On 11 February, the SafetyNet Medication Safety working group held an event to explore the potential of clinical decision support (CDS) in enhancing medication safety within electronic health systems. Hosted by the NIHR Central London PSRC, the event brought together experts from across the PSRCs, as well as commercial partners and policymakers. Presentations highlighted projects advancing CDS across various settings, showcasing innovative approaches to improve patient outcomes and address biases in healthcare data.

Chaired by Dr Yogini Jani, a panel – Professors Avery, Foy and Franklin – set the scene with the origins and early experience of clinical decision support in primary and secondary care electronic prescribing systems in England.

Commercial partners Jacob Robinson (Epic) and Polly Shepperdson (FDB Health) highlighted the increasingly sophisticated approaches using data science and artificial intelligence, enabling precision in decision support at patient as well as clinician level.

Professor Duncan Smith, UCL, described the evolving regulatory landscape, considering the intersection of medical device regulation with software and artificial intelligence and how this applies to clinical decision support in digital health technologies.

Rahul Singal, Chief Pharmacy and Medicines Information Officer, NHS England, posited the key role of digital health systems, including clinical decision support in optimising medicines use and safety. He mapped areas of national health policies where the alignment of clinical decision support, digital health technologies and medicines management have potential to optimise patient outcome, enabling the use of effective treatments, with efficiency. 

A focussed session on the people perspective of CDSS featured a simulation of healthcare professional’s experience of multiple alerts from various digital systems by Serena Williams.  Drs Aoife Shields and Yogini Jani highlighted the importance of considering diversity in big data studies and machine learning to overcome historical biases and allow for the processing of complex, intersectional patient characteristics e.g., interplay between social, economic, cultural and ethnic backgrounds, to identify subtle patterns in treatment requirements and to enable future learning from diverse populations.

NIHR SafetyNet researchers rounded the day with rapid presentations of current projects that are advancing the field, showcasing the use of clinical decision support across a range of settings and topics.

The event concluded with a workshop to identify priority areas that the NIHR Patient Safety Research Collaborations can collectively progress in collaboration with commercial partners to generate evidence for emerging advanced clinical decision support, underpinned by artificial intelligence, to inform national and international practice and policy.

Related