Researchers at the Nuffield Department of Primary Care Health Sciences' Bennett Institute for Applied Data Science have received two major awards from Wellcome: £7 million to support groundbreaking research into mental health treatment outcomes, and a £10 million investment in new techniques for secure access to NHS data for research.
The first award will, for the first time, enable researchers to analyse anonymised NHS Talking Therapies data alongside GP records, in a highly secure setting. The Talking Therapies service is a major part of the NHS, delivering NICE-recommended psychological therapy for depression and anxiety disorders to over 670,000 patients in England each year. The therapies include cognitive-behaviour therapy, counselling, and guided self-help. Uniquely, outcome data is collected from 98% of people who have a course of treatment.
Incorporating this data into the OpenSAFELY platform will help answer many vital questions about mental health treatment, including:
- How talking therapies affect long-term health outcomes
- Which approaches work best for specific conditions and patient groups
- The best way to deliver services
- The relationship between mental health treatments and physical health
The research will build on the success of OpenSAFELY, the secure analytics platform developed at Oxford during the COVID-19 pandemic. OpenSAFELY has been delivering whole population data analysis, using innovative new methods to protect patients’ privacy while allowing researchers to conduct their research. The platform's findings directly informed UK public health policy decisions during the pandemic, particularly regarding protection for vulnerable groups.
Building on the Bennett Institute's existing collaboration with NHS England, the project will analyse outcome data from millions of patients who have used NHS Talking Therapies services while maintaining strict privacy controls. No identifying patient information leaves NHS systems, as all analysis takes place within the secure OpenSAFELY environment. Researchers design and write their analysis using randomly generated fake “dummy data”. They then submit it for automatic remote execution on the real patient data. This means that they don’t need to interact directly with real patient records. As part of the mental health project, OpenSAFELY and NHS England will also explore new mechanisms for data linkage, where datasets are minimised before moving between NHS England controlled data centres.
Read the full story on the Nuffield Department of Primary Care Health Sciences website.