We talk about ‘Data’ a lot nowadays. Sometimes it can sound impersonal and remote; at other times it can sound personal and intrusive. However, when we’re talking about data in mental health research, it really comes down to a very simple question – can we learn from people’s experiences and improve mental health and/or mental healthcare as a result?

Traditionally, information on mental health has been collected in research studies that recruit participants and follow them up over time. More recently, anonymised information from healthcare records and other routine data sources has been used to answer research questions. The two approaches complement each other – research studies provide a lot of detail but tend to focus on selected groups of people, routine data can be huge in size and better represents the population. Both could do better at including those from marginalised underserved groups such as those from ethnic minorities.

The most important task in ‘data science’ is to ensure that patients and carers are closely involved – not only in the security of technologies and platforms being used to support research, but also in the research studies themselves, making sure that they are addressing the questions that most need answering. Clearly this is particularly important in mental health research because of the sensitive issues frequently being covered and the often more personal nature of the data involved.

Advances in healthcare have always been rooted in learning from past experiences. Increasing volumes of information provide a genuine opportunity to accelerate this progress – to develop new ways of delivering care and address the longstanding disadvantages that people with mental health conditions have had to face, particularly in achieving good physical health.

DATAMIND, our Health Data Research Hub for Mental Health Informatics Research Development, brings together a community of mental health data science experts across the UK to make a difference. Most importantly, patient and public involvement and engagement is central, so that priorities can be shaped appropriately and so that we understand how people do or don’t want their data used in research.

Specific projects will develop infrastructure for new clinical trials, for child and adolescent mental health, for under-served communities, for the interface between physical and mental health, for improving the detail of healthcare information through text mining, and for the discovery of new treatments.

And in the coming months, DATAMIND will be working to make more mental health datasets available on the HDR Innovation Gateway and producing papers on some of the challenges facing mental health data science.

Discover and request access to the current collection of 43 mental health datasets on the Gateway.

Find out more about DATAMIND

Find out more about World Mental Health Day