Our results show that key workers in essential service sectors (e.g., food chain, utility, transport, and public security or safety) had consistently higher depressive and anxiety symptoms over the entire one-year study period. In contrast, teacher/childcare workers had fewer depressive symptoms than non-key workers at the start and for the duration of the study. Despite the disproportionate research attention paid to the mental health of healthcare key workers during the pandemic, we did not find evidence that those working in the ‘health/social care’ and ‘public service’ sectors had levels of depressive and anxiety symptoms that were higher than the general population.

This is now the third study to have found worse mental health outcomes in key workers employed in sectors other than healthcare.^11,18 In our study, key workers within essential services reported consistently, on average, 2-point higher levels of depressive symptoms and 1-point higher anxiety symptoms than non-key workers independent of potential confounding factors. There are a number of potential explanations for why this group may have been particularly badly affected. First, qualitative studies of such key workers have documented particular challenges such as adapting to duties under more stressful circumstances (e.g., increased workloads and fears of transmitting the virus to family members¹⁵). Second, individuals in essential services key worker roles (e.g., transport, postal, and retail workers) have received less recognition from the public or their employers for their efforts, which could have resulted in a feeling of inadequacy and further exacerbate their mental health.¹⁵ Third, essential services key workers are disproportionally more likely to have lower levels of educational attainment, to be in more routine occupational roles, and to experience financial hardship than other categories of key workers and the general population. Studies focusing on the experiences of people in roles of lower socio-economic position have consistently shown poor mental as well as physical health,²² these findings could therefore reflect existing socio-economic health inequalities within society.²³ Finally, key workers in essential services roles have also been at particularly high risk for contracting COVID-19.^4,13 It is notable that the distinctions between the mental health experiences of essential services key workers and other key worker roles were exacerbated when taking into account having to work outside the home, suggesting that exposure to risk may have had an adverse effect on mental health. This theory is also supported by qualitative work highlighting the challenges faced by essential services key workers in often being unable to access personal protective equipment (PPE) through their employers in comparison to other key worker groups such as health/social care workers and teachers where although challenges with PPE remained, there was more attention given to safe working environments.¹⁵

Another key finding from this study was that health and social care workers did not show higher levels of anxiety or depressive symptoms than non-key workers, although their levels were descriptively higher than some other groups such as teachers/childcare key workers and public service key workers. However, our findings should not necessarily be taken to imply that mental health has not been adversely affected amongst health and social care workers. First, this study looked just at symptoms of anxiety and depression and only measured from the start of the first lockdown in 2020. By this point, hospitals were already overwhelmed with COVID-19 patients. As we lack data on mental health in this group prior to the pandemic, it is therefore unclear whether symptoms recorded here were higher than usual baselines. Second, this study looked at general symptoms of anxiety and depression, but other studies have suggested effects on wider aspects of mental health that these measures may not have captured such as post-traumatic stress.¹¹ Finally, our category combined the experiences of health and social care workers without differentiating between frontline workers vs those working in other roles that might not have involved exposure to COVID-19 patients. As documented in other studies,^5,6 the nature of health and social care roles within the pandemic has been shown to have differential effects on mental health. Therefore, future studies are encouraged that explore in more depth the types of health and social care roles that may have been most adversely affected.

It is also notable that key workers in the teacher/childcare sector appeared to have fewer depressive symptoms than non-key workers at the beginning of the pandemic in March 2020 and over the course of follow-up, although the differences were slight. There are several possible explanations for this. First, the start of lockdown in 2020 in the UK involved the closure of schools. This reduced the workload for many in these professions as most schools were unable to deliver a full curriculum online and also reduced changes of exposure to the virus amongst this group. At the same time, teachers did not face the stress of furlough schemes or unemployment as their jobs were recognised as essential when society did reopen, which may contrast with non-key worker sectors where a greater proportion of people were furloughed. Nonetheless, some other workers within this key worker category such as nursery workers did experience furlough and redundancy. As the pandemic continued and schools reopened, teaching and childcare then offered an opportunity to maintain face-to-face social interactions with a population less likely to be physically affected by the virus (children and adolescents),²⁴ which could have helped improve their wellbeing during lockdowns and when the strictest physical distancing measures were in place.

Our study did not find any statistical differences between the mental health of public service workers and non-key workers. Many in the former group may enjoy greater job security and lower furlough rates, as well as the ability to work from home. Whilst many non-key workers may have also been able to work from home, they may have been faced with fears of job losses, financial concerns, and stress due to not being able to leave home for work. Although the aim of this study was to report trajectories rather than prevalence, non-key workers evidenced anxiety symptoms which were higher than averages in other studies pre-pandemic using the same measure (2.7-3.2²⁵). These findings point to the importance for monitoring and supporting mental health in the population as a whole in the current and in future pandemics.

This study has a number of strengths. It utilised a large sample with sufficient heterogeneity to include good stratification across all major socio-demographic groups. The analyses were weighted on the basis of population estimates of core demographics, with the weighted data showing good alignment with national statistics from the Labour Force Survey;²¹ a nationally representative study. Due to the richness of the dataset, we were able to employ advanced statistical analyses to examine the trajectories of depressive symptoms and anxiety amongst key workers in various sectors since the first lockdown in the UK across different stages of the pandemic over 12 months. Despite these strengths, the limitations of our study raise important points for future search on mental health amongst key workers. First, despite the effort to make our sample representative to the working population in England, there is still the possibility of potential biases due to omitting other demographic factors that could be associated with survey participation in the weighting process. Second, we were only able to analyse data with respondents who reported themselves in government identified ‘key worker’ roles at the start of the pandemic, but the definition of this changed throughout the pandemic. Future study is required to capture how changes in key worker status designation may have impacted mental health of these groups. Moreover, we lacked data on participants’ mental health prior to the COVID-19 pandemic. It therefore remains unclear whether the levels of depressive and anxiety symptoms had already been consistently high amongst the key workers in sectors such as utility, food-chain, transport, and delivery prior to COVID-19, or whether the conditions exacerbated their mental health during the pandemic. Future research is encouraged to look at the longer-term mental health trajectories of key worker in these groups, including after the current pandemic is under control. Although due to our large sample size we were able to examine more key worker categories than prior studies, there was still some heterogeneity within the key worker groups in our study. Due to data limitations, we also were unable to include information on the frontline nature of key workers’ roles, which will be an important confounder to explore in future studies. Finally, we were unable to account for whether and to what extent individual key workers within each category interfaced directly with the public, which might affect their levels of depression and anxiety.^5,6,17

Our findings indicate that the mental health of individuals in key worker roles such as essential services that have been less visible than health and social care has been worst affected during the current pandemic, suggesting the potential importance of mental health screening in at-risk occupations during pandemics. More mental health support will therefore be needed to deal with these symptoms as the pandemic eases. There is also a need for more fine-grained analyses of the mental health of different types of key workers, particularly those working in utility, transport, and food chain roles to identify individuals in specific occupations who will need this support most. Future research should also seek to understand ways in which workplace measures to mitigate risk may have been inadequate during the current pandemic, to inform policies for future pandemics.