Longitudinal data for mental health research: one step at a time

Longitudinal studies follow the same group of people over a period of time, allowing researchers to examine inter-individual differences and intra-individual changes in mental health. The Landscaping International Longitudinal Datasets project funded by the Wellcome Trust involved conducting a global search for existing longitudinal datasets with the potential to advance scientific understanding of anxiety, depression, and psychosis. As part of the project, MQ led on the delivery of a theory of change process.

The process was designed to frame the steps needed to utilise the baseline findings curated by the global search.

The process adopted a ‘theory of change’ approach. A theory of change is a tool used to describe and understand the process and pathways through which a desired goal or impact could be achieved. In practical terms, it outlines the steps, or intermediate outcomes, that need to happen in order for a final outcome to be realised.

The approach has been applied to the development and evaluation of public health interventions, including those in mental health, but is in theory applicable to any initiative aiming to achieve change. In the context of longitudinal data in mental health research, a theory of change can be used to identify the key steps and activities required to bring about improvements in mental health outcomes.

While it can seem daunting to attempt to solve a huge international, systems level challenge, all problems are easier when you break them into smaller chunks.

Using a methodology such as Theory of Change helps you to structure the process of breaking it down: starting with the impact you want to see and breaking it down piece by piece right down to the first action that needs to happen. The challenge is to be disciplined and constantly ask “HOW would we achieve that?” After that, it’s all about getting the right people, with range of expertise and going through the process.

To develop a theory of change that incorporates the perspectives of all mental health research stakeholders, a co-production methodology was adopted, with the inclusion of individuals with lived experience of mental illness, as well as researchers, policymakers, and other stakeholders, including those in low- and middle-income countries.

The co-production process began with a series of pre-workshop submissions, where stakeholders were asked to submit their perspectives on the current state of mental health research and the challenges that need to be addressed in the context of using longitudinal data to enhance mental health outcomes. This information was then used to inform the design of the workshop, which was structured to facilitate open and inclusive dialogue.

During the workshop, participants worked together in smaller groups to try and identify the key drivers of change and the intermediate actions that need to be achieved in order to bring about improvements in mental health outcomes.

The outcome of this co-production process is a visualisation and understanding of the perspectives and needs of the mental health community which can be used to guide future research and impact efforts.

The discussion covered:

• How could each of the actions be broken down even further to develop a research or project plan that could help to move forward this vision?

• What resources would be needed to make this all a reality?

• How could you take this methodology and apply it to your own work – perhaps to move something forward to impact that you thought was never possible?

From our experience of prioritisation methodologies and engaging multiple stakeholders to design more democratised processes, it has been shown again and again when you rely on processes being delivered in roundtables, either in person or virtual, you inevitably end up with bias. Whether one person dominates the conversation, or others are influenced by certain people in the room, who are for instance more senior than other attendees, the results could be pushed in a direction by a small contribution of thoughts.

Part of this project’s process design relied on a virtual workshop, but this was carefully orchestrated with surveys and questions sent out to stakeholders in advance. The workshop was then designed to specifically validate the conclusions we had drawn for the individually prepared answers and to test assumptions that had been generated. This deliberative democratised process yields far better results and allows those with new ideas and ways of thinking brought to the forefront of the conversation. It is also particularly valuable when engaging people with lived experience alongside academic and industry partners – offering a level playing field for all. We had an informal briefing meeting with the Lived Experience Experts ahead of the workshop to ensure they had all the information they needed and felt comfortable and confident to participate in the multi-stakeholder workshop.

Longitudinal datasets hold great promise for understanding the full range of factors that influence the origins and development of mental health conditions, and the risk and resilience factors associated with them. Given the huge cost of establishing new longitudinal studies, augmenting existing initiatives would be a far more cost-effective way to obtain longitudinal data relevant to mental health. Linking such sites to one another will facilitate sharing of expertise, tools, and experience, enable coordinated studies and analysis of data, and provide more generalisable conclusions.

The theory of change developed through this project provides a starting point for realising this vision, one step at a time.

It identifies the key areas of activity required to develop a global network of mental health-enabled longitudinal studies and for the effective use of these resources to inform policy and practice.