Benefits and Challenges of Deterministic, Probabilistic, and Patient-Mediated Approaches to Linking Patients’ Clinical Trial Data (CTD) and Their Real-World Data (RWD)
Author(s)
Zabell T1, Jeswani N1, Hahn K2, Coutcher J2, Papsch R1
1IQVIA, London, LON, UK, 2IQVIA, Cambridge, USA
Presentation Documents
OBJECTIVES: Whilst depth of CTD is unparalleled, RWD can expand our view multi-dimensionally (over time and across settings) to build richer understanding of drug value. Linkage of patients’ CTD and RWD can provide insights on disease / treatment journey before, during, and after trials and across primary care, pharmacies, emergency, specialist tertiary care centers, and home settings. We aimed to understand different methodologies for bridging CTD and RWD, characterising use cases, strengths, weaknesses, and operational considerations.
METHODS: We conducted a pragmatic literature review to identify precedent of CTD-RWD linkages, and RWD-RWD linkages with an eye toward models that could be adapted for CTD.
RESULTS: We identified three predominant approaches: (1) deterministic linkage, connecting CTD and RWD by unique patient identifiers (e.g., social security number); linkage rates depend on patient overlap between trial and RWD populations; (2) tokenisation, encrypting personally identifiable information to pseudonymise patient-level RWD into tokens from fields including name, DoB, and provider postcode, supporting probabilistic matching; linkage rates depend on trial-RWD population overlap, matching algorithm, and unique identifiability of data underlying tokens. Successful CTD-RWD linkage via both methods has been demonstrated in US, while emerging in Canada and Europe; (3) direct patient-mediated access to RWD, in which patients consent for third-party collection and integration of their CTD and RWD.
CONCLUSIONS: Benefits of CTD-RWD bridging hinge on working principles such as obtaining all required consent at trial enrolment without delaying trial operations, aligning use cases to linked cohort sizes achievable, and minimising site/patient burden. Tech-enabled solutions play a key role in overcoming challenges of linkage whilst protecting patients' data privacy. Top-down, standardised tokenisation engines on RWD can enable scalable linkage whilst privacy analytics help mitigate re-identification risk. Bottom-up, patient-mediated access empowers autonomy patient over their data in an increasingly digitalised world, and could exponentiate the value of insights through connecting patients' multi-dimensional data.
Conference/Value in Health Info
Value in Health, Volume 25, Issue 12S (December 2022)
Code
RWD140
Topic
Real World Data & Information Systems, Study Approaches
Topic Subcategory
Clinical Trials, Data Protection, Integrity, & Quality Assurance, Health & Insurance Records Systems, Reproducibility & Replicability
Disease
STA: Biologics & Biosimilars, STA: Genetic, Regenerative & Curative Therapies, STA: Medical Devices, STA: Personalized & Precision Medicine