While our knowledge about the development of mental health problems has been steadily increasing over the past few decades, this has failed to curb the rise in the number of people living with mental health problems.

Undertaking research to better understand the onset, development and recurrence of disorders such as anxiety, depression and psychosis is crucial for finding rapid and efficient ways to predict, intervene and ultimately stop the harmful outcomes of mental illnesses.

Insightful questions and reliable data are the foundation of solid mental health research. With recent scientific advances, we have developed sophisticated tools for conducting innovative research. One tool that has stood the test of time is longitudinal studies, which have generated a wealth of valuable data to answer questions about the development of mental health problems.

The value of longitudinal data as a research tool for mental health research

A longitudinal study is a prospective observational study that follows the same group of people repeatedly over a period of time. Individuals followed by longitudinal studies can be part of the general population or groups with particular diseases or conditions. Data collected across several years are extremely valuable as they allow for the examination of inter-individual differences and intra-individual changes. (To find out more about the value of longitudinal studies, have a look at CLOSER, the home of longitudinal research in the UK.)

The Wellcome Trust, one of the most important funders of mental health research in the world, sees longitudinal data as a critical resource to help researchers advance our understanding of how the brain, body and environment interact in the trajectory and resolution of psychiatric disorders. Investment in existing or new datasets may be important for answering key questions in relation to mental health problems.

But here’s a note of caution from the late Prof Sir Michael Rutter. He wrote, nearly 30 years ago “… it is crucial that researchers do not rush impetuously into the premature use of longitudinal studies … longitudinal studies are expensive, time-consuming, and need to be reserved for circumstances when their considerable research power can be used to maximum advantage and not wasted on exploratory investigative forays into new territories. Moreover, longitudinal studies must be science driven if their expense is to be justified.” (Rutter, 1994).

Rather than investing in setting up new longitudinal studies, we can maximise the potential of existing studies by enabling them to collect further data or by providing insight from various stakeholders, including people with lived experience of mental health problems.

Aims of this new project

In accord with this statement, the Wellcome Trust has tasked us with conducting a global search for existing longitudinal datasets to advance scientific understanding of how the brain, body and environment interact in influencing the course of anxiety, depression and psychosis. Wellcome are interested in opportunities to enrich these datasets with the ultimate aim of finding new and improved ways to predict, identify and intervene as early as possible in ways that reflect the priorities and needs of those who experience mental health problems.

We are delighted and excited to have been tasked by Wellcome to search the world for large, longitudinal datasets that have the potential to answer questions related to mental health. More specifically, we are looking for datasets:

  • Drawn from different sectors, such as academia, government, healthcare, education and industry;
  • From any discipline, including those focused on specific health problems, economic and social outcomes, occupational cohorts and beyond;
  • That are established or planned within the next three years (must be funded already);
  • Include, or have potential to include, mental health data on participants at some point between the ages of 14 and 30;
  • From teams that are active and in contact with the participants, or plan to collect new data in the future.

We still do not know for sure how many longitudinal datasets we will be processing in the coming weeks, but we are certain that we will discover many studies we did not know about before.

Building a strong partnership for landscaping across the world and across sectors

We are a team based at King’s College London, the Institute of Psychiatry, Psychology and Neuroscience and we know a thing or two about mental health research and longitudinal studies. We created and developed the Catalogue of Mental Health Measures, a platform designed to facilitate mental health research and maximise the uptake of mental health and wellbeing data already collected by longitudinal studies.

For this new project, we will work with key partners to ensure we leave no stone unturned.  We developed a partnership of researchers, charities, industry and lived experience experts to work with us on this project. They include: the Open Data Institute who will focus on landscaping datasets outside academia; MQ Mental Health Research who will lead on organising a workshop to gather the views of various stakeholders on how we can enrich existing datasets; the Centre for Global Mental Health who will help in uncovering datasets from low and middle income countries; DATAMIND who will help in discovering different types of datasets; and Wellcome’s lived experience advisors who will provide feedback throughout the whole project.

A meticulous step by step process

We have only 6 months to complete this mammoth task and so we must be super organised! We have split our work into 5 different stages.

Firstly, we are searching the world for the most promising large scale longitudinal datasets. We are searching widely and deeply using tools that make datasets discoverable. We are also visible on social media to make people aware of our activities and how they can tell us about longitudinal datasets they are aware of. We are especially interested in hearing from people outside academia.

Secondly, we are screening datasets to ensure they are ongoing or will commence within the next three years with funding already in place. Datasets may be focused on mental health with measures of anxiety, depression and psychosis already collected, but others may not. Studies do not need to be run in English, but the data holders need to be able to communicate with us in English.

Thirdly, we will soon start surveying more closely the datasets that made the cut to ensure they fit the predetermined criteria. Some of those criteria refer to the number of participants in the dataset and the granularity of the assessments.

Fourthly, we will assess how the relevant datasets could be enriched to maximise their potential for answering questions about anxiety, depression and psychosis. We will not do this alone; we are in the process of organising a workshop that will gather a range of stakeholders to consider the best approach to move the field forward.

Lastly, we will write a report summarising our findings. We aim to make recommendations to Wellcome about how they can enrich existing longitudinal datasets.

We will soon make the list of longitudinal datasets we have identified available on our website so that everybody can make discoveries along with us.

Visit the Catalogue of Mental Health Measures


Rutter, M. (1994). Beyond longitudinal data: Causes, consequences, changes, and continuity. Journal of Consulting and Clinical Psychology, 62, 928-940.