The sponsor should ensure integrity and confidentiality of data; apply QC to relevant data handling stages and focus QA and QC, including data review, on higher-criticality data and metadata; pre-specify in the protocol what data are collected and how, with detail such as a data flow diagram in documents like a data management plan; validate fit-for-purpose data acquisition tools before use; implement documented processes for data integrity across the life cycle; safeguard blinding during entry and processing and describe unblinding (who, when, why, who stays blinded, safeguards); guide sites, providers and participants on data capture, changes, retention and disposal; not change investigator or participant data unless justified, agreed in advance and documented, while allowing justified, source-supported corrections; give investigators timely access to protocol data including external data, without sharing unblinding information; not hold exclusive control of data in acquisition tools; ensure investigators can access data for retention and know how to navigate systems; seek investigator endorsement at set milestones; define and document pre-analysis data management steps for each analysis; manage access during interim analysis and restrict edit access before final analysis and unblinding; use unambiguous participant codes; protect privacy and confidentiality; describe handling of data on withdrawal; protect data from unauthorised access, alteration, destruction or loss; and have processes to report incidents, including security breaches, with significant impact on trial data to relevant parties including regulators.
The graph holds this control, the 0 it maps to, and the evidence behind each claim, over MCP and REST.