Informatics and data science

The informatics and data science group provides a clinical informatics platform that underpins the delivery of all BRC research. It enables the effective management of, access to and integration of research data and routinely collected data to provide reliable outcome ascertainment and broader personal and health characterisation. The program will coalesce around our leading programs in record linkage, and development and application of tools for data sharing, integration, analysis and secure co-analysis. 

Management of BRC research data (Led by Professor John Macleod)

We provide logistical support and consultancy on optimal hardware and software for secure capture, storage, integration and analysis of research data across all our research themes. We provide training as appropriate.

Enhancement of health research through record linkage (Led by Professor John Macleod, Professor Kate Tilling and Professor Harvey Goldstein ).

We have extensive experience of and expertise in linkage both to established data assets such as those held by the Health and Social Care Information Centre (HSCIC) and Clinical Practice Research Datalink (CPRD), and have developed bespoke approaches – working directly with NHS software suppliers to allow patient follow–up using primary care records. Based on this experience, our work in ALSPAC and our collaboration in the MRC Farr Institute, we will enable linkage to routine data (in particular primary and secondary medical care data) using deterministic methods where unique identifiers are available and probabilistic methods where they are not. This will allow ascertainment of health service use alongside clinical risk and outcome data: for example to assess health outcomes in patients undergoing cardiac and orthopaedic surgery (cardiovascular and surgical innovation themes) and health and educational outcomes in children born through assisted conception (perinatal and reproductive health theme). 

Application and development of methods and tools for data sharing, integration and co-analysis (Led by Professor William Browne and Professor Harvey Goldstein).

This transdisciplinary stream, addressing a broad range of data-types, will closely complement the record linkage programme and will draw strength from our internationally-recognised expertise in modelling complex hierarchical data. We will use novel methods and tools that we are already developing and evaluating, to provide a range of approaches to privacy-protected analysis, anonymization and secure federated analysis. 

Environmental and personal assessment (Led by Professor Ian Craddock, Professor William Browne and Professor John Macleod).

Working with the SPHERE project,we are investigating how feasible it is to ask how feasible and feasibility and acceptability of both static and worn sensors to measure health related behaviours and physiological parameters in approximately 100 households. We will also investigate the scaleability to a larger sample of families in ALSPAC. We will work with the research themes to apply these methods for example in the follow up of patients undergoing orthopaedic surgery (surgical innovation theme).