rights management, audit trails, and quality assurance activities that help ensure research data remain reliable and compliant with regulatory standards.
0/12
Data Quality Management and Query Resolution
This topic is about ensuring that data is accurate, complete, and consistent. It covers identifying, investigating, and correcting errors or discrepancies in data. It also focuses on timely resolution of data queries to maintain data integrity and reliability.
0/13
Introduction to R for Clinical Data Management.
This topic introduces R, an open-source programming language used for statistical analysis, data management, and visualization. In clinical research, R is applied to clean, analyze, and report data in a reproducible and reliable manner.
0/1
Clinical Research Data Management Course

Database freeze and database lock are important milestones near the end of a study or analysis period. The terms may vary by institution, but generally a database freeze is a temporary restriction or snapshot used to support review, interim analysis, or final cleaning. Database lock is the formal point at which data are considered final for analysis and no further changes should occur without a controlled unlock process.
Before database freeze, the data manager should ensure that expected data are present, key checks have been run, major discrepancies have been queried, and outstanding issues are documented. A freeze may be used to allow statisticians to prepare preliminary analysis while final queries are resolved. It should be clear which dataset version is frozen, when it was exported, what exclusions or unresolved issues remain, and who approved the freeze.
Before database lock, the requirements are stricter. All critical queries should be resolved or documented. Primary outcome data should be complete or missingness explained. External data should be reconciled. Serious adverse events should be reconciled with safety databases or logs. Protocol deviations should be reviewed. Coding should be finalized. Audit trail concerns should be addressed. The data dictionary and metadata should be complete.
The statistician should confirm that the dataset is ready for final analysis.
Lock should be a controlled decision involving the data manager, principal investigator, statistician, monitor or sponsor where applicable, and sometimes quality assurance. The lock decision should be documented. After lock, changes should be rare and should require formal approval, documentation of reason, database unlock, correction, relock, and version update.
This protects the credibility of final analyses. Database lock is not only a technical action. It is a declaration that the study team has completed required data management activities and that the dataset is fit for final analysis. A rushed lock can lead to analysis delays, rework, and credibility concerns. A well-planned lock reflects continuous quality management throughout the study.