Research and mental health during COVID-19—advice and some requests

This journal receives many papers on COVID-19 research on mental health difficulties every week and few make it through the editors to receive peer review. At one time this was 80 or 90 submissions a week and now it is a trickle of 30–40. We spend a lot of time wading through them each week and so have decided to provide some advice to researchers interested in this area. This might help submissions pass through our editorial and peer review filters and authors to produce helpful publications on how to deal with mental health of the public during a pandemic.

The context for research during COVID-19

Mental health research usually depends on some face-to-face contact, to help people to complete questionnaires, participate in in-depth interviews or to provide treatment, although scientists were moving to web and app-based data collection pre-COVID. The pandemic disrupted traditional research practices and accelerated the switch to remote working and remote assessment if our studies were to continue. Many studies were not able to adapt and came to an abrupt halt. These often depended on the good will of clinical staff to recruit participants into the studies who due to high workloads and new ways of working were unable to deliver anything except clinical care under the constraints of COVID-19. Other studies staggered, but then successfully “pivoted” to new systems for recruitment and assessment with the ingenuity and creativity of researchers, and with the generosity of people using mental health services. As well as being a challenge for research, the pandemic provided opportunities to understand how a global stressor affects the public’s mental health. But what was this stressor and how have we interpreted it?

Before effective treatments and vaccines were discovered, the pandemic was a shared public stressor—anyone could catch it and suffer the consequences. Good sense meant we stopped socialising as much in person, even without governments’ mandating lockdowns. It was a little while before we learned that some groups were more at risk of severe consequences from the SARS-CoV-2 virus, including death. Early on older people, those with physical and mental health diagnoses (including a diagnosis of schizophrenia (Bitan et al., 2021) or mood disorder) and people who deliver basic and frontline services (e.g. grocery store workers, health care providers) and within these jobs those from minority ethnic groups were highlighted as at greatest risk (Mathur et al., 2021). Some of these groups were shielded by being at home and limited their social contact, others like bus and delivery drivers, personal care workers and most in health care could not do this and had to continue working and be prepared for the risks. For those shielding, these protective actions reduced the threat of catching COVID-19, but for many also had a negative effect on their mental health by reducing social contacts. The lockdowns across the world also produced other social pressures by limiting finances and access to the work environment and supplies (Sheridan Rains et al., 2021). These different types and levels of stressors, and their mitigation through community supports grew during the pandemic. These factors should be the biopsychosocial markers of interest to mental health researchers. The length of time of the pandemic and the different layers of lockdown in different parts of the world also add to the opportunities for mental health studies to understand how to alleviate and even to prevent mental health difficulties arising in the future.

Unemployment following the economic downturn also adds to the problems by increasing loneliness (Li & Wang, 2020) and financial insecurity which contribute to and maintain mental health problems (Bond & Holkar, 2020). We know that some contact with the news and social media helps with information, but too much time searching the web is associated with mental distress (Geirdal et al., 2021) (although that relationship is circular and difficult to disentangle, and related to many different variables including secondary traumatization).

The Journal of Mental Health has and continues to receive submissions about the mental health consequences of COVID-19. We have published several editorials and book reviews on this topic over the last year (e.g., Byrne & Wykes, 2020; Guha, 2021; Lim, 2021; Taggart et al., 2021; Willis & Chalder, 2021) in addition to peer reviewed research from around the globe. The research has told us that the digital divide still exists and interferes with remote therapy (e.g., (Watson et al., 2021), that parental responses are important for suicidal ideation, particularly during lockdowns when parents spend more time with their children (Chiang et al., 2021), that a SARS-CoV-2 infection produced anxiety and depression, and for those with chronic medical conditions, the depression lasted a long time (Shousha et al., 2021). We also published important studies on the mental health effects in groups of people who were at high risk of infection such as health care workers (Alsolais et al., 2021; Patelarou et al., 2021; Villalba-Arias et al., 2021) as well as the effects of lockdown in different countries (Lawal, 2021; Lopez Steinmetz et al., 2021; Sapara et al., 2021; Schou-Bredal et al., 2021). These studies were carried out early in the pandemic, were convenience samples and mostly used well-known mental health measures which allows them to be put into context. Our previous understanding of dangerous infectious diseases comes from studies of SARS and Ebola carried out in Africa, Canada, and Asian countries. It turns out that the mental health effects are similar, especially for frontline healthcare workers and those who survived the disease, with the most prominent mental health effects being post-traumatic stress disorder, continued high levels of anxiety and intermittent sleep problems and depression. Although these difficulties resolved for most, some people continued to experience difficulties.

In addition, we have published studies on the mitigating effects on coping, perceived social support etc (Budimir et al., 2021; Burton et al., 2021; Simblett et al., 2021) and a paper and a series of commentaries on what the research priorities and goals should be following the pandemic (Inchley et al., 2021; King, 2021; Shenow et al., 2021; Teesson & Morgan, 2021; Wykes et al., 2021).

The Journal of Mental Health has already contributed to an understanding of the immediate and the short-term effects of the pandemic and associated lockdown and we want to do so in the future.

Research limitations

During a pandemic there were few opportunities for contact with the public, so researchers learnt quickly how to shift to online methods. Some of these studies were at the request of governments to inform decision making (Johnson et al., 2021). These, of course, have problems that we have discussed above and at the very beginning of the pandemic (Torous & Wykes, 2020). Some of the previous criticisms of observational data in this journal (Wykes et al., 2019) are closely related to COVID-19—data collected with online surveys. Firstly, and most obviously, the sample may not be representative of the population. To take part you need to see the survey advertisement, be willing to complete it and, of course, provide valid data (including gender). This journal has received multiple cross-sectional papers describing a sample of nurses, medical students, the general population, pregnant women and more, without taking this representativeness limitation into account. Large numbers do not mean the data are representative or any more important than a perfectly formed representative small sample. For example, young people who are more digitally skilled may be able to complete these surveys, but choose not to, and older people may not even see an advertisement for a survey because they rarely go online. Even during COVID-19 digital skills and the digital divide are likely to prevent or promote participant inclusion. Most surveys are also completed predominantly by women, presumably because they are more involved with social media and more altruistic, but this does mean we have a gap—what are likely to be the mental health effects in men? A simple solution to this problem would be purposive sampling. Most submissions to this journal do not address this representation problem.

A further limitation is the study measures. They are often ones produced by the study authors that have no information about their reliability or validity. If authors used some consistent, psychometrically sound measures then there would be some ability to compare different studies by sharing samples with similar characteristics. There may be other limitations, but we will not labour the point. We want scientists to take an objective look at their data and analysis and consider whether it uses sound methods. Then decide whether the paper adds to what we already know from the 11,469 studies currently available after a quick search of Web of Science using the terms “COVID-19” and “mental health”.

Potential future studies

The pandemic has also been in our lives for the past 20 months and with the appearance of more variants such as Omicron is likely to be here for a while longer. This context provides ample opportunity to develop longitudinal research addressing questions such as the level of continued resilience, the role of community support and engagement, reactions to the vaccine rollout and reactions to varying lockdowns. We look for these types of studies, but rarely receive them.

We also now need to see more interesting data sets and analyses. Lockdowns differ between countries and some of their characteristics may promote wellbeing or have detrimental effects on mental health. We have seen few of these studies submitted to our journal, although some researchers have taken this approach to provide more definitive data on the social and economic supports that mitigate the negative mental health effects of lockdowns (Brulhart et al., 2021).

We already know quite a lot about behaviour, immediate mental health effects and social consequences of lockdown and, in cross-sectional models, the effects on mental health. We do not need more of this research. This is a call for mental health scientists to be more creative. If you have cross-sectional data, expand it into a longitudinal study if you can, consider how you might compare your data with others by using similar measures and investigate how varying levels of risk produce mental health effects in different groups of people. This last suggestion is easy even if using convenience samples like medical students. For example, these students in different years of study vary in their contact with the public or spend more or less time in healthcare settings. This is a varying risk factor and could be explored both cross-sectionally and over time. This study could provide us with information about how to titrate the support students need depending on their year of study. You could investigate new topics like vaccine hesitancy and the impact of prior mental health—especially as people with particular diagnoses and experiences (Bond & Holkar, 2020) are likely to suffer serious consequences from COVID-19.

This has been a difficult time for mental health researchers with a lack of access to the usual research methods, but that doesn’t mean we drop our standards. We know that there are mental health effects, and they are similar across different groups of people. The pandemic is not over, new variants are appearing and will produce further mental health effects. We are depending on science not just to discover treatments and vaccines for the physical health effects but to generate more information to help understand and respond to the pandemic’s mental health consequences.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

The author(s) reported there is no funding associated with the work featured in this article.