In search of vulnerability mechanisms for adolescent depression has the working title of 'What influences our mood?' This research project examines adolescents' mood to see if we can identify factors that will help us understand why some young people experience low mood more often than others.


Depression is a highly prevalent illness that often recurs over the lifetime, with earlier onset being predictive of episode length and severity. Previous research has highlighted certain markers to be salient predictors, such as neuroticism, cognitive biases, and stress, so this work examines these markers alongside other potential factors. The overall aim is to identify markers for adolescent depression in order to predict mood over time; this would lead to being able to identify those at risk to ensure they receive early intervention. We can also identify which markers are modifiable and develop interventions to target them.


The project is led by Stella Chan, an academic clinical psychologist, with a collaboration between Clinical Psychology (involving Dr Stella Chan and Dr Elaine Gray) and Psychiatry at the University of Edinburgh (involving Dr Heather Whalley and Dr Toni Clarke). The project is funded by the Wellcome Trust. Additional support has been received from the Centre for Cognitive Ageing and Cognitive Epidemiology to extend our protocol to include genetic analyses.


Since June 2018, we have been working with young people aged 12 to 18 and assessing their mood, psychological/cognitive factors (including thinking styles, emotional biases, and personality traits), physical/biological factors (including genetic factors (DNA methylation to measure epigenetic age), sleep quality (using actiwatches), and hair cortisol (the so-called stress hormone)), and social/interpersonal factors (including interpersonal relationships and social media use). It is possible that these factors may help us explain health and mental health outcomes in the future, so we will also examine whether these factors can predict mood over time.



Overview of project stages:

Project development: February 2018-May 2018 (complete).

Baseline assessment in schools and online: June 2018-August 2019 (complete).

6-month follow-up assessment online: December 2018-January 2020 (complete).

Workshops in school (dissemination): March 2020 (complete).

Data analyses and write up: ongoing.


Who took part?

A community sample of 425 adolescents aged between 12 and 18 was recruited. The adolescents were from 12 Scottish secondary schools (10 state and two private) over four council areas. Participants had to self-identify as being fluent in English in order to participate. Written informed consent was obtained from participants or parents/guardians if participants were under the age of 16 (along with child assent). Participants were recruited over a 13-month period.



What was involved for participants?

1. Participants completed three paper and pencil questionnaires in school to report their mood (depression, anxiety, and mental wellbeing).


2. In school, participants were asked to provide a sample of hair for us to measure cortisol (the so-called stress hormone) levels in the hair over the past few months. This was to see whether the cortisol levels related to their reported stress and to their mood.


3. In school, participants were asked to give a saliva sample, which involved spitting into a tube for a few minutes (see photos below). This was to examine DNA methylation (to measure epigenetic age) and to see whether this DNA information related to mood.

4. In school, participants were asked to take away and wear an actiwatch (see photo below) for a night. Actiwatches are like a wristwatch that measures sleep quality. This was to see whether sleep quality relates to mood.

5. Participants were asked to complete some more questionnaires online and received a £10 gift card for doing so. These questionnaires were to measure different factors thought to be linked to mood and included:


Demographic information



   Postcode (for Scottish Index of Multiple Deprivation index to estimate socioeconomic status)

   Cultural background

Physical, health, and lifestyle

   Height and weight (to calculate body mass index)

   Mental health history (self)

   Mental health history (family)

   Medication (self)

   Physical activity in last seven days (from Health Behaviours in School-aged Children (HBSC))

   Physical activity (frequency) (from HBSC)

   Physical activity (duration) (from HBSC)

   Tobacco use at present (from HBSC)

   Tobacco use in past 30 days (from HBSC)

   Alcohol use at present (from HBSC)

   How many times have they ever been really drunk (from HBSC)

   Cannabis use in past 30 days (from HBSC)

   Adolescent Food Habits Checklist


   Brief Resilience Scale

   Ruminative Response Scale

   Dysfunctional Attitudes Scale

   Eating Attitudes Test 26

   Eysenck’s Short Neuroticism Questionnaire

   Perceived Stress Scale

   Short Ambiguous Scenarios Task for Depression in Adolescents

   Self-referent task


   Number of social groups they are part of

   How many social media sites they used

   How they felt after using specified sites

   Social media usage (on school days, on non-school days)

   Other as Shamer

   Level of Expressed Emotion

   Bullying and cyberbullying perpetration

   Bullying and cyberbullying victimisation

6. Participants have been contacted 6 months after completing the first part of the study and were asked to repeat the three mood questionnaires (depression, anxiety, and mental wellbeing), so we can see how participants' mood has changed over time and if it relates to any of the measures. A total of 212 (49.88%) completed the 6-month follow-up study.

This research has been approved by the Research Ethics Committee at the School of Health in Social Science and the Local Education Authorities of the schools taking part in the project. 

All the information collected from participants is kept confidential and will not be shared with anyone. Only the members of the research team have access to the raw data. Data are anonymised and stored against participant codes only, with only one separate file to link participant codes to participant names. Participant names and all other identifying information is kept separate. All the information collected is stored securely in encrypted password-protected files on the University of Edinburgh's secure network. Data will be reviewed for destruction every five years and once the study is complete only an anonymised master copy of the data will exist in University archives. The findings of this study will be written up and published in academic journals and presented at conferences, but names and all other identifying information will never be disclosed.

If you have any questions please contact Elaine or Stella.

© 2023 by