The primary goal of Augur is to predict future symptoms of common mental health problems, alongside and treatment seeking an outcome. In an ideal world, we would hope to be able to do so with absolute certainty, but this is rarely possible; people are complex and they live in changeable and complicated environments. As a result, the best thing we can do is to be as transparent as possible about this uncertainty, and be sure that we communicate it to our users.

Our analysis approach uses Bayesian modelling to provide uncertainty estimates for our predictions. This is central to our goal of informing users not just of how likely someone is to develop mental health problems, seek treatment, or respond to treatment, but to also convey a sense of confidence in these predictions. However, these models can be complex and computationally expensive to fit. This is particularly true when we consider the large datasets we are working with.

The primary goal of our tool is to help mental health researchers. We want to do this in ways that are both powerful and accessible, providing both flexible and extensible software packages alongside easy to use web applications. In this post, we’ll talk about our goals for supporting mental health researchers with Augur. 1. Immediate predictions for researchers The first goal of Augur is to provide researchers with a tool that can make predictions about mental health outcomes, without any need for specialist skills or coding.

In November 2022, Toby and I attended the Data Science Support Workshop organised by the Wellcome Mental Health Data Prize, which focused on open science and FAIR principles. Open science was never new to us, and in fact, all our team members have practised open science in some way, and have received some recognition regarding our contributions to open science[1,2]. To be fair, however, one of the main reasons that we believed (and are still believing) in open science is we all agree science could have been done better, in the era of replication crises.

Lived experience engagement is critical for ensuring that research in applied areas (such as mental health) follows the principle of ‘nothing about us, without us’ - a policy that, it turns out, has a long history in European political tradition. However, lived experience engagement isn’t always easy, and we would argue that it can be a challenge in data-driven research especially. Why do we think this? Because ensuring that lived experience input is meaningful, rather than tokenistic, requires: That input is given at multiple stages of a project, from conception to dissemination That input guides the questions and methods used, rather than just the wording of those questions and methods That those with lived experience and those who work as researchers can understand each other to make truly joint decisions