The PHUSE Real World Data Autumn Event returned for its second multi-day edition, uniting global experts to share insights, practical approaches and applications of real-world data (RWD).
Held from 30 September to 1 October, the event brought together attendees from 81 companies and 25 countries.
Over the two days, sessions explored how quality management, standards and regulatory frameworks can improve confidence, interoperability, traceability and regulatory readiness in RWD. Speakers highlighted reproducible validation, auditable decision-making and practical ways to manage complexity and generate trustworthy real-world evidence. One attendee commented that “learning about different approaches to the same problems I face was enriching. All presentations stood out… it was a fantastic session.”
Presentation slides and recordings are available on the PHUSE website.
Real World Data Autumn Event Day 1: Real World Data Quality and Lineage
Berber Snoeijer – Assessing Quality and Usability of Real-World Data Sources
Berber Snoeijer outlined the PHUSE Real World Evidence Working Group’s work on assessing RWD quality, reusability and regulatory suitability. The presentation covered evolving FDA and EMA guidance, fit-for-purpose assessments, DPRFIs, anonymisation, data lineage and NLP governance, while highlighting the group’s white paper and new data-lineage project.
Sowmya Lakshmi Gudimella – Beyond Standard Checks: A Reusable Data Quality Validation Framework for OMOP CDM in Real-World Evidence Studies
Sowmya Lakshmi Gudimella introduced a reusable OMOP Common Data Model validation framework covering structural, vocabulary, temporal and cross-table checks. Applying consistent assessments across studies improves confidence in standardised healthcare data, reduces repetitive quality control, accelerates study start-up and supports more reliable real-world evidence.
Yogesh Kumar Gupta – RWD Signal Integrity Modelling: A Framework for Evaluating the ‘Health’ of Real-World Data Before Use
Yogesh Kumar Gupta proposed a framework for evaluating the health of real-world datasets before use. Its Signal Integrity Score combines pre-analysis integrity, temporal shifts, missingness, bias and source reliability, enabling structured dataset comparisons, earlier risk identification and more proactive selection for research and regulatory use.
Zhimin Liu – A Knowledge Graph Foundation for Real-World Data Sourcing, Lineage and Evidence Traceability
Zhimin Liu showed how knowledge graphs and AI-assisted metadata extraction connect research questions, datasets, publications and evidence. Preserving provenance, version history and decision rationale improves traceability, supports data reuse, reduces duplicated effort and strengthens future evidence generation and regulatory scrutiny.
Real World Data Autumn Event Day 2: Real World Data Standards and Regulations
Kieran Mace – Same Analysis, Many Sources
Kieran Mace compared four approaches for harmonising real-world evidence analyses across multiple sources: common data models, source-specific code, configuration-based mappings and canonical data stores. Each involves trade-offs in terminology, interpretation and timing. Organisations should select the approach that best supports their analysis volume and traceability needs.
Seong Jun Byun – Implementing ICH M14 in Fragmented Claims-EMR Settings
Seong Jun Byun set out a four-domain framework for implementing ICH M14 across fragmented claims, electronic medical record and registry data. Covering source suitability, feasibility, comparator selection and validation, it emphasises transparent documentation of assumptions, decisions and limitations so that evidence can withstand regulatory review.
John Owen & Berber Snoeijer – From Criteria to Connection: Enabling Diverse Participation with USDM Through READI
John Owen and Berber Snoeijer showed how READI uses USDM to connect digitised study criteria with electronic health records, helping identify eligible patients from underrepresented populations. Standardised study definitions improve patient matching, streamline research and support interoperable digital data flows across European clinical research.
Miten Mistry & Abhishek Dabral – Operationalizing CDISC Biomedical Concepts as the Semantic Layer of an eSource Platform
Miten Mistry and Abhishek Dabral demonstrated how CDISC Biomedical Concepts provide the semantic foundation for an eSource platform. Their metadata-driven approach enables machine-readable forms, automated EHR-to-EDC transfer and AI-assisted extraction, improving consistency, scalability, interoperability and standards-based regulatory outputs.
Taken together, the eight presentations showed how robust quality assessment, shared standards, transparent governance and interoperable technologies can strengthen RWD. The programme progressed from assessing data fitness to demonstrating how data can be connected, reused and translated into credible evidence. Ultimately, trustworthy results depend on consistent methods, traceable decisions and collaboration across the research ecosystem.
Real World Evidence Pavilion: 14–18 March 2027
Continue the conversation at the in-person Real-World Evidence Pavilion!
Taking place at the US Connect 2027, the Real-World Evidence Pavilion will be a space for attendees to meet with regulators, sponsors, technology leaders and patient advocates to explore the latest challenges and developments in data transparency.
Review your ticket options and register by 23 October to enjoy the Early Bird discount!
Become a Pavilion Sponsor!
Sponsorship packages are available for the Real-World Evidence Pavilion, for organisations to showcase their expertise, raise their profiles and connect with key decision-makers across the real-world evidence community.
Download the prospectus for more information on available packages.
