key: cord-0839498-m8w3gwmu
authors: Kennedy, Kathryn E. R.; Bastien, Célyne H.; Ruby, Perrine M.; Killgore, William D. S.; Wills, Chloe C. A.; Grandner, Michael A.
title: Nightmare content during the COVID‐19 pandemic: Influence of COVID‐related stress and sleep disruption in the United States
date: 2021-08-18
journal: J Sleep Res
DOI: 10.1111/jsr.13439
sha: 6cb458881ee8128ae26263d4344641b08f5b6556
doc_id: 839498
cord_uid: m8w3gwmu

Nightmares are often associated with psychiatric disorders and acute stress. This study explores how the COVID‐19 pandemic may have influenced the content of nightmares. A sample of N = 419 US adults completed online surveys about sleep and COVID‐19 experiences. Participants were asked about the degree to which they agreed with statements linking greater general stress, worse overall sleep and more middle‐of‐the‐night insomnia with the COVID‐19 pandemic. They were also asked if, during the pandemic, they experienced nightmares related to various themes. Logistic regression analyses examined each nightmare content as outcome and increased stress, worse sleep and more middle‐of‐the‐night insomnia as predictors, adjusted for age, sex and race/ethnicity. Those who reported greater general COVID‐related stress were more likely to have nightmares about confinement, failure, helplessness, anxiety, war, separation, totalitarianism, sickness, death, COVID and an apocalypse. Those who reported worsened sleep were more likely to have nightmares about confinement, oppression, failure, helplessness, disaster, anxiety, evil forces, war, domestic abuse, separation, totalitarianism, sickness, death, COVID and an apocalypse. Those who reported worsened middle‐of‐the‐night insomnia were more likely to have nightmares about confinement, oppression, failure, helplessness, disaster, anxiety, war, domestic abuse, separation, totalitarianism, sickness, death, COVID and an apocalypse. These results suggest that increased pandemic‐related stress may induce negatively‐toned dreams of specific themes. Future investigation might determine whether (and when) this symptom indicates an emotion regulation mechanism at play, or the failure of such a mechanism.

A nightmare is a bad dream that evokes a strong emotional response in the sleeper and leads to being awoken. Nightmares are associated with particular personality types more susceptible to threat perception and proposed by Nielsen and Levin (2007) to involve dysfunction of a network comprising the limbic, paralimbic and prefrontal regions of the brain involved in emotion regulation. A nightmare may either be idiopathic (of unknown cause) or post-traumatic (following a traumatic event) and typically takes place during rapid eye movement (REM) sleep (Fisher et al., 1970; Spoormaker et al., 2006) .

Nightmares are more common in children than adults (20-30% vs.

11% of the general population) (Bixler et al., 1979; Leung & Robson, 1993) , but those who experience them in adulthood are most likely to be young females (Spoormaker et al., 2006) . This prevalence in females is thought to decline with age (Spoormaker et al., 2006) .

Nightmares are significantly associated with psychiatric disorders and anxiety, with persistence of nightmares linked to irregular sleep schedules, avoidance of addressing them, and fear of going to sleep (Hublin et al., 1999; Spoormaker et al., 2006) . This can result in sleep deprivation, which increases anxiety levels and thus perpetuates the cycle (Pires et al., 2016) . Robert and Zadra (2014) conducted a study to assess the most common nightmare themes and found these to include aggression, fear and unfortunate endings. Interestingly, they found that nightmares were less likely to reflect issues of personal competence and more likely to reflect negative external triggers. This falls in line with the Threat Simulation Theory, which suggests that a key function of dreams is to play out threatening scenarios while asleep and thus help prepare for and avoid or confront a threat in waking life (Revonsuo, 2000) .

A general increase in dream and nightmare frequency during the COVID-19 pandemic has been reported around the world. Scarpelli and colleagues (2021) and Gorgoni and colleagues (2021) both noted an increase in dreams with strong emotional content in Italy. In a sample of Canadian university students, an increase in nightmares was reported, with females disproportionately affected (Kilius et al., 2021) . Another Canadian study noted an increase in dream imagery related to pandemic themes among participants (MacKay & DeCicco, 2020) .

In the survey during the first lockdown in France, an increase in nightmare frequency was also observed (Ruby, 2021) . Themes related to the health crisis were the most frequently reported. This was also seen in a Brazilian study that analyzed dream themes and found a higher proportion of themes around contamination and cleanliness (Mota et al., 2020) , and in a Finnish study where participants reported more night-time awakenings and nightmares as compared to pre-pandemic times, with over 50% of these nightmares about pandemic-specific themes, including sickness, physical distancing and family concerns (Pesonen et al., 2020) . Among adolescents living in Italy, Croatia and Romania, an increase in nightmares and dreams with pandemic-related themes were also reported (Guerrero-Gomez et al., 2021) .

General sleep disturbances during the pandemic have been particularly prevalent among healthcare workers (Wang et al., 2020) , whereas other studies have found females in general to be at higher risk (Dal Santo et al., 2021; Kilius et al., 2021; Pinto et al., 2020; Scarpelli et al., 2021) . A large study using Fitbit data from approximately 160,000 users across six major US cities found that compared to 2019, sleep duration had increased during the 2020 pandemic (Rezaei & Grandner, 2021) . Bedtime also shifted to later and variability in sleep duration decreased. Increased sleep time was also observed in France during the first lockdown, as seen in a survey that covered all regions of the country (N = 3337) (Ruby, 2021) .

In another study comparing pandemic-related sleep to prepandemic norms, a team from the Netherlands recruited a population consisting of both good and bad sleepers (for which previous data existed) and assessed subjective sleep quality changes during the pandemic. Kocevska and colleagues (2020) found no uniform effect on sleep, with some individuals experiencing improvements as compared to pre-pandemic times, whereas others noted worse sleep that was often associated with pandemic anxiety. This was also seen in Italy, France and Spain, where worse sleep quality was more frequently reported than improved sleep quality Dal Santo et al., 2021; Ruby, 2021) .

More time than usual spent indoors could be responsible for some of the sleep disturbances that have been seen. Morin and colleagues (2020) postulated that not only could stress be responsible, but also disturbed circadian rhythms as a result of confinement. The usual light and social cues used for circadian entrainment are no longer present when individuals are restricted to extended periods of social distancing, at home, indoors. With this lifestyle change has also come increased reliance on the use of screens for work and social connections (Majumdar et al., 2020) . Blue-light-emitting devices are known to suppress melatonin production and thus disturb sleep (Cajochen et al., 2011) . These factors may have contributed to the increase in sleep medication use seen during confinement by Alfonsi and colleagues (2021).

Given the known implications of stress and sleep loss on health and well-being, and the need for further understanding of the pathology of nightmares, the aim of the present study was to determine whether COVID-related general, family or media stress, subjective perception of worsened sleep, increased onset insomnia or middle-of-the-night insomnia were associated with specific nightmare themes in a United States sample after seeing similar effects in other parts of the world. We hypothesized that pandemic-related worsening of sleep quality and pandemic-related increased stress would be associated with increased frequency of nightmares that would similarly reflect pandemic-related themes.

These data were collected as part of the Coronavirus and Impact on Dreams (CoVID) study, which was approved by the Institutional Review Board (IRB) of the University of Arizona. Data were collected from June to November 2020. Adults aged 18 and older who were also fluent in English were eligible. Recruitment included flyers posted locally in Tucson, Arizona, as well as social media and online advertisements accessible nationwide. Data were collected in the form of a REDCap survey, for which informed consent was required in order to proceed. Analyses were conducted on all completed surveys (N = 419).

Participants were asked to indicate the degree to which they agreed or disagreed with the following statements about COVID-related stress: "COVID-19 is increasing my general stress," "COVID-19 is increasing stress about my family" and "The media coverage of COVID-19 has increased my stress." They were also asked to indicate the degree to which they agreed or disagreed with the following statements about their sleep: "Since quarantine, my sleep has worsened," "Since quarantine, I'm struggling to fall asleep" and "Since quarantine, I'm waking up more during the night." These six items, which record general stress, family stress, media stress, worsened sleep, onset insomnia and middle-of-the night insomnia, respectively, had the following response options: "Strongly Disagree," "Disagree," "Agree" and "Strongly Agree." These were coded numerically as 0-3. Participants were also asked, "Have you had nightmares about any of the following during quarantine?" They indicated yes or no to the following themes inspired from the French survey (Ruby, 2021) : "confinement," "claustrophobia," "suffocation," "oppression," "drowning," "failure," "feeling helpless," "natural disasters," "anxiety," "evil forces," "war," "domestic abuse," "separation from loved ones," "totalitarian regime," "being chased," "dangerous animals," "sickness," "death," "covid-19 virus" and "end of the world."

Participants completed the Patient Health Questionnaire (PHQ-9) (Kroenke et al., 2001) , which screened for depression, and the Generalized Anxiety Disorder 7-item instrument (GAD-7) (Spitzer et al., 2006) . Participants also completed the Mannheim Dreams questionnaire (MADRE) (Shredl et al., 2014) , which asked about dream and nightmare experiences, including frequency of nightmares. 

Logistic regression analyses examined the likelihood of each nightmare as a binary variable, predicted by each of the stress and sleep variables, treated as pseudo-linear independent variables. All logistic regression analyses were adjusted for age, sex and race/ethnicity.

Results are displayed as adjusted odds ratios (ORs) and 95% confidence intervals (CIs), along with z score of the OR and p-value.

Because there were 20 different outcomes, all p-values were subject to Holm-Bonferonni adjustment (Holm, 1979 

Results of logistic regression analyses investigating relationships between COVID-related stress and nightmare content are reported in Media stress was not associated with an increased likelihood of any specific nightmare content.

Results of logistic regression analyses investigating relationships between COVID-related sleep disturbances and nightmare content are reported in Table 3 . In unadjusted analyses, following a p-value adjustment for multiple comparisons, worse sleep was associated with an increased likelihood of nightmares about confinement, oppression, failure, helplessness, natural disasters, anxiety, evil forces, war, domestic abuse, separation from loves ones, totalitarian regimes, sickness, death, COVID-19 and an apocalypse. After adjustment for demographic covariates, the relationships with evil forces and domestic abuse were no longer statistically significant in Holm-Bonferroniadjusted analyses. Z-score plots for sleep disturbances and nightmare theme associations are presented in Figure 2 .

of nightmares about all topics except dangerous animals. After adjustment for demographic covariates, the relationships with suffocation, drowning, evil forces, being chased, sickness and COVID-19 were no longer statistically significant in Holm-Bonferroni-adjusted analyses.

Middle-of-the-night insomnia was associated with an increased likelihood of nightmares about confinement, oppression, failure, helplessness, anxiety, war, domestic abuse, separation from loved ones, sickness, death, COVID-19 and an apocalypse. After adjustment for demographic covariates, the relationships with confinement, COVID-19 and an apocalypse were no longer statistically significant in Holm-Bonferroni-adjusted analyses.

To determine the effects of depression and anxiety on nightmare content, logistic regression analyses investigating relationships between PHQ-9 and GAD-7 scores and nightmare themes were performed. These are reported in 

Nightmare frequency was assessed using the Mannheim Dream questionnaire (MADRE), with the question: "How often have you experienced nightmares recently (in the past several months)?"

The answer options included: several times a week, about once a

week, two to three times a month, once a month, two to four times a year, once a year, less than once a year, and never. Data were grouped into: more than once a week, more than one a month, more than one a year, and less than one a year, and assessed as an additional covariate in the models. In almost all cases, the relationship between stressor and nightmare content, or sleep complaint and nightmare content was maintained, indicating there was no influence of nightmare frequency on specific nightmare themes (unreported data). These findings suggest that increased stress or dysregulated emotions may result in more frequent nightmares and that the pandemic may have given rise to specific associated nightmare themes.

Although this was also seen in locations outside the United States Mota et al., 2020; Ruby, 2021; Scarpelli et al., 2021) , unique sociopolitical stressors specific to the United States occurring in parallel may have also influenced the themes of totalitarianism, war and helplessness.

Data from Germain and Nielsen (2003) indicating that posttraumatic nightmares were more closely associated with insomnia than idiopathic nightmares suggest that the findings seen in the present study are indicative of traumatic responses to pandemic-related events. Uncertainty about the future could have led to nightmares about helplessness, totalitarianism or war. Experiencing concerns regarding personal or familial sickness are also possible sources of anxiety that can manifest in negatively toned dreams. Interestingly, media stress was generally not associated with an increased likelihood of any specific nightmare themes. Rather, general and family stress seemed to evoke the strongest responses. This suggests that nightmares could be less influenced by societal issues portrayed in the media and more closely influenced by personal issues that pose imminent threats to an individual's health and well-being.

Negatively toned dreams are associated with higher stress levels (De Koninck & Brunette, 1991) , and nightmares are accompanied by increased autonomic activity and sympathetic tone (Paul et al., 2019; Perogamvros et al., 2019) . Although it's unclear which specific mechanisms are involved in the brain that experiences frequent nightmares, as outlined by Nielsen and Levin (2007) , norepinephrine may be involved. The success of the drug Prazosin in reducing posttraumatic stress disorder (PTSD)-related nightmare frequency could be a result of decreased norepinephrine, although this remains to be determined (Spoormaker et al., 2006) .

The hormone cortisol may also play a role in nightmares. One study found that frequent nightmares were associated with a blunted cortisol awakening response in women, who comprised 74% of our sample (Nagy et al., 2015) . Cortisol is known to act on the hypothalamic-pituitary-adrenal (HPA) axis, suggesting that the HPA may be involved in nightmare generation. The association between disrupted cortisol levels and sleep disturbances, including insomnia, has also been well documented (Drake et al., 2017; Grimaldi et al., 2020) . If individuals have insufficient or poor-quality sleep, emotion processing in the brain can become impaired (Walker & van der Helm, 2009 Measures of sleep were also limited to very few questions, thus limiting the extent to which disturbances and poor sleep could be determined. Lastly, this survey was advertised largely online and those without internet access or those without social media accounts will have been less likely to have participated. Although online survey distribution is now commonplace, it's important that future studies take this portion of the population into account, so as not to exclude any individuals.

The mechanisms underpinning the emotion processing involved in dream manifestation -or perhaps the lack thereof -during a stress response, or as seen in sleep loss, are not yet fully understood.

However, this study highlights the effects of stress and reported sleep difficulties on specific nightmare themes, particularly with regards to a global pandemic or similar health crisis and the sociopolitical stressors that occurred in parallel within the United

States. This may be beneficial for future sleep and mental health interventions.

There are no conflicts of interest to disclose. 

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Kathryn E. R. Kennedy https://orcid.org/0000-0001-5087-3780

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

Prevalence of sleep disorders in the Los Angeles metropolitan area

Evening exposure to a light-emitting diodes (LED)-backlit computer screen affects circadian physiology and cognitive performance

Early Impact of the COVID-19 Outbreak on Sleep in a Large Spanish Sample

Presleep suggestion related to a phobic object: successful manipulation of reported dream affect

Familial risk for insomnia is associated with abnormal cortisol response to stress

A psychophysiological study of nightmares

Sleep pathophysiology in posttraumatic stress disorder and idiopathic nightmare sufferers

Pandemic dreams: quantitative and qualitative features of the oneiric activity during the lockdown due to COVID-19 in Italy

Autonomic dysregulation and sleep homeostasis in insomnia

Dreaming in adolescents during the COVID-19 health crisis: Survey among a sample of European School Students

A simple sequentially rejective multiple test procedure

Nightmares: Familial aggregation and association with psychiatric disorders in a nationwide twin cohort

Pandemic nightmares: COVID-19 lockdown associated with increased aggression in female university students' dreams

Suicidal ideation during the COVID-19 pandemic: The role of insomnia

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

The PHQ-9: Validity of a brief depression severity measure

Pandemic dreaming: The effect of COVID-19 on dream imagery, a pilot study

COVID-19 pandemic and lockdown: Cause of sleep disruption, depression, somatic pain, and increased screen exposure of office workers and students of India

Sleep and circadian rhythm in response to the COVID-19 pandemic

Dreaming during the Covid-19 pandemic: Computational assessment of dream reports reveals mental suffering related to fear of contagion

Frequent nightmares are associated with blunted cortisol awakening response in women

Nightmares: A new neurocognitive model

Nightmares do result in psychophysiological arousal: A multimeasure ambulatory assessment study

Impact of the novel coronavirus (COVID-19) pandemic on sleep

Increased heartbeat-evoked potential during REM sleep in nightmare disorder

Pandemic dreams: Network analysis of dream content during the COVID-19 lockdown

Sleep quality in times of Covid-19 pandemic

Effects of acute sleep deprivation on state anxiety levels: A systematic review and meta-analysis

The reinterpretation of dreams: an evolutionary hypothesis of the function of dreaming

Changes in sleep duration, timing, and variability during the COVID-19 pandemic: Large-scale Fitbit data from 6 major US cities

Thematic and content analysis of idiopathic nightmares and bad dreams

Ce que le rêve nous apprend sur le vécu des françaises et des français

Pandemic nightmares: Effects on dream activity of the COVID-19 lockdown in Italy

Not so sweet dreams: Adults' quantity, quality, and disruptions of sleep during the initial stages of the COVID-19 pandemic

The Mannheim Dream questionnaire (MADRE): Retest reliability, age and gender effects

A brief measure for assessing generalized anxiety disorder: the GAD-7

Nightmares: From anxiety symptom to sleep disorder

Overnight therapy? The role of sleep in emotional brain processing

Sleep disturbances among medical workers during the outbreak of COVID-2019

Nightmare content during the COVID-19 pandemic: Influence of COVID-related stress and sleep disruption in the United States