As a major Research Hospital, UCLH is developing ways to utilise digital tools to enhance the way clinical research is conducted. Underpinning this aspiration is structured clinical data which is recorded in a standards-compliant manner.
In April 2021, UCLH took the strategic decision to transition away from a proprietary diagnosis list and implemented SNOMED-CT – a structured clinical vocabulary – for diagnosis recording in our electronic health record system, Epic. SNOMED CT gives clinical IT systems a single shared language, which makes exchanging information between systems easier, safer and more accurate. UCLH has become the first English-speaking Epic site in the world to implement SNOMED-CT, by deploying a novel methodology funded by the BRC.
Use of SNOMED-CT will facilitate multicentre clinical trials, particularly trials embedded in electronic health record systems. Any data-enabled trials at UCLH will be easy to scale to other UK sites and international partners.
The transition to SNOMED-CT has immediately enabled UCLH to build a proof-of-concept interface between our electronic health records and Keytrials, a locally developed clinical trials discovery platform that presents UCLH clinicians with an intuitive, searchable list of trials enabling clinicians to match patients to potential clinical trials.
This has particular value for an NHS Trust such as UCLH which has a large number of patients with rare diseases. It also aligns well with our strategies to empower patients and their care-givers through the democratisation of data and knowledge.
Adopting an interoperable data format has also paved the way for UCLH to act as a pilot site to evaluate an EHR-to-EDC (electronic data capture) solution with a large commercial trial sponsor, to significantly streamline the data management process in clinical trials. Several future Epic sites in the UK are utilising UCLH’s work in data harmonisation.
We predict that by 2025 potentially up to 20% of the UK population will be covered by SNOMED-CT diagnosis, driving a national step change in opportunities for large clinical trials.
The value of data-enabled trials
Using an evidence-based medicine approach to determine optimal treatment decisions has become the accepted gold standard for clinical practice. However, many treatments routinely administered in hospital lack a strong evidence base as randomised controlled trials (RCTs) are expensive to conduct and have a long lag time to effecting change in clinical practice.
Further methodological advances designed to improve trial efficiency have led to the increased use of Randomised Embedded Multifactorial Adaptive Platform (REMAP) study designs. In these trials, data collection is embedded within the electronic health record system – leading to improved efficiency, together with the use of adaptive randomisation to optimise treatment allocation.
Multiple treatments may be evaluated concurrently, making these trials highly pragmatic in nature. Finally, multiple research questions are evaluated using the same infrastructure (Platform), reducing costs.
Within the PROSPECTOR study, UCLH researchers are studying the feasibility of conducting electronic point-of-care randomisation through an Epic alert directly focused on clinical decision makers.
Where clinicians have equipoise for a given decision, they may choose to follow the randomised treatment allocated. Equally, where they lack equipoise, they continue to follow their preferred course, but instead of representing a protocol violation under a standard RCT design, they continue to collect data on the patient characteristics which may have influenced the lack of equipoise.
This information can then contribute to future research cycles by allowing increasing scrutiny of patient sub-groups likely to derive benefit. PROSPECTOR will study whether magnesium supplementation will prevent new-onset Atrial Fibrillation (AF) within critical care setting.
