Substance Misuse Case Finding
Quickly identify potential service users that might never have sought treatment.
Rapidly increase caseloads in treatment and better utilise your capacity by saving time looking for more patients.
Expedite the treatment of service users before their need escalates.
Understand the interaction of population health indicators using machine learning models, to improve service focus and design.
Substance misuse services are highly effective. For example, alcohol services return £3 to the NHS and wider society for every £1 invested. However, 82% of those who could benefit from substance misuse intervention do not access it.
Increasing service support uptake requires overcoming stigma, fear and lack of awareness. Given 70% of those receiving substance misuse treatment also experience mental illness, mental health services across primary and secondary care are well-positioned to drive service take up. The UK Drug Strategy emphasises the importance of acting at the earliest opportunity to prevent escalation, and the government is providing more and more ringfenced funding to local authorities to boost service access.
how it works
Provider datasets are combined with ONS & census data. Both clinical and non-clinical features (or markers), developed in partnership with clinicians, are calculated for service users who have been on addiction teams’ caseloads.
Service users on other caseloads are compared to these patients using these same features and a score is produced. This score indicates the likelihood a service user is eligible for substance misuse treatment based on their case history and other factors.
Substance misuse staff access a central platform to analyse, review and make notes on these services users, allowing them to flag these patients and expedite their referrals to treatment.
A flexible, easy-to-use interface, that allows users to quickly and efficiently review patients.
A robust machine learning model which considers features that matter, improving the more it is used.
Population mapping tools powered by machine learning to understand relative prevalence, to help plan tactical and strategic interventions.
Case studies
Finding service users on mental health caseloads for addictions treatment, with a predictive machine learning model
Find out more