In April 2023, the DATAMIND Data Science event included workshops and conference for researchers to share talks and presentations. Dr Amy Ronaldson, an MQ Research Fellow and pictured above, shares her highlights of attending the experience.
Towards the end of April (2023), researchers, data scientists, clinicians, and experts by experience came from across the UK to attend the two-day event held in the London HQ of Deutsche Bank. On Day One, we attended an interactive roundtable workshop which focused on using clinical data for mental health research. The Data Science Meeting took place on Day Two with presentations, panel discussions, and Q&A sessions from researchers at all career stages, as well as from experts by experience.
It is clear we are entering a sort of ‘golden age’ for data science in mental health and MQ’s Data Science event held in collaboration with DATAMIND is real evidence of this.
A little bit about me – I am an MQ Research Fellow trying to understand why people with severe mental illness (SMI) are at an increased risk of dying from infectious diseases. My work relies heavily on big data from electronic health records, so meeting researchers using similar at events like these is invaluable.
Do Not Let Perfect Be The Enemy Of Good – and other lessons
MQ and DATAMIND’s workshop kicked off with an introduction to clinical data where we explored the benefits and challenges of using this sort of data for research. Matthew Broadbent and Amelia Jewell from the NIHR Maudsley Biomedical Research Centre (BRC)’s Clinical Informatic Service explained that although this type of data is not collected specifically for research, it can be made to be “research-friendly”. Even though there are inherent issues around data quality and validation, the take-home message here was do not let perfect be the enemy of good (basically…although clinical data is not perfectly suited for research, it still has huge potential as long as we approach it in the right way).
Using clinical data for research is an iterative process where you learn as you go, often tweaking research questions and hypotheses based on the sort of data that is available. This was made clear for all of us by Dr Risha Govind (NIHR Maudsley BRC) who led us through an interactive task where we had to design a study assessing the impact of clozapine on COVID-19 incidence using electronic health records.
This hands-on task really demonstrated how clinical records can be leveraged to answer important research questions, as long as those questions are framed correctly. Dr Katrina Davis (King’s College London (KCL)) then used the UK Biobank to showcase how data linkages can be used for the validation of important clinical concepts such as diagnoses and medication usage. A striking example of research from Scandinavia illustrated that all may not be what it seems when using clinical data.
Measures of Mental Health and Transparency
In the afternoon, Professor Louise Arsenault (KCL) and her team showcased The Catalogue of Mental Health Measures – a catalogue they have developed of British cohort and longitudinal studies which contain measures of mental health and wellbeing. A helpful demonstration showed how the advanced search function allows you to filter for detailed characteristics like sample size, age at recruitment, and study design.
The day was brought to a close by Professors Ann John (Swansea University) and Rob Stewart (KCL) who co-direct DATAMIND. Together, they facilitated a really lively discussion about how DATAMIND might support early career researchers in mental health data science.
The Data Science Meeting began the following day with Johnny Downs (KCL) and Pauline Whelan (University of Manchester) discussing embedding data science and digital mental health within clinical services. Pouria Hadjibagheri then delivered a fascinating keynote. Pouria was the Tech Lead for the UK Government COVID-19 dashboard and described how much data, time, and work went into maintaining the dashboard we relied upon so heavily at that time. The biggest take home message for me from Pouria’s talk, however, was the importance of open data and transparency.
Highlight of the Day
My personal highlight of the meeting took place after a short break for coffee – a panel discussion with the DATAMIND Super Research Advisory Group. This group is made up of service users and carers from various backgrounds who have an interest in data. They provided invaluable insight into how to incorporate lived experience into ‘big data’ research, stressing the importance of the user perspective and co-production.
Later in the afternoon, Professor Sonia Johnson and Dr Nafiso Ahmed (University College London) presented the results of their systematic review looking at mental health in Europe during the COVID-19 pandemic. What really stood out to me was the inclusion of a ‘lived experience commentary’ section in the review where experts by experience provided their interpretation of the findings.
Bringing the voice of lived experience to academic publications struck me as such an excellent idea and will be something I incorporate into papers and reviews going forward.
Lunch provided some time for networking (and sandwiches…) as well as the research poster session. Early career researchers then gave exciting three-minute thesis competition style presentations on topics.
After lunch, food for thought
In the afternoon, Professor Cathie Sudlow showcased the British Heart Foundation Data Science Centre which links cardiovascular health data with other largescale data resources (e.g. mental health services), in order to improve the nations cardiovascular health.
Professor David Osborn (UCL) then gave a fascinating whistle-stop tour of his work on the physical health of people with SMI (this was of particular interest to me, seeing as I want to understand why death from infection is higher in people with SMI…).
Prof Osborn talked to us about the damaging effect of diagnostic overshadowing (when a health professional attributes a patient’s symptom to a psychiatric problem, when they may actually have a comorbid condition) in the health of people with SMI, and the importance of lived experience in big data clinical research.
The meeting then came to an end with a ‘soft launch’ of the DATAMIND website which went live that week.
I learned so much at this data science event and left with a lot of food-for-thought. I am now really excited to get my MQ Fellowship off the ground and hope I have some exciting ‘big data’ findings to share with you all at the next meeting in the Autumn.
And MQ can’t wait to hear what Amy has to share! Our thanks to Amy for sharing their experience with us. You can learn more about future events like this by signing up to our research round up newsletter and also by visiting our events page.
