Mental Health Award: Leveraging longitudinal data to transform early intervention in mental health

Longitudinal data offer a rich yet underutilised resource that could transform early identification and intervention in mental health. This call will fund projects that utilise existing longitudinal datasets and integrate experimental research to predict the onset or progression of anxiety, depression or psychosis. You can explore over 1,500 longitudinal datasets through the Atlas of Longitudinal Datasets.

Successful applications will develop data-driven or hypothesis-driven early identification models and advance their translational potential, aiming to create a step-change in early intervention for anxiety, depression and/or psychosis, either directly or in time. 

We will define longitudinal data as any data collected over time with a minimum of three time-points. Longitudinal datasets include, but are not limited to: 

• cohorts
• electronic health records
• data from experimental research collected over time

Projects will have the potential to enrich existing longitudinal datasets, including linkage to other sources, harmonisation and collection of additional data. Establishing entirely new cohorts will be out of scope. 

We expect proposed research projects to: 

• Use cutting-edge data analytics on existing longitudinal data to develop prediction models that predict the onset and progression of anxiety, depression and/or psychosis. We define these mental health conditions broadly to include bipolar disorder, obsessive compulsive disorder, post-traumatic stress disorder and perinatal mental health problems. Projects may also focus on symptoms strongly associated with these conditions.
• Validate their model in independent existing dataset(s), which can be enriched for validation purposes, including new data collection. Validation datasets need to be separate from the datasets used to develop the model, meaning that subsets or test/validation splits are not sufficient.
• Include innovative experimental research that will transform our understanding of early identification factors and/or advance the translational potential of the prediction models. The experimental research component can extend from the molecular and cellular to animal or human research and intervention trials.
• Meaningfully integrate relevant lived experience expertise in the research team.
• Include collaborations between diverse and multidisciplinary teams covering mental health and data science expertise. 

Although it is not mandatory, we encourage projects to consider: 

• Research within and across low- and middle-income countries in an equitable manner.
• The use of qualitative/unstructured data including, but not limited to:
  • transcribed text files
  • data from wearable devices
  • audio and video recordings

You must submit your application by 17:00 BST on the deadline day. We cannot accept late applications.

The funding call will launch on 23 April 2025 with full eligibility information. In the meantime, we encourage you to initiate discussions about research projects, foster collaborations and explore available resources. To facilitate this, you may find the following resources helpful: 

• The Atlas of Longitudinal Datasets is an online, searchable platform that maps thousands of longitudinal datasets from around the world. It enhances discoverability and accessibility of data, enabling users to uncover new datasets relevant to their research.
• We have partnered with the online networking platform Neuromatch as an option to help you establish novel collaborations. Using artificial intelligence and a bespoke algorithm, Neuromatch will connect mental health researchers, data scientists and data holders around the world who work in compatible areas and wish to collaborate. We will update this section with a link to sign up once registration opens.
• On 7 May, we will deliver a webinar outlining the details of the call and answering attendees’ questions. You can register for our funding webinar now. The webinar will also be recorded and uploaded on the funding call webpage.