Linking mental health records for research: Eating disorders as a test case

Mental health conditions affect many people, but research to better understand them through drawing on routinely collected data is limited. This is because data from different parts of the healthcare system is not easily linked.

Avon & Wiltshire Mental Health Partnership NHS Trust (AWP) provides specialist mental health services, and its electronic health records contain valuable information about patient care. However, these records are not currently connected to other health and research data sources.

The South West Secure Data Environment (SDE) is a secure platform that allows trusted researchers to link and analyse health data from multiple sources. By creating safe processes to bring AWP data into the South West SDE, we can better understand mental health care across the whole system.

Project aims

We want to build a data pipeline to allow AWP’s electronic patient records to be securely linked to wider health and research data in the South West SDE. We will use eating disorders as an example to test this process. The pipeline can then be re-used for other projects.

We will develop systems to process structured and unstructured data, set up governance and legal frameworks and ensure ethical approvals. We will also create protocols for linking AWP data with other sources, such as GP, hospital and social care records, and research datasets like Children of the 90s.

What we hope to achieve

By linking these data sources, we can map treatment pathways for people with eating disorders. We can understand things like how long they wait for care, what treatments they receive and their contacts with other NHS services.

We will also explore patterns in research cohort data, such as Children of the 90s, to understand risk factors and outcomes.

This work will help identify opportunities for prevention and early intervention, evaluate services and investigate inequalities in access and outcomes.

Finally, we will produce a blueprint for creating linked mental health datasets, enabling similar research for other conditions in the future.