Digital triage is transforming healthcare, but evaluating its performance is far from straightforward. Different countries and studies use vastly different methods, making it hard to know what 'good' really looks like - or whether the numbers being quoted are even comparable in the first place.

In this presentation, we explore the landscape of digital triage benchmarking: the hidden complexities, the surprising pitfalls, and why commonly quoted accuracy figures can be deeply misleading. 

Drawing on a deep dive into the published literature, we unpack the methodological challenges that make benchmarking in this space so difficult - from the absence of a gold standard metric, to the impact of study design, patient cohort, and how triage categories are defined.

Attendees will leave with a practical framework for interpreting triage performance claims critically - knowing what questions to ask, what to look for in a benchmarking study, and why the same number from two different products can mean something entirely different.