key: cord-0933987-9mswj20b
authors: Zerbini, Giulia; Taflinger, Shannon; Reicherts, Philipp; Kunz, Miriam; Sattler, Sebastian
title: Perceived risk of COVID‐19 exposure and poor COVID‐19 prognosis impair sleep: The mediating and moderating roles of COVID‐19‐related anxiety and knowledge
date: 2022-03-02
journal: J Sleep Res
DOI: 10.1111/jsr.13569
sha: c7388304b59f2df30aa976c4a498d8065c8b85d8
doc_id: 933987
cord_uid: 9mswj20b

The ongoing COVID‐19 pandemic has been linked to increased levels of stress, depression, and anxiety in many people around the world. Therefore, identifying individuals at risk of psychosocial burden during this unprecedented crisis is essential for developing prevention measures and treatment options for mental health issues. To this aim, we investigated two risk groups: individuals at higher risk of exposure to the virus and individuals at higher risk of poor prognosis if they contract the virus. We conducted a survey (N = 4167) with a representative sample of the German population and assessed perceived risk of COVID‐19 exposure and poor prognosis if infected, COVID‐19‐related anxiety, problems with sleep and daytime functioning, as well as self‐reported knowledge about the disease. Results showed that perceived risk group membership was linked to increased problems with sleep and daytime functioning via elevated levels of COVID‐19‐related anxiety. This mediated effect was further moderated by self‐reported COVID‐19 knowledge, but only for individuals who rated themselves at higher risk of COVID‐19 exposure. Thus, knowledge buffered the negative effect of exposure risk on anxiety and ultimately on sleep in this risk group. Reaching individuals at increased risk of exposure with clear information about the disease, how to prevent infection, and treatment options could be an effective strategy to contain anxiety levels and promote good sleep, which is important for general well‐being.

The COVID-19 pandemic has been severely impacting lives around the world since its emergence in December, 2019. Over 5 million people have died from the virus (World Health Organization, 2021) and many more have grieved their losses. Efforts to slow the spread of the virus have led to the imposition of lockdowns, subjecting individuals to a variety of changes in their everyday routine, such as sudden school and work closures. From a psychosocial perspective, the pandemic can be seen as a prolonged, intense stressor, as it challenges people on many different levels (e.g., economic, social, health, political) , which in turn, can have negative physical and mental health consequences (Godinic et al., 2020; Pearlin & Bierman, 2013) .

Many studies describe the negative effects of the COVID-19 pandemic on the mental health of the general population (see Xiong et al., 2020 for a systematic review) and especially of healthcare workers, who have been constantly and intensively exposed to the virus (see Sahebi et al., 2021 for an umbrella-review). Anxiety, depression, and stress are commonly reported psychological burdens of the general population in the context of the pandemic (Salari et al., 2020) . Longitudinal research has further demonstrated a causal impact of the pandemic on the worsening of depression symptoms and acute stress (Holman et al., 2020) .

Closely related to physical and mental health, another important aspect of well-being to consider is sleep. Studies exploring the impact of the pandemic on sleep quality have yielded mixed results. As described in a recent systematic review (Lin et al., 2021) , some report a deterioration (Cellini et al., 2020; Robillard et al., 2021; Salfi et al., 2021) and others report no change, or sometimes even improvements (Korman et al., 2020; Leone et al., 2020; Salfi et al., 2021) . Kocevska et al. (2020) found, for example, that the impact of the pandemic on sleep depended on pre-pandemic sleep quality and mood changes, with pre-pandemic good sleepers reporting a worsening of both sleep and mood during the pandemic. The pandemic seems not only to have heterogenous effects on the population, but also on distinct aspects of sleep quality, as some aspects ameliorated (e.g., sleep duration and daytime functioning) and others worsened (e.g., use of sleep medications and sleep efficiency) (Alfonsi et al., 2021) .

The relationship between sleep and anxiety has been well documented over the course of the pandemic with many studies showing that higher levels of anxiety are associated with sleep problems (Al-Ajlouni et al., 2020; Cellini et al., 2020) . Although the relationship between anxiety and sleep problems is likely bidirectional (Alvaro et al., 2013) , one can hypothesize that in the context of a pandemic, first anxiety levels increase and then, as a consequence, sleep is affected. Thus, the question arises what specific pandemic-related factors might decrease sleep quality by increasing anxiety levels.

One potential cause of increased anxiety during the COVID-19 pandemic might be perceived risk group membership, being at higher risk of exposure to COVID-19 (e.g., because of job-related reasons) or at higher risk of poor COVID-19 prognosis in case of infection (e.g., because of pre-existing medical conditions). Perceived risk of contracting the virus as well as perceived risk of dying from COVID-19

were both found to be predictors of increased levels of anxiety and fear (Harper et al., 2020; Lin et al., 2020) . The relationship between increased risk of exposure and poor prognosis and sleep has also been explored in the context of the COVID-19 pandemic. Frontline healthcare workers, who interact directly with COVID-19 patients and, thus, are at higher risk of exposure, have more anxiety and lower sleep quality compared to non-frontline healthcare workers (Lai et al., 2020; Qi et al., 2020; Zerbini et al., 2020) . Similarly, individuals with chronic illnesses, who are at increased risk of poor COVID-19 prognosis, report lower sleep quality than individuals without any chronic illnesses Stanton et al., 2020) . Thus, associations between being at increased risk of exposure or poor prognosis if infected and both anxiety and sleep problems have been reported previously. Here, we test a mediation model exploring the relationships between these three variables and hypothesize that COVID-19-related anxiety can help to explain the relationship between perceived risk group membership and sleep problems.

Additionally, we wanted to explore whether subjective knowledge about COVID-19 moderates the relationship between perceived risk group membership and COVID-19-related anxiety.

Situations that are associated with harm, increased uncertainty, and partially unclear or unknown coping options are often appraised as threatening (cf., Lazarus & Folkman, 1984) . Consequently, we expect individuals who feel informed about the symptoms and consequences of contracting COVID-19 to develop less anxiety that arises from their perceived risk-group membership. Exploring a moderating effect of knowledge is also practically relevant as lack of or incorrect knowledge can be addressed through public health campaigns (Anker et al., 2016) . Therefore, a knowledge-moderated pathway would provide key insights for policy makers seeking to reduce anxiety and insomnia in populations at risk for COVID-19 as well as similar existing and upcoming diseases or health-threats.

Such a moderating role of knowledge has been theorized in the context of the pandemic , but not yet been tested in our study context, and while many studies have examined the effect of the pandemic on psychological health and sleep, to the best of our knowledge, the effect of risk group membership on sleep difficulties mediated by COVID-19 anxiety has also not yet been examined (see Figure 1 depicting the proposed conceptual model). (Zerbini et al., 2022) were collected between December 16 and 29, 2020. At that time Germany was under a lockdown, with schools, hotels, bars, and restaurants closed. Most individuals worked remotely and there were severe restrictions to social gatherings.

We used the seven-item Athens Insomnia Scale for Non-Clinical Application (AIS-NCA; Sattler et al., 2021) . Responses were assessed on a five-point scale. The AIS-NCA provides two subscores, which describe sleep problems (sample item: "I could usually get to sleep (after turning off the lights)..." with response options ranging from "immediately" to "after a very long time") and impaired daytime functioning (sample item:

"Throughout the day, my level of (physical and mental) performance was usually..." with response options ranging from "very good" to "very bad"), and a total score. Items were averaged to calculate each score, with possible values ranging from 1 to 5. Higher scores indicate more problems with sleep and daytime functioning. Reliability was good for both of the two subscales and for the total score (range of Cronbach's α = .77-.85).

Two self-report indicators for belonging to a risk group were used.

The first was higher risk of exposure to the virus and the second was higher risk of poor COVID-19 prognosis. Some examples were given to the participants to clarify the meaning of risk of exposure (e.g., because of working in a hospital or school) and of poor COVID-19 prognosis (e.g., because of suffering from diabetes or a lung disease) (cf., Gouin et al., 2021) . Responses options were "no" [0] and "yes"

[1] in each case.

We used three items of the German version of the Corona Anxiety Scale (CAS; Hölzel & Willenborg, 2020; Lee, 2020) to measure dysfunctional COVID-19 related fear and anxiety. Respondents assessed how frequently they experienced physiologically-based symptoms aroused by COVID-19-related information and thoughts (e.g., "I felt dizzy, lightheaded, or faint, when I read or listened to news about the coronavirus.") on a five-point scale from "not at all"

[1] to "nearly every day over the last 2 weeks" [5] . Responses to the three items were averaged to calculate the total score, with possible values ranging between 1 and 5. Reliability was good (Cronbach's α = .85). 

Two moderated mediation models were computed using the PROCESS macro (Model 7) (Hayes, 2017) percentile bootstrap confidence intervals (95% CI Boot ) (N = 10,000) were used, thereby a 95% CI Boot that does not include zero indicates a statistically significant effect. We report the conditional indirect effects of risk group membership at three different values of COVID-19-related anxiety ("low" at 1 SD below the mean, "average" at the mean, and "high" at 1 SD above the mean). All reported 1 In an elaborate scoring and control process, the panel is subjected to permanent quality control. effects are unstandardized coefficients. We included age and sex as control variables since both variables have been shown to play a role in relation to membership to the aforementioned risk groups (eg., Rommel et al., 2021) , fear of COVID-19 (eg., Niño et al., 2021) , and sleep (eg., Luca et al., 2015) .

Nearly every fifth (18.33%) respondent perceived themselves as being at risk of exposure to COVID-19, while three out of ten (30.19%) perceived themselves as being at risk of a poor COVID-19 prognosis.

While the majority of the respondents (89.3%) reported they (almost) never experience anxiety symptoms, approximately 1% did so almost every day during the past 2 weeks. No or hardly any problems with sleep and daytime functioning occurred in 16.3% of the respondents, while 5.3% reported (very) severe problems. More of such problems were reported by people who perceived themselves at high risk of exposure to COVID-19 or a poor COVID-19 prognosis (see Figure 2 ).

Self-reported COVID-19 knowledge was high, with 22.7% of the respondents choosing the two highest options (very high knowledge) on the 11-point scale, while only 1.0% indicated the two lowest options (very low knowledge). See Table 1 for further descriptive statistics and pairwise correlations of the examined variables.

The mediator model reported in Model 1 (Table 2) 

The mediator model reported in Model 2 (Table 2) 

Previous research suggests that the COVID-19 pandemic had heterogenous effects on sleep health in the population (Lin et al., 2021) . In More specifically, we found that two in ten individuals classified themselves as being at increased risk of COVID-19 exposure and three in ten as being at increased risk of poor COVID-19 prognosis. The respective type of risk group is indicated in the header of the table; pathway a moderated by self-reported COVID-19 knowledge; 95% CI Boot , 95% percentile bootstrap confidence interval (N = 10,000); SE Boot , percentile bootstrap standard error.

Anxiety levels were also positively associated with sleep problems and daytime malfunctioning, consistent with previous studies (Al-Ajlouni et al., 2020; Cellini et al., 2020) , and mediated the relationship between perceived risk and sleep. Individuals who perceived themselves at increased risk of COVID-19 exposure (especially those with low self-reported knowledge about COVID-19) reported increased COVID-19 anxiety levels, which in turn negatively impacted their sleep. The same mediation pathway (without the moderating effect of knowledge) was found for individuals at risk of poor prognosis.

We also found a residual direct effect of risk group membership on sleep, indicating that belonging to one of the two risk groups was associated with sleep problems independent of anxiety. A potential cause of the direct effect of risk of exposure on sleep could be the increased workload that individuals in certain jobs faced because of the pandemic (Bell et al., 2021) . Similarly, pre-pandemic research shows that many chronic illnesses (e.g., heart disease and high blood pressure which are risk factors for poor COVID-19 prognosis) are associated with insomnia (Taylor et al., 2007) .

COVID-19 moderated the association between risk of exposure and COVID-19-related anxiety. Information about how to prevent infection from COVID-19 (e.g., social distancing and wearing F I G U R E 3 Predicted values (with standard errors) of COVID-19related anxiety (CAS) depending on risk of exposure to COVID-19 (left) and risk of poor COVID-19 prognosis (right) and as a function of different levels of self-reported COVID-19 knowledge. Notes:

The range of the COVID-19-related anxiety scale is 1-5, with higher scores indicating higher levels of anxiety. Self-reported COVID-19 knowledge is predicted at three different values: "low" at 1 SD below the mean □, "average" at the mean , and "high" at 1 SD above the mean ■. (Cheng et al., 2021) . The study also highlights the challenges of obtaining correct and reliable information, as well as the complex interaction between information coping styles and strategies in determining how individuals respond to the overwhelming amount of information available.

We also acknowledge the limitations of our study. First, we used singleitem instruments to assess perceived risk group membership concerning exposure to the COVID-19 virus and poor prognosis in case of infection.

Although we provided respondents with exemplary indicators, future studies could provide a full list of indicators to improve the validity of such self-attributed categorizations. Similarly, we also assessed subjective knowledge about COVID-19 with one-single item, which only captured general knowledge, and could be improved using a more detailed assessment with multiple items (e.g., about effective ways to prevent infection, treatment options, consequences of specific symptoms, or local incidence rates). Second, the aforementioned measures as well as the sleep and anxiety measures were subjective, thus prone to self-report biases. Although subjective measures are highly relevant (especially when assessing psychological health), future studies could complement them with objective measures (e.g., actimetry to assess sleep and knowledge tests to assess knowledge about COVID-19). A third limitation of this study concerns self-selection into online surveys, which could be addressed in future studies by using probability samples drawn from the general population. In addition, given the retrospective and cross-sectional nature of the data, longitudinal research re-examining the relationships found in this study is warranted. Longitudinal analyses are also needed to identify long-term effects of the pandemic on sleep and other indicators of physical and psychological health.

Altogether, this study contributes to our understanding of how the COVID-19 pandemic has particularly affected specific groups of individuals and stimulates ideas for possible interventions. Sleep quality of individuals that perceived themselves at risk of COVID-19 exposure or at risk of poor COVID-19 prognosis could be improved by decreasing COVID-19-related anxiety levels. Public health campaigns which provide information about the virus and protection measures (Anker et al., 2016) could be one strategy not only to prevent infections, but also to provide the general population, and especially individuals at risk, with practical tools to protect themselves while simultaneously lowering their anxiety levels. Decreasing anxiety levels and, thereby, improving sleep quality is an important goal to promote well-being, given the central role of sleep in mental and physical health (Freeman et al., 2020) .

We thank those who helped conduct this study, especially Dina

Maskileyson and Floris van Veen for programing the survey. Open access funding enabled and organized by ProjektDEAL.

The Authors declare that there is no conflict of interest. 

Informed consent was obtained from all study participants.

The data that support the findings of this study are openly available in "PUB-Publikationen an der Universität Bielefeld" (Zerbini et al., 2022) .

https://orcid.org/0000-0002-5348-9212 Sebastian Sattler https://orcid.org/0000-0002-6491-0754

Anxiety and depressive symptoms are associated with poor sleep health during a period of COVID-19-induced nationwide lockdown: A cross-sectional analysis of adults in Jordan

COVID-19 lockdown and poor sleep quality: Not the whole story

Disparities in coronavirus 2019 reported incidence, knowledge, and behavior among US Adults

A Systematic review assessing bidirectionality between sleep disturbances, anxiety, and depression

Measuring the effectiveness of mass-mediated health campaigns through meta-analysis

Challenges facing essential workers: A cross-sectional survey of the subjective mental health and well-being of New Zealand healthcare and 'other' essential workers during the COVID-19 lockdown

Changes in sleep pattern, sense of time and digital media use during COVID-19 lockdown in Italy

Maladaptive coping with the infodemic and sleep disturbance in the COVID-19 pandemic

Effects of economic uncertainty on mental health in the COVID-19 pandemic context: social identity disturbance, job uncertainty and psychological well-being model

Socio-demographic, social, cognitive, and emotional correlates of adherence to physical distancing during the COVID-19 pandemic: A cross-sectional study

Functional fear predicts public health compliance in the COVID-19 pandemic

Introduction to Mediation, Moderation, and Conditional Process Analysis, 2nd ed. A Regression-Based Approach

The unfolding COVID-19 pandemic: A probability-based, nationally representative study of mental health in the United States

Sleep quality during the COVID-19 pandemic: Not one size fits all

COVID-19-mandated social restrictions unveil the impact of social time pressure on sleep and body clock

Factors associated with mental health outcomes among health care workers exposed to coronavirus disease 2019

Stress, Appraisal, and Coping

Coronavirus anxiety scale: a brief mental health screener for COVID-19 related anxiety

Effects of lockdown on human sleep and chronotype during the COVID-19 pandemic

Knowledge, attitudes, impact, and anxiety regarding COVID-19 infection among the public in China

Burden of sleep disturbance during COVID-19 Pandemic: A systematic review

Age and gender variations of sleep in subjects without sleep disorders

Race and ethnicity, gender, and age on perceived threats and fear of COVID-19: Evidence from two national data sources. SSM -Population Health

Current Issues and future directions in research into the stress process

The evaluation of sleep disturbances for Chinese frontline medical workers under the outbreak of COVID-19

Profiles of sleep changes during the COVID-19 pandemic: Demographic, behavioural and psychological factors

Population with an increased risk of severe COVID-19 in Germany

The prevalence of anxiety and depression among healthcare workers during the COVID-19 pandemic: An umbrella review of meta-analyses

Prevalence of stress, anxiety, depression among the general population during the COVID-19 pandemic: A systematic review and meta-analysis

Demographic, psychological, chronobiological, and work-related predictors of sleep disturbances during the COVID-19 lockdown in Italy

Public stigma toward people with drug addiction: a factorial survey

Assessing sleep problems and daytime functioning: A translation, adaption, and validation of the athens insomnia scale for non-clinical application (AIS-NCA)

Sleep in times of crises: A scoping review in the early days of the COVID-19 crisis

Depression, anxiety and stress during COVID-19: Associations with changes in physical activity, sleep, tobacco and alcohol use in Australian adults

Psychological impact of the COVID-19 pandemic on health care workers in Singapore

Comorbidity of chronic insomnia with medical problems

Response effects of prenotification, prepaid cash, prepaid vouchers, and postpaid vouchers: An experimental comparison

WHO Coronavirus (COVID-19) Dashboard

Impact of COVID-19 pandemic on mental health in the general population: A systematic review

Psychosocial burden of healthcare professionals in times of COVID-19 -a survey conducted at the University Hospital Augsburg

Data publication for: Perceived risk of COVID-19 exposure and poor COVID-19 prognosis impair sleep: The mediating and moderating roles of COVID-19-related anxiety and knowledge