Longitudinal changes in mental health and the COVID-19 pandemic: evidence from the UK Household Longitudinal Study

Publication type

Journal Article

Published in



Michael Daly, Angelina Sutin and Eric Robinson

Publication date


To examine longitudinal changes in the prevalence of mental health problems before and during the COVID-19 crisis in a large-scale population-based study and to identify population subgroups that are psychologically vulnerable during the pandemic.
Design and setting:
Longitudinal observational population study using data from the 2017 – 2019 and April 2020 waves of Understanding Society, the United Kingdom Household Longitudinal Study (UKHLS), a nationally representative sample of adults in the United Kingdom.
12,090 men and women with a mean age of 49.9 (range 18 – 92) living in private households in the United Kingdom were drawn from the representative wave 9 (2017 – 2019) sample of the UKHLS and followed up between the 24th and 30th of April 2020.
Main outcome measure:
Mental health problems were measured using the General Health Questionnaire 12 (GHQ-12), a validated and widely used measure of common mental health symptoms in population studies. The GHQ-12 score cut-off (score of 3 or more) for “psychiatric caseness” (i.e. likely to present with psychiatric disorder) was used to estimate prevalence of mental health problems.
The percentage of participants classified as experiencing mental health problems increased from 23.3% in 2017-2019 to 36.8% in April 2020. In a multivariate mixed effects logistic regression model all population subgroups examined showed statistically significant increases in mental health problems. Increases were most pronounced among young adults, females, and those with a higher level of education. The rise in mental health difficulties was 8.6 percentage points (95% CI: 4.1 to 13.2) greater among those aged 18-34 compared to those aged 50 – 64, 6.9 points (95% CI: 4.0 to 9.8) greater in females compared to males and 4 points (95% CI: 1.2, 6.9) greater amongst those with a university degree compared to others. Additional analyses of the full UKHLS dataset from 2009-2020 showed that the substantial increase in mental health problems observed in April 2020 was unlikely to be due to seasonality effects or year-to-year variation.
This study contributes to a rapidly growing evidence base suggesting that mental health problems may have risen substantially during the COVID-19 pandemic.




Psychology, Demography, Public Policy, Well Being, Health, Life Course Analysis and Covid 19


Open Access preprint