Clinical Decision Support in 2024: A broad and fast-evolving field offers insights into common success factors
Clinical Decision Support (CDS) is revolutionizing patient care, by unlocking the power of patient data to:
Identify at-risk patients for screening assessments
Accelerate and improve accuracy of diagnosis techniques and services
Improve treatment decision-making (for integrated therapy + digital protocols)
Optimize patient experience & outcomes based on patient needs and monitoring data
This is an extremely dynamic area, with many different players including Health Systems, Pharma, MedTech and various flavours of HealthTech. In this article I recap some of the recent news has caught my eye, highlighting the direction of travel for CDS, and providing insights & inspiration for those interested in this space.
To begin
A recent Nature publication provides a good 101 for those who are new to CDS, or anyone interested in a handy recap: Advancing clinical decision support: The role of artificial intelligence across six domains — ScienceDirect
News demonstrates the wide range of activity, and gives insights into common success factors
Whilst there is lots of untapped potential in CDS, there is also plenty happening right now. These examples show just how wide that range is, ith cases across the patient journey (diagnosis to treatment), a range of therapy areas, and global geographies (including Europe, US, and Asia). There are pragmatic cases (e.g. predicting response to immunotherapy based on just two readily-available datapoints) as well as (of course!) AI-based solutions.
The examples that are reaching the scale-up stage highlight some common success factors — Health system co-creation, grounding in pragmatic health system needs (automation, reducing burden), early demonstration of value, and leveraging pre-existing scaling pathways.
AI-based decision support to optimize complex care for preventing medication-related falls
Brainomix 360 platform helping decision making in stroke endorsed by NICE
AI-based decision support boost post-stroke prevention
Investigation on explainable machine learning models to predict CKD prognosis
Optimizing warfarin dosing for patients with atrial fibrillation using machine learning
UC San Diego Risk Stratification AI Predicts Sepsis, Reduces Mortality
GE Healthcare leverages routinely collected clinical data to predict response to immunotherapy
Lenus moves to scale-up stage for diagnostic automation and CDS product in heart failure
Coming Up
These cases highlight the increasing technical feasibility of CDS. Additional barriers to overcome in scaling CDS impact include at-scale distribution to large populations across diverse IT infrastructure, and ensuring HCPs adopt the solution into their workflow. Q1 has offered some early insights here too, which I will cover in a separate Medium article or LinkedIn post.
