We have an ambitious programme of multidisciplinary Health Data Science and Informatics research underway within the University, falling broadly within the following research themes:
Connected Health Cities
The University is a lead partner in the North West Coast Connected Health Cities Programme funded by the UK’s Department of Health, hosting the Healthcare Data Laboratory – a multidisciplinary group of analysts and clinicians working in close partnership with key stakeholders involved in the planning and delivery of care.
Working across three exemplar care pathways, the programme is developing innovative approaches to interrogating, analysing and sharing healthcare data to support the creation of a learning healthcare system – generating better ways to evaluate pathway improvement initiatives and creating actionable analytics to guide more effective interventions.
The University of Liverpool has the largest Department of Pharmacology in the UK. It hosts the Wolfson Centre for Personalised Medicine and the MRC Centre for Drug Safety Science; both pre-eminent in their work in personalised medicine. Our aim is to combine our expertise in clinical pharmacology with state of the art technologies including genomics and proteomics, to identify biomarkers which allow for disease stratification, identification of individuals who respond well to medicines and those who respond adversely to medicines, as well as develop algorithms to individualise dosing.
This programme of research is led by Professor Sir Munir Pirmohamed who also leads the UK Pharmacogenetics and Stratified Medicines Network, is a member of the Expert Advisory group for Precision Medicine Catapult, is lead for the pharmacogenomics sub-domain for the 100k genome project, and has sat on advisory boards for A*Star and the Mayo Clinic.
The Centre for Genomic Research also includes a dedicated team of 15 experienced bioinformaticians and software engineers working in conjunction with laboratory specialists ensuring that large and complex genomic datasets can be analysed efficiently and effectively.
The University is at the forefront of the development of alternative robust epidemiological methods to demonstrate the health inequalities effects of public policy decisions. Our vision is to build the evidence-base to reduce health inequalities, applying robust methods for policy evaluation to innovative data resources developed through a wide collaborative network.
Working with Liverpool CCG, Liverpool City Council and Alder Hey Children's Hospital we are establishing an anonymised population level child health and development dataset for all children in Liverpool, nested around the universal health check data, linked to birth records, data on use of local health and social care services, educational outcomes, and National Child Monitoring Programme data (NCMP). This will form the basis of an electronic population cohort to allow epidemiological assessment of the early origins of health inequalities in Liverpool.
A collaboration between NIHR CLAHRC North West Coast, the ESRC funded Consumer Research Data Centre and 10 local authorities across the North West has also enabled the establishment of an Integrated Longitudinal Research Resource (ILRR) of linked neighbourhood datasets, enabling the tracking of the determinants of health and health outcomes within neirghbourhoods along with novel neighbourhood level contextual indicators of networked distances to health care facilities, health related commercial outlets (fast food, alcohol) and health assets (leisure facilities, green spaces).
Infectous Disease is one of three University level research themes recognised globally for its international research excellence. An interdisciplinary approach seamlessly brings together medical and veterinary science to focus on diverse challenges such as new medicines for children and tackling HIV disease, through to developing diagnostics, treatments and vaccines for both humans and animals. Major programmes in health and biomedical informatics include the Integrate programme, led by Professor Sarah O'Brien; SAVSNET, led by Dr Alan Radford; vaccines research led by Professor Neil French; and Enhance led by Professor Matthew Baylis.
Analytics, Methods and Standards
The University of Liverpool has one of the largest Departments of Biostatistics in the UK undertaking cutting edge research in areas such as joint modelling of longitudinal and time-to-event data, multivariate analysis, multi-source evidence synthesis, and statistical genetics and pharmacogenetics. Combined with a strong group of geographers with expertise in spatial statistical analysis and a cross-faculty analytics special interest group, this offers high-level analytical support and innovative methods development opportunities to the Institute.
The Department of Biostatistics hosts the University's Healthcare Data Laboratory - led by Dr Keith Bodger; linking our critical mass of analytical expertise with a programme of engagement of front-line clinicians across the North West Coast, supporting co-production of clinically validated analytics and data visualisations for defined patient pathways.
The University also hosts the MRC North West Hub for Trials Methodology Research (NWHTMR), the Clinical Trials Research Centre, and the COMET initiative (www.comet-initiative.org). COMET aims to better align measurement and annotation of appropriate outcomes in research and the Electronic Health Record. NWHTMR undertakes evaluation of approaches to e-recruitment and data sharing. CTRC is delivering a number of NIHR funded e-trials including ISDR, CF START, REACT and OVERT, using informatics in a way similar to the GSK Salford Lung Study, and developing web and app based solutions to collect patient reported outcome data. We are working with CPRD on point of care trials, including the AZ-funded DECIDE study. We are developing knowledge, guidance and platforms for efficient trials that can be shared across the Institute.
Clinical code sets, rules and algorithms to interrogate healthcare data
The development and validation of code sets, hierarchical rules and algorithms to interrogate healthcare datasets in more meaningful ways is a core activity of the Healthcare DataLab and CHC programme, bringing together front-line clinicians and subject experts to iteratively refine methodologies.
Despite the use of standardised classification systems and terminologies within structured routine administrative or clinical datasets, there is well-recognised variation in the way specific clinical concepts may be coded, including phenotype (e.g. symptoms/signs, diagnosis, complications, co-morbidities), interventions (e.g. procedures) and relevant visits or contacts within a care pathway (e.g. attendances for specific types of service).
Although variability can relate to coding errors or changes to administrative processes, there is also variation arising from the use of codes that have similar clinical meaning or represent legitimate alternatives to recording the same clinical concept.
This presents particular challenges when re-using routine data to study complex, chronic diseases and multimorbidity, both for the identification of clinically-meaningful patient cohorts and for tracking patient journeys and outcomes over time and across datasets. Creating linkage between datasets that use differing clinical coding systems adds further complexity. By creating, testing and sharing code libraries, metadata and data processing scripts, this work aims to improve the transparency, reliability and reproducibility of analyses.