Common metrics in mental health research

Without objective measures of mental health, we rely on collecting subjective, often self-reported data regarding people's thoughts, feelings, and behaviour to advance mental health science. However, this process of reporting is done inconsistently across studies and with a variety of different measures.

As a member of the International Alliance of Mental Health Research Funders’ Common Measures Board for Mental Health Science, Wellcome is committed to identifying and adopting common measures so that:

• outcomes can be more readily compared across studies 
• in time, data from different studies can be easily combined to answer crucial research questions.

We have identified and agreed measures for child and adult depression and anxiety, and adult functioning. We are looking at ways of providing the information researchers need to implement these common metrics, and of facilitating selection of further essential metrics.

Teams planning to conduct research projects with human participants, which will include collecting data on anxiety and/or depression must use (as a minimum) one or more of the following recommended measures, as relevant to their research question: 

• Patient Health Questionnaire PHQ-9 (adults - depression)   
• General Anxiety Disorder questionnaire GAD-7 (adults - anxiety)   
• Revised Child Anxiety and Depression Scale RCADS-25 (children and adolescents – depression and anxiety) 
• World Health Organisation Disability Assessment Schedule WHODAS-12 (adults – impact on functioning) 

Information must be captured and reported at the level of the individual item response to allow comparison between studies at the required level of detail.

Researchers may also use as many additional measures as they wish, and we do not require that these measures be the primary outcome measures in any analysis plan.

In exceptional circumstances when these measures are not feasible, we may agree to the use of shorter versions (for example, PHQ-2, GAD-2, RCADS-10), or not at all. However, the reasoning for this must be clearly justified.

The final decision on which measure(s) is/are relevant for a given proposal can be agreed at the point of funding. 

With increasing uptake of these common measures, we anticipate that more consistent data will be available for secondary research. Teams planning to conduct secondary research with data on anxiety and depression may consider:

• Checking relevant repositories for sharing scientific data that include these common measures, such as the NIMH Data Archive.  
• Using resources such as the ‘Catalogue of Mental Health Measures’ to identify cohort and longitudinal studies that use the same measures. Note that this resource is specific to those in the United Kingdom.  
• Drawing on novel tools or harmonisation projects, such as Harmony (a Wellcome Mental Health Data Prize 2022 winner), that use natural language processing to help researchers make better use of existing data from different studies by supporting them with harmonisation of measures and items. Harmony is available online and as a Python and R package.