Scottish COVID-19 Mental Health Tracker Study: Wave 3 Report

Executive summary

This report presents the longitudinal and cross-sectional findings for a range of mental health outcomes for Wave 3 of the Scottish COVID-19 (SCOVID) Mental Health Tracker Study, specifically looking at differences between particular subgroups in the sample. These findings are based on questionnaire data collected between 1^st October and 4^th November 2020 (a period which coincided with an increasing of restriction measures in the central belt in Scotland on 1^st and 7^th October 2020). The main mental health outcomes include: depressive symptoms, anxiety symptoms, suicidal thoughts, psychological distress (measured with the General Health Questionnaire and a single distress item), and mental wellbeing. Other correlates of mental wellbeing are also included, such as loneliness and life satisfaction, which are reported on more briefly.

Two kinds of findings are reported within this report; cross-sectional and longitudinal. The Wave 3 cross-sectional sample is comprised of 1625 respondents, which is 50% of the original Wave 1 sample, as well as a further 327 respondents who were recruited during Wave 3 to boost the sample due to loss at follow-up. The longitudinal sample is comprised of the 1298 respondents who have completed all three waves.

Wave 3 cross-sectional findings show:

• 32.0% of the sample reported psychological distress and a possible psychiatric disorder (based on responses to the GHQ-12),
• 21.4% reported moderate to severe depressive symptoms,
• 16.2% reported moderate to severe anxiety symptoms,
• 9.9% of respondents reported suicidal thoughts within the week prior to completing the Wave 3 survey.

Consistent with the cross-sectional findings from Wave 1 and Wave 2, this report suggests that particular subgroups within the sample are reporting higher rates of mental health problems during Wave 3. These groups include:

• young adults (18-29 years),
• women,
• individuals with a mental health condition,
• respondents with a physical health condition,
• individuals in a lower socio-economic group (SEG^[1]).

Longitudinal analysis suggests overall poorer mental health during Wave 3 compared to previous waves, although this was not consistent across all mental health outcomes. Specifically:

• Depressive and anxiety symptoms increased from Wave 1 to Wave 3, although there were no notable changes between Wave 2 and Wave 3.
• Suicidal thoughts decreased for the whole sample from Wave 2 to Wave 3, although these were still higher at Wave 3 than at Wave 1.
• Psychological distress (as measured by both the GHQ-12 and by a single item) increased from Wave 2 to Wave 3,
• Loneliness increased from Wave 2 to Wave 3,
• Life satisfaction decreased between Wave 2 to Wave 3.

It is important to note particularities of the Wave 3 sample, which has informed and impacted the analysis underpinning this report. Due to attrition (i.e., loss at follow-up) at Wave 3 (particularly of young adults), a booster sample was recruited to allow for more robust subgroup analysis at Wave 3. Respondents in the booster sample cannot be included in the longitudinal analysis (i.e., reflecting changes over time) as they have not completed previous waves. Therefore, two samples are reported within this report: the longitudinal sample comprising respondents who have completed all three waves of the SCOVID Study (n=1219; 46.8% of the Wave 1 sample), and the cross-sectional Wave 3 sample which includes the booster sample (n=1625).

Due to the loss at follow-up at Wave 3, a number of demographic groups are under-represented in the findings reporting changes across waves, specifically young adults (aged 18-29 years). Although the data are weighted to adjust for this loss to follow-up, when weighting data there is a risk of bias as the weights may inflate or suppress the data from subgroups in the sample. As young adults in particular are underrepresented, this may skew some of the findings for this group. Therefore, this age group will only be reported within the Wave 3 cross-sectional analysis. Only statistically significant differences are reported within this summary and the wider document.