Abstract 1974P
Background
The relevance and utility of real world data to enhance clinical knowledge and support clinical decision-making is increasingly recognized. Electronic medical record systems are not well suited to the design, implementation and analysis of multi-site, real world studies. Clinician accessible solutions to enable distributed, real world clinical data collection and exploration are urgently needed.
Methods
MediGrid is a cloud-based solution for managing, storing and analysing clinical data. The solution uses ‘event sourcing’ enabling high quality, traceable and immutable data. The solution encourages collaboration between institutions supporting multi-centre studies. It uses a decentralized data visiting approach based on FAIR (Findable, Accessible, Intraoperable, Reusable) principles. Data is stored and managed locally and is shared via licencing. This leads to better data sharing based on open science, high quality data, strict authorization and regional compliance. A patient facing information entry application can be used to allow real-time, multi-source data entry.
Results
MediGrid was utilised to integrate & analyse real world patient data to better understand oncological immune-related adverse events (irAEs). The interface enabled efficient query, visualisation and analysis of irAE outcome data e.g. visualisation of incidence data, survival implications and heterogeneity in the presentation, complications and management of irAEs. Clinicians rapidly gained insights by comparing selected clinical outcomes across patient populations of interest e.g. irAE incidence in patients with autoimmune disease. The resulting data groupings, tabulations and visualisations were reviewed by collaborators via the platform. Results were exportable for use in other digital applications.
Conclusions
With the increasing recognition of the utility of real world data, effective and user-friendly interfaces are required to transform such data into clinically relevant and actionable knowledge. The MediGrid solution enabled clinicians to rapidly access, query, visualise and explore a large collection of real world data resulting in new and accessible insights into their patient populations.
Clinical trial identification
Editorial acknowledgement
Legal entity responsible for the study
The authors.
Funding
Roche.
Disclosure
A.C. Olsson-Brown: Honoraria (self): Roche; Research grant/Funding (institution): Roche; Honoraria (self): Bristol-Myers-Squibb; Research grant/Funding (self): Bristol-Myers-Squibb; Research grant/Funding (institution): Eli Lily; Research grant/Funding (institution): Novarits; Research grant/Funding (institution): UCB Pharma. J. Scheppers, H. Bossenbroek, M. van Mierloo: Full/Part-time employment: Luminis. M. Coles: Honoraria (self): Roche. J. Wagg: Full/Part-time employment: Roche.