Dr Jo Inchley, Dr Judith Brown and Prof Daniel Smith tell us all about the valuable mental health big data research taking place at the University of Glasgow, Scotland
Several areas of medicine are currently benefiting from the application of innovative data science and health informatics approaches, including mental health. The Medical Research Council (MRC) in the UK have the ambition to establish a UK-wide Mental Health Data Platform to drive research, innovation and clinical translation for mental illnesses. As a first step, in 2018 they awarded nine UK research centres initial funding to set up Mental Health Data Pathfinder Projects.
The University of Glasgow’s Mental Health Data Pathfinder Award will build on our excellent record for interdisciplinary health informatics research in mental health and will be delivered in partnership with the NHS in Scotland, the voluntary sector and with both local and national government. The project has two primary areas of activity: the development of a Scotland-wide schools health improvement research network (called SHINE); and enhancing existing clinical and eHealth cohorts to drive discovery science for mental health.
The Schools Health and Wellbeing Improvement Research Network (SHINE)
Promoting the mental health and wellbeing of children and young people is a top priority. Pupil health and academic attainment are intertwined as healthier pupils have higher attainment. However, a lack of evidence-based, up-to-date, local data has been identified as one of the barriers to improvements in health and wellbeing. SHINE is a pilot study to establish a national infrastructure for school-based health improvement research in Scotland with the aim of building stronger data partnerships between schools, policymakers and academic researchers.
SHINE follows the successful Schools Health Research Network (SHRN), established in Wales in 2014. SHRN began with a small number of secondary schools participating in the Health Behaviour in School-aged Children (HBSC) study, but by 2018 all Welsh secondary schools had joined the Network. Fundamental to its success has been the production of individual school-level health and wellbeing data reports which enable monitoring of health and wellbeing within the school community, targeting of interventions and an evidence base for health improvement planning.
In Scotland, the SHINE network aims to support schools in addressing their health and wellbeing needs, with a focus on mental health, by using a data-driven, systems-level approach to health improvement. Underpinning SHINE is the Scottish HBSC survey, led by Dr Jo Inchley who is also the International Coordinator of the HBSC study, currently involving 49 countries. HBSC collects internationally comparable data every four years on 11-, 13- and 15-year-old boys’ and girls’ health and well-being, health behaviours and their social context. Data from the HBSC survey is now being fed back to participating SHINE schools via an individual school health and wellbeing data report and will be used to inform health improvement planning and facilitate actions which promote positive health and educational outcomes.
Self-report mental health data have several key strengths, particularly within the context of population-level health promotion and disease prevention. They can be collected at low cost, and are relatively non-invasive and hence can achieve high response rates and greater representativeness. While a wide range of self-report mental health measures currently exists, there is a need for further validation among the adolescent population and for increased harmonisation across surveys. SHINE will review existing self-report mental health measures and undertake validation work to produce a suite of recommended indicators which are valid for use in the UK and international surveys. SHINE will also establish mechanisms for linkage of adolescent mental health data to a range of existing educational and health data to enhance understanding of the causes and consequences of mental ill-health.
Building capacity for big data research in mental health
The NHS in Scotland is unique in having a dedicated service for health informatics, the NHS Information Services Division (ISD). We have a range of projects currently in progress with ISD, one of which was focused on using national-level linked health data to assess changes in patterns of prescribing for bipolar disorder. This work was published recently in the British Journal of Psychiatry. In brief, we found that a quarter of patients with bipolar disorder were being prescribed antidepressant medication on its own, a practice that is at best unhelpful (antidepressants are ineffective for bipolar depression) and, at worst, potentially harmful (antidepressants may cause more frequent episodes of mania). Furthermore, only 1 in 20 patients were taking lithium monotherapy. Lithium is the gold standard treatment for bipolar disorder and is recommended as first-line treatment by clinical guidelines in the UK and in many other countries around the world. To find so few patients on lithium monotherapy is a major concern and suggests that many patients in Scotland with bipolar disorder may be missing out on the best available treatment options.
This work has been covered widely in the media and we hope it will act as a stimulus for psychiatrists in Scotland to review their prescribing preferences for bipolar disorder. Another area of activity within the Glasgow Mental Health Data Platform Award is the development of large NHS clinical cohorts as future resources for data science research. For example, the Glasgow Psychosis Clinical Information System is a clinical database tracking over 10,000 patients with schizophrenia and bipolar disorder. We have started innovative work to link these patients to routinely-collected data (such as brain MRI scans and blood monitoring) to provide a much more comprehensive set of phenotype and outcome data for research. Additionally, we are collecting consent from these patients so that we can store a routine sample of blood for future pharmacogenomics and stratified medicine research projects.
Overall, this is a very exciting time for mental health big data research. Health informatics and data science approaches have the potential to radically change the way that we approach the assessment, diagnosis and treatment of very large numbers of people with psychiatric disorders. As our SHINE project highlights, this work also has enormous potential for public health approaches to improving mental health at a population level. We are very grateful for the support provided to us by the MRC in the UK for this work.
Please note: This is a commercial profile
Dr Judith Brown
MHDP Project Manager
Institute of Heath & Wellbeing
University of Glasgow
Glasgow G12 8RZ
Tel: +44 (0)141 330 2023