RAPPORT: Developing a Rapid AI-based Policy Probing and Observational Research Tool

    12 th December 2022 Introducing Noemi: Hi, I’m Noemi. For the RAPPORT project I am providing expert input around the technical aspects of ‘targeted learning’ and ‘causal forests’. I am a Health Economist, based at the Centre for Health Economics (CHE) at the University of York. My research has focussed on evaluating the causal impact of interventions and policies on health. Thanks to funding by the Medical Research Council, I currently lead a research team that investigates how machine learning techniques could be employed for better targeted health policies. As part of the Global Health Economics team I also do research  on health policy issues in low- and middle-income countries, for example the physical and mental health impacts of conflict violence. In my free time I enjoy outdoors things like hiking or cycling, and indoors activities such as playing music on my recorder.

12 th  December 2022 Summary of experiences so far on RAPPORT…. Myself and my team were really delighted and excited to be funded for this first ‘Discovery’ stage of the Wellcome Data Prize in Mental Health. Firstly, it was an opportunity to explore newer epidemiological methods, that combine the predictive power of machine learning with the explanatory capacity of traditional statistical methods. Specifically, this was an opportunity to deploy these newer methods in the field of mental health research. These are ‘targeted learning’, which is used increasingly in other fields for causal modelling, and; ‘causal forests’ which can evaluate how different groups of people respond differently to interventions, treatments or ‘exposures’. Secondly, for some time I have become increasingly uncomfortable with the lack of transparency and reproducibility of research using observational health data, especially once machine learning gets involved! Machine learning can make things more comp...

  10 th December 2022 Introducing Lewis: Hi I’m Lewis. I will be leading on the analysis for the project, with technical support from Paul and Noemi. I will also be working with Lauren to help incorporate lived-experience into the analyses. I am a Lecturer in Data Science at the University of York. My background is in Bayesian Statistics, and I got a PhD in this area at Durham University. I have a general interest in a wide range of quantitative methods, including machine learning and targeted learning! My post-doctoral work mainly focussed on applying these techniques to medical selection and regulation, but more recent work has focussed on mental health. Outside of work, I am a keen golfer and supporter of Darlington FC, although both activities can at times be more frustrating than enjoyable!

  Whole team meeting, 2 December 2022   The whole team, including the lay researchers with lived experience met to discuss the feedback provided by the young people earlier in the week. We clarified roles in the team, and discussed how we could all work together to develop Directed Acyclic Graphs (DAGs), that is - causal diagrams, which will help facilitate our application of machine learning and help us achieve the project objectives. To facilitate their involvement, the lay researchers were introduced to the concept of DAGs, which they found interesting. The lay researchers are going to help us develop a framework also to ensure that our use of machine learning is transparent and reproducible so that other researchers in the field can apply these methods too.

  Reflections on the involvement of young minds and people with Lived Experience   Alongside myself, the RAPPORT team includes four lay researchers that have lived experience of depression and/or anxiety; this includes two individuals that have lived experience in a parental capacity. Given that some of the data included in the Millenium Cohort Study has been obtained from parents, we believe it is particularly important to hear things from their perspective too. There were some initial delays involving people with lived experience, which was outside of our control; however, during the week commencing 28 November 2022 we met with the lay researchers, as well as a diverse group of young people to collaborate on the project.     The young people had an interest in research and were keen to hear about the project. We discussed machine learning and the aim of the project, which is to establish whether exercise causes improvements in young peoples’ mental health. ...