What the engagement delivered, in numbers.
The first service in its health system to use AI for vaccination cohort grouping, live in the operational estate.
Cohorts identified across the whole population in a fraction of the previous manual time.
An accuracy threshold agreed with clinical governance and met before go-live; explainability built into every output so a clinician can see why.
A time-bound specialist team on a client-owned platform, with exit criteria tied to the clinical-governance thresholds.
A national task done by hand.
Deciding who is eligible for a vaccination programme, and when they should be called, meant working through population-scale data with rules that change by programme and by year. It took time, relied on a small number of experienced people, and individual decisions were hard to evidence afterwards.
What this shows
AI in health earns its place by being explainable and by meeting the bar clinicians set.
A time-bound specialist team on a client-owned platform is the right shape when the capability has to stay with the organisation.
The pattern, supervised learning with explainability and a governance threshold agreed up front, transfers to any programme where the rules are known and the stakes are high.
“Identifying who to call for vaccination, and in what order, was a national task done largely by hand. We agreed the accuracy bar with clinical governance before anything was built, and the service met it before it went live. Every output can be explained to a clinician, and the decision still sits with a person.”
Head of Data
Talk to us about your organisation.
Tell us where your organisation is starting from.