Eastern England Secure Data Environment and VISTA Early Adopter Project: Final Report

This report highlights how the VISTA project enhanced the capabilities of the Eastern England Secure Data Environment (EE‑SDE) by deploying and evaluating new tools and standards designed to make health data research safer, more transparent and scaleable. VISTA focused on three areas: helping SDEs demonstrate their operational integrity (through the SATRE standard), enhancing the way research outputs are checked before leaving the environment (using the SACRO semi‑automated disclosure‑control tools), and ensuring that machine‑learning models trained on sensitive data can be assessed safely before release (through SACRO‑ML). Rather than introducing separate processes, the project embedded these capabilities directly into existing EE‑SDE workflows. This helped teams understand how automation can support, but not replace, expert judgement and strong governance. The work also highlighted the importance of clear communication, consistent training and well‑structured documentation, particularly as AI becomes more common in research. This report summarises practical improvements identified during the project, including clearer assurance processes, more consistent output‑checking pathways and a structured approach to assessing AI‑related risks. It also identifies areas for further development, such as reducing technical friction, improving guidance for new users and refining national standards as TREs mature. Overall, VISTA has been a highly constructive learning exercise that has strengthened the EE-SDE, contributed directly to national TRE standards and built a foundation for future AI-ready, privacy preserving research.