Understanding Mental Health in the Research Environment

The Royal Society and Wellcome Trust are interested in better understanding the mental health needs of researchers, and what interventions could be used to support them. This reflects the recent focus on mental health among undergraduate students in the UK, and the concern that others in the academic and wider research environment may have mental health needs that have not been as well explored and considered. This study aims to establish what is currently known about the mental health of researchers based on the existing literature.

Context

Over 6 million working-age people in England have a mental health condition at a given time. The most common diagnosable difficulties among working age adults are anxiety and depression, each of which includes a number of different conditions. Less common but still widespread mental health diagnoses include personality disorders and psychoses such as bipolar disorder and schizophrenia. Many people will have more than one diagnosis at a time, or receive different diagnoses over time.

The causes and triggers of poor mental health are complex and not fully understood. There is evidence that the vast majority of people who experience poor mental health in adulthood first experienced difficulties as children, often from a young age. Risk factors for poor mental health include having a parent with mental health difficulties, growing up in prolonged poverty and housing insecurity, experiences of abuse, neglect and bullying, and traumatic experiences during childhood. Some groups of people have a heightened risk of poor mental health, including some black and ethnic minority communities, people with long-term physical conditions, lesbian, gay, bisexual and transgender people, and people with disabilities.

There is mixed evidence about the extent to which a person's experiences of work contribute to their having a mental health difficulty. Survey evidence suggests that workplace factors such as bullying, insecurity and a lack of control are major causes of mental ill health among staff. On the other hand, there is also evidence that work helps many people to recover from an episode of poor mental health, and there is clear evidence that unemployment is a major risk factor for mental and physical ill health.

Mental ill health and work-related stress are key issues for the labour market as they affect productivity through absenteeism and presenteeism, and are associated with high economic costs for individuals, employers and the economy at large. It has been estimated that poor mental health costs employers in the UK £26 billion nationwide each year, equivalent to £1,035 for every employee in the workforce (Centre for Mental Health, 2007).

Approach

Little is known about how mental health needs vary across working environments, or about how to tailor interventions to address different working populations.

The aim of this study was to assess what is known about mental health in research environments through a literature review, and it focused on the UK and comparable research systems. A better understanding of researchers' mental health needs will enable the design of more effective interventions to address them, while a better understanding of evidence gaps can also help guide future research efforts in this area. The following research questions guided the study:

• How are "mental health" and "wellbeing" understood in the context of research environments?
• What is currently known about researchers' mental health and wellbeing, and does it differ from that of other populations?
• What interventions are used to support researchers, and what evidence is there of their effectiveness?
• What are the strengths and limitations of the evidence base in this area?

How Are "Mental Health" and "Wellbeing" Understood in Research Environments?

Overall, the existing literature offers little insight into what sets the research environment apart from other workplaces, or into how mental health, stress, and wellbeing are defined in these contexts. Rather, the majority of the literature identified focuses on describing the levels of stress amongst the academic workforce and, in particular, identifying contributory factors within the workplace. There is little available evidence based on objective clinical assessment about the prevalence of clinically defined mental health conditions and their treatment in this context. The focus on wellbeing raises the issue that although the presence of common mental health conditions does correlate with some of the wellbeing scales used commonly in the literature, more serious (e.g. psychotic) mental illnesses are not necessarily aligned with measurement of wellbeing.

The literature is also almost exclusively focused on universities, with many studies covering all university staff, which will include both researchers and non-research staff. Some studies focused more specifically on researchers, and a more limited group within that looked at particular groups of researchers—most commonly PhD students, reflecting the wider focus on (typically undergraduate) students in the literature around this topic. The majority of the existing research is based on survey data, which is subject to sampling biases, relies on self-reporting, and was not triangulated with other objective indicators, such as absence data.

What Is Currently Known About Researchers' Mental Health and Wellbeing, and How It Differs from Other Populations?

Evidence on the Prevalence of Work-Related Stress and Mental Health Problems

Despite widely reported anecdotal evidence and press coverage of a "mental health crisis" in academia, there is limited published evidence regarding the prevalence of specific mental health conditions among researchers. The majority of the literature on prevalence identified through this review relates to the experience of work-related stress (and arguably the risk of developing a mental health condition as a result of exposure to identified stressors) among academic staff and postgraduate students in university settings.

• Survey data indicate that the majority of university staff find their job stressful. Levels of burnout appear higher among university staff than in general working populations and are comparable to "high-risk" groups such as healthcare workers.
• The proportions of both university staff and postgraduate students with a risk of having or developing a mental health problem, based on self-reported evidence, were generally higher than for other working populations.
• Large proportions (>40 per cent) of postgraduate students report symptoms of depression, emotion or stress-related problems, or high levels of stress.

UK national statistics indicate that only 6.2 per cent of staff disclosed a mental health condition to their university, though academics have been found to be among the occupational groups with the highest levels of common mental disorders with prevalence around 37 per cent. It should be noted, however, that prevalence may generally be over-reported in surveys of occupational groups.

Personal Factors That Contribute to Mental Health Outcomes in the Research Workplace

Gender was the key personal factor that emerged as a determinant for mental health (or its reporting), with women reporting more exposure to stress than men, as well as greater challenges around work-life balance. There was also evidence that personality and perceived competence affect mental health as self-critical personalities are more susceptible to stress, though it is also possible that they are more aware of it or more willing to report it. However, it was unclear whether stress was a result of working conditions in the research environment, or whether research settings attracted particular types of individuals. The results on whether age affects mental health were inconclusive, partly as age is often difficult to disentangle from discussions about rank and seniority. Other factors such as disability, sexuality and minority status were mentioned in a small number of articles in the sample, and these articles indicated that these personal factors generally increase stress.

Environmental Factors Commonly Considered in Surveys of Mental Health and Wellbeing in Workplaces

Based on the Health and Safety Executive's framework, and evidence from the wider literature, we identify six key aspects of work that can affect workers' stress levels: work demands, job control, change management, work relationships, support provided by managers and colleagues, and clarity about one's role.

• These aspects of the work environment can be sources of stress or they can help counteract it.
• Findings from studies of university staff and researchers were consistent with the wider understanding of factors that contribute to stress in workplaces.
• Factors including increased job autonomy, involvement in decision making and supportive management were linked to greater job satisfaction among academics, as was the amount of time spent on research. Opportunities for professional development were also associated with reduced stress.

UK higher education (HE) and research staff report worse wellbeing in most of the six aspects, as compared to staff in other sectors.

• In large-scale surveys, UK higher education staff have reported worse wellbeing than staff in other types of employment (including education, and health and social work) in the areas of work demands, change management, support provided by managers and clarity about one's role.
• The only area where higher education staff have reported higher wellbeing in large-scale surveys is in job control, though even here results are mixed across studies. Wide variability was seen among respondents in relation to the level of support provided by managers and colleagues.
• Job insecurity (real and perceived) appears to be an important issue for those working in the research environment, and particularly for early-career researchers, who are often employed on successive short-term contracts.

PhD students face similar challenges to other researchers and higher education staff.

• The main factors associated with development of depression and other common mental health problems in PhD students are high levels of work demands and work-life conflict, low job control, poor support from the supervisor and exclusion from decision making.
• Believing that PhD work is valuable for one's future career helps reduce stress, as does confidence in one's own research abilities.

Some studies suggested that changes to the UK higher education system had brought increased job stress.

• These studies discussed changes that had occurred in the UK higher education system from the 1990s onwards, and had resulted in increased emphasis on accountability, efficiency and performance management. Study authors suggested that these changes could have brought about increases in job stress for staff working in this system.
• However, data explicitly linking the changes to an increase in stress are limited, partly due to a lack of comparable data from before the 1990s.

Staff who can devote a large proportion of their working time to research have better wellbeing.

• Studies found that spending a larger percentage of one's time on research was associated with reduced stress, and that research-only staff reported lower levels of work-life conflict and had better wellbeing than other higher education institution (HEI) staff. However, this may be to some extent confounded by other characteristics of such researchers (e.g. they may be more senior).

Research on emotionally challenging topics can put staff wellbeing at risk.

• Studies showed that staff involved in research on sensitive topics, such as trauma or abuse, may be emotionally affected by the material they encounter in their work and should receive greater support to mitigate the negative impacts of this work.

Outcomes Related to Poor Mental Health and Wellbeing

In addition to considering the extent to which individuals in research environments suffer from mental health issues, it is important for employers and institutions to recognise that these issues have further implications:

• Job stress and poor workplace wellbeing can contribute to reduced productivity—both through absence and, more importantly, through presenteeism, where researchers attend work and are less productive.
• They can also lead to lower levels of commitment to their research and to institutions—which can be seen in high levels of turnover and through negative attitudes in the workplace.
• Effects on job satisfaction are less clear because of the satisfaction researchers gain from intrinsic factors such as the intellectual stimulation of their work. Several studies note that high levels of job-related stress can coexist with high levels of job satisfaction.
• Effects can also spill over into personal and family life.

The overall effects of these negative outcomes on the sector have not been fully quantified, but estimates drawing on broader experience suggest that the costs could be high. An estimate from Shutler-Jones et al (2008) which has several caveats and assumptions, suggests that the costs to the UK HE sector could be more than £500 million per year (c. 5 per cent of the sector's total annual income). Costs to the economy and the country more widely could also be significant due to the lost potential for scientific advances and due to impacts on the availability of research talent if PhD students fail to complete their studies or choose to leave research subsequently.

What Interventions Are Used to Support Researchers, and What Evidence Is There of Their Effectiveness?

Though poor mental health at work is often related to difficulties that are not caused by work (e.g. childhood adversity, family life and other stressors), support in the workplace can offer benefits. However, the evidence around the effectiveness of interventions to support the mental health of researchers specifically is thin. Few interventions are described in the literature and even fewer of those have been evaluated. Where evaluations have been conducted, they are often of limited utility, either because of the evaluation design or the length of follow-up.

Interventions typically focus on stress and wellbeing rather than clinical mental health conditions, reflecting the wider focus in the literature as described above. In addition, the majority of interventions identified aim to support researchers to deal with workplace stress, but they may not be effective in addressing the root causes of that stress or stresses relating to life outside work. The interventions identified can be broadly classified into four groups: policy changes, communication activities, training, and health-promotion activities.

Focusing specifically on the UK, a range of interventions were piloted and evaluated (to a limited extent) as part of a wellbeing initiative by the Higher Education Funding Council for England (HEFCE) around 2009–2011. These offer scope for further investigation and potentially evaluation now that more time has elapsed. Additionally, the project, though completed in 2011, has spawned a network that is now managed by the Universities and Colleges Employers Association (UCEA), which may offer a route to identify further ongoing initiatives and potentially a space to pursue and evaluate efforts to address these issues in the HE sector.

What Are the Strengths and Limitations of the Evidence Base in This Area?

The existing evidence base is limited, meaning it is not possible to draw robust conclusions about the mental health status and needs of researchers, and how researchers may differ from other populations in this regard. More work is needed to understand both the mental health needs of researchers and how they can be addressed. Particular gaps include the effectiveness of interventions, prevalence of specific mental health needs (rather than stress) among researchers, and any evidence about researchers outside the academic setting. There are also limitations to the quality and design of many of the studies conducted, such as lack of long-term follow-up and absence of control groups.

What Next?

Based on the evidence gaps identified and the information available, we suggest the following avenues for further research on this topic:

• Study the prevalence of mental health conditions amongst postdoctoral researchers: Further work on prevalence could use a targeted approach building on the recent work by Levecque et al. (2017), who used a survey to assess the presence of psychological distress and potential psychiatric disorders in a sample of PhD students and compared the results to those of three other sample populations, and Eisenberg et al. (2007), who surveyed a sample of undergraduate and postgraduate university students to assess prevalence of depressive and anxiety disorders and took steps to address the issue of non-response bias. In particular, we suggest a similar study focusing on postdoctoral researchers, a group that is particularly poorly addressed in the existing literature.
• Map mental health policies and procedures at UK HEIs: The current standard of mental health policies and procedures in UK research institutions is not well understood. We suggest that a mapping of the current policies in place across institutions could be valuable, and could build on standards such as those set out in the Mindful Employer Charter (Mindful Employer, 2017).
• Evaluate the interventions introduced through the HEFCE wellbeing and engagement initiative: The wellbeing initiative established by the HEFCE and subsequently maintained as a network by UCEA offers a range of interventions for evaluation. In the project reporting in 2011, many of the institutions noted that it was too soon to tell whether their interventions had been effective. Though these initiatives generally focus on wellbeing rather than clinical mental health conditions, there is scope to explore with the relevant institutions whether those interventions have developed over the years, and whether data are now available (or could be collected) to provide more useful evaluation of the interventions introduced.
• Investigate and develop the HSE management standards as a framework for workplace mental health management in research environments: As well as providing a framework for workplace stress used in several important surveys, the Health and Safety Executive (HSE) have also set out management standards that describe an approach to identifying sources of workplace stress and addressing them at an organisational level. It could be useful to work through that approach with a university or a research organisation to identify the mechanisms at play in those environments. Doing so could establish the relevance of the approach in this context, and potentially provide a model that could be used more widely in the sector.
• Conduct more and higher-quality evaluations of mental health interventions and publish their results: Broadly, better-quality evaluations are needed to identify what works in this area. There is a need for high-quality studies to test the effectiveness of interventions.