What goes around comes around: Nightmares and daily stress are bidirectionally associated in nurses

Abstract

Nurses may experience frequent nightmares due to stressful work environments. Nightmares may also exacerbate stress among nurses, although this has yet to be tested empirically. We examined daily bidirectional associations between stress severity and nightmares, and moderation by posttraumatic stress disorder (PTSD) symptoms. 392 nurses (92% female; 78% White) completed 14 days of sleep diaries to assess previous-day stress severity and nightmare occurrence and severity. PTSD symptoms were assessed at baseline. Multilevel models were used to examine bidirectional, within-person associations between daily stress and nightmares, and cross-level moderation by baseline PTSD symptoms. 47.2% of nurses reported at least one nightmare across the two weeks. Days with greater stress were associated with higher odds of experiencing a nightmare (OR = 1.22, p = 0.001), as well as greater nightmare severity (b = 0.09, p = 0.033). Nightmare occurrence (b = 0.15, p < 0.001) was associated with greater next-day stress severity. Daily stress and nightmare associations were similar for those with and without PTSD symptoms. Nightmares and stress may occur in a bidirectional fashion. Nurses face intense occupational demands and frequent exposure to potentially traumatic events. Studies should explore whether targeting nightmares and stress may improve nurses’ health and well-being.

Nightmares, defined as disturbing dreams that result in awakening, are common among individuals experiencing high stress levels (Blagrove, Farmer, & Williams, 2004). Approximately 5% of the general adult population reports regular nightmares (i.e., at least once per week) (Li, Zhang, Li, & Wing, 2010), and those experiencing acute or chronic stress report more severe and frequent nightmares (Wood, Bootzin, Rosenhan, Nolen-Hoeksema, & Jourden, 1992).

Theoretical and empirical evidence suggests nightmares and daily stress may influence each other bidirectionally. According to the stress-acceleration hypothesis of nightmares, stress from trauma or adversity can lead to increased fear sensitivity and less effective fear extinction (Nielsen, 2017). According to the Nightmare-Affect Network Dysfunction model, this maladaptive fear response may be mediated by alterations in the amygdala, hippocampus, anterior cingulate cortex, and medial prefrontal cortex, particularly during rapid eye movement (REM) sleep when nightmares occur (Nielsen, 2017). Physiologically, daily stress activates the sympathetic nervous system and hypothalamic pituitary adrenal axis, which release the stress hormones cortisol, norepinephrine, and epinephrine (Irwin, 2015; Wright et al., 2015). These hormones in turn can lead to hyperarousal that may increase the likelihood of subsequent stress and sleep disturbances, including nightmares (Sladek, Doane, & Breitenstein, 2020). These nightmares may serve as a stressor itself for individuals, resulting in fear of sleep and exacerbating next-day stress and anxiety (Pillai, Steenburg, Ciesla, Roth, & Drake, 2014; Roberts, Lennings, & Heard, 2009; Van Laethem et al., 2015).

Cross-sectional studies have confirmed strong links between stress and nightmares. For example, among pregnant women, more stressors are associated with greater nightmare frequency (Schredl et al., 2019), and among undergraduate students, nightmare frequency is associated with poorer well-being (Zadra & Donderi, 2000) and more distress (Schredl, 2003). Few studies have examined how stressors experienced in everyday life are associated with subsequent nightmares using an intensive longitudinal design. Results from these studies indicate that among those with frequent nightmares or high levels of dissociation, daily nightmares are significantly and positively associated with daily distress (Gehrman, Harb, Cook, Barilla, & Ross, 2015; Lancee & Schrijnemaekers, 2013; Soffer-Dudek & Shahar, 2011). Other longitudinal studies have examined nightmares in relation to posttraumatic stress disorder (PTSD), which is a common psychological condition following exposure to severe traumatic events (Kilpatrick et al., 2013), and entails intrusive trauma-related thoughts, avoidance of trauma triggers, negative alterations in thoughts and affect, and arousal (American Psychological Association, 2013). These studies have shown that among individuals with PTSD, daily negative mood is marginally associated with greater odds of nightmares (Miller, Jamison, Gala, & Woodward, 2018), and greater daytime PTSD symptoms predict subsequent nightmares (Short, Allan, Stentz, Portero, & Schmidt, 2018).

Although these studies provide important preliminary findings on links between daily stress and nightmares, samples were relatively small (N = 31 to 105) or consisted of specific populations (e.g., female undergraduates, male veterans), limiting generalizability of results to other populations such as nurses. Nurses may be particularly likely to experience both nightmares and high stress levels, given their demanding work environment and potential exposure to occupational trauma (e.g., witnessing patient death or violent injury, performing futile care to terminally ill patients, treating severe bleeding and open wounds, being physically threatened by patients) (Mealer, Burnham, Goode, Rothbaum, & Moss, 2009). In fact, studies estimate that up to 35% of nurses experience nightmares related to workplace events and stressors (Mealer et al., 2009). Further, it is plausible nurses with PTSD symptoms exhibit stronger associations between nightmares and daily stress (Short, Allan, & Schmidt, 2017; Short et al., 2018). Approximately 18% of nurses meet diagnostic criteria for PTSD (Mealer et al., 2009). Nightmares are one of the hallmark symptoms of PTSD (Germain, 2013), and ~72% of those with PTSD symptoms report nightmares (Leskin, Woodward, Young, & Sheikh, 2002; Li et al., 2010). These posttraumatic nightmares (i.e., nightmares following trauma) are associated with greater psychological impairment than idiopathic nightmares (nightmares occurring for other reasons) (Schreuder, Kleijn, & Rooijmans, 2000).

Research supports the notion that PTSD may exacerbate daily stress-nightmare associations. For example, on days when an individual experiences a minor hassle, the chronic stress of PTSD may deplete coping resources, rendering them more susceptible to the harmful effects of stress on nightmares or vice versa (Serido, Almeida, & Wethington, 2004). Indeed, chronic stress exacerbates the effects of acute stress on psychological health and well-being (Pike et al., 1997; Serido et al., 2004). Individuals with PTSD also may experience more distressing nightmare content (e.g., related to their traumatic experience), which may intensify next-day stress and arousal. Research shows that the more similar the nightmare content is to the memory of traumatic events, the more distressing those nightmares are perceived (Davis, Byrd, Rhudy, & Wright, 2007). However, others have proposed that idiopathic and PTSD nightmares function similarly (Nielsen, 2017), highlighting a need to test this idea empirically. Overall, understanding if the links between daily stress and nightmares are stronger among nurses experiencing PTSD symptoms may help identify those nurses most in need of interventions.

The Current Study

Expanding on previous literature, we examined: 1) the within-person, bidirectional associations of daily stressor severity with nightmare occurrence and severity in a sample of nurses; and 2) whether associations between daily stressor severity with nightmare occurrence and severity were moderated by baseline PTSD symptoms. We hypothesized that days with greater stress severity would be associated with greater odds of experiencing a nightmare, as well as greater nightmare severity during the subsequent sleep interval. We also hypothesized that more frequent and severe nightmares would be associated with higher levels of next-day stress. Finally, we expected that nurses with higher PTSD symptoms at baseline would have stronger positive bidirectional associations between daily stress and nightmare occurrence and severity, compared to nurses without PTSD symptoms.

Methods

Procedure

This study was part of a larger investigation on the effects of sleep on antibody response to the influenza vaccine that occurred between September 2018 and November 2018. Participants were recruited from two regional hospitals through nursing staff presentations, notification through employee email systems, and flyers that directed them to an initial online consent form. Four-hundred and sixty-one nurses provided online consent and were asked to complete initial online Qualtrics surveys to collect demographic information as well as retrospective self-report estimates of recent health. Participants were then invited to enroll in the main portion of the study in the early fall (i.e., the start of the influenza season), which included completion of in-person informed consent approximately one month later. At this time, participants were given instructions on completing the daily sleep diaries and wearing the actigraphy device, which they completed for the subsequent 14 days. All daily surveys were completed electronically using REDCap and time-stamped. Participants were instructed to complete the daily surveys upon awakening, reporting on their stress and sleep during their main sleep interval in the past 24 hours. All study procedures were approved by the University of North Texas and Medical City Plano Institutional Review Boards.

Participants

Inclusion criteria were: 1) not yet received the current season’s influenza vaccine, 2) between the ages of 18 and 65, and 3) registered nurses actively working at least part-time at one of two regional hospitals. Exclusion criteria were: 1) pregnant/nursing or planning to become pregnant or 2) having an egg allergy (due to potential adverse reactions to the vaccine administered for the larger study). Table 1 reports demographic characteristics for the entire sample. Generally, participants matched national demographics of nurses in the United States (Health Resources and Services Administration, 2013). Most participants were female (92%), White (78%), and non-Hispanic (89%). A significant portion of the sample also self-identified as Asian (10%) or Black/African American (7%). A majority (63%) of the sample was currently married and most had children (65%), and 26% reported working at least one night shift (work between 9pm and 6am) during the 14-day study period (M = 1.34 night shifts, SD = 2.62, median = 0, range = 0 to 14). Approximately 10.5% of the sample (n = 41) was above the cutoff for probable PTSD (i.e., scores ≥ 33 on the PTSD Checklist for DSM-5 [PCL-5]) (Bovin et al., 2016).

Table 1.

Participant Characteristics for the Entire Sample and by Probable Posttraumatic Stress Disorder

	Whole sample	Nurses without probable PTSD	Nurses with probable PTSD	p	d or OR
	Mean (SD) or n (%)	Mean (SD) or n (%)	Mean (SD) or n (%)
n	392	350	41
diary stress	0.81 (0.60)	0.76 (0.56)	1.23 (0.75)	<0.001	0.714
nightmare frequency	0.10 (0.17)	0.09 (0.15)	0.19 (0.25)	<0.001	0.494
nightmare severity	0.10 (0.18)	0.09 (0.15)	0.21 (0.31)	<0.001	0.490
2-week total nightmare frequency	1.28 (2.15)	1.17 (2.08)	2.15 (2.60)	0.006	0.413
PCL-5 total scores	19.07 (10.51)	16.19 (5.73)	43.17 (10.83)	<0.001	3.113
Age (years)	39.54 (11.15)	39.71 (11.24)	38.41 (10.41)	0.484	0.119
Sex (% female)	360 (91.8%)	325 (92.9%)	34 (82.9%)	0.033	0.374
Shift work (% recent night shift worker)	101 (25.8%)	88 (87.12%)	13 (12.9%)	0.365	1.067
Race
White	305 (77.8%)	273 (78.0%)	31 (75.6%)	0.729	0.976
African American/Black	26 (6.6%)	22 (6.3%)	4 (9.8%)	0.400	1.035
American Indian/Alaska Native	6 (1.5%)	4 (1.1%)	2 (4.9%)	0.066	1.038
Asian	41 (10.5%)	37 (10.6%)	4 (9.8%)	0.872	0.992
Multiracial	7 (1.8%)	7 (2.0%)	0 (0.0%)	0.362	0.980
Other	7 (1.8%)	7 (2.0%)	0 (0.0%)	0.362	0.980
Ethnicity (% Hispanic/Latinx)	42 (10.8%)	36 (10.4%)	6 (14.6%)	0.409	1.481
Marital Status
Married	248 (63.3%)	222 (63.4%)	26 (63.4%)	0.999	1.000
Single	101 (25.8%)	89 (25.4%)	11 (26.8%)	0.846	1.014
Divorced	33 (8.4%)	31 (8.9%)	2 (4.9%)	0.387	0.961
Separated	7 (1.8%)	5 (1.4%)	2 (4.9%)	0.116	1.035
Widowed	3 (0.8%)	3 (0.9%)	0 (0.0%)	0.553	0.991

Note. M = mean, SD = standard deviation, PCL-5 = Posttraumatic Stress Disorder Checklist for DSM-5, PTSD = Posttraumatic stress disorder symptoms (Nurses without probable PTSD: scores < 33 on the PCL-5; Nurses with probable PTSD: scores ≥ 33 on the PCL-5), d = Cohen’s d. OR = odds ratio. Nightmare frequency is the average number of nightmares reported per night across the 2 weeks. Nightmare severity is the average nightmare severity (0 = not at all to 3 = very) reported per night across the two weeks.

Measures

Nightmare occurrence and severity.

Upon awakening each day, participants reported on nightmare frequency and severity from the previous sleep interval: “How many nightmares did you have that woke you up?” and “How would you rate the overall severity of your nightmares?” (scale of 0 [not all all] to 3 [very]). The generalizability coefficients for within-person variations averaged over time (time nested within people) were 0.98 and 0.97, respectively, indicating high reliability (i.e., consistency) in individuals’ experiences of nightmares across the 14 days. In other words, the same people were likely to continue have nightmares over time. Because participants reported >1 nightmare on only 15% of all nightmare occasions (n = 67 occasions), for the nightmare occurrence variable, we dichotomized nightmare frequency into 0 (no nightmares that night) and 1 (one or more nightmares that night) for all analyses. Previous research supports collapsing across categories when modeling sparse ordinal variables (DiStefano, Shi, & Morgan, 2020).

Daily stress severity.

In the daily sleep diary, participants reported on their stress severity levels during the previous day using the item “I felt stressed,” rated on a scale of 0 (not at all) to 4 (extremely). Previous studies have used similar single item measures to assess daily stress severity (e.g., Slavish et al., in press).

Baseline PTSD symptoms.

The PTSD Checklist for DSM-5 (PCL-5) is a 20-item self-report measure used to assess PTSD symptom severity in the past month (Weathers et al., 2013). We used the version of the PCL-5 that examines past month PTSD symptoms without assessing Criterion A trauma exposure (Weathers et al., 2013). The measure is summed to obtain a total score ranging from 0 to 80, with higher scores indicating greater symptom severity. A score ≥ 33 indicates a positive screen for PTSD (i.e., moderate to severe symptoms) (Bovin et al., 2016). The PCL-5 has good psychometrics (Blevins, Weathers, Davis, Witte, & Domino, 2015). In the current study, the PCL-5 demonstrated good internal consistency (α = .94).

Statistical Analysis Plan

All analyses were conducted in the open-source statistical program R (R Core Team, 2013). Data and code are available at: https://osf.io/gv53p/?view_only=41b684901e3a48e6a53f9cd6c0e9d040. Multilevel models were conducted using the R package lme4 (Bates, Mächler, Bolker, & Walker, 2015), and tables were created using the R package sjPlot (Lüdecke, 2019). For all multilevel models, level 1 days were nested within level 2 people. Restricted maximum likelihood (REML) was used, which is a robust method for handling missing data using all available information to estimate the model (Raudenbush & Bryk, 2002). Intercepts were allowed to vary randomly across people. All models controlled for sex (0 = male, 1 = female) and age, given previous studies showing robust differences in stress, sleep, and nightmares by these variables (Dijk, Beersma, & Bloem, 1989; Reyner, Horne, & Reyner, 1995; Schredl & Reinhard, 2011; Walsleben et al., 2004). For analyses examining nightmares predicting subsequent stress, stress data were lagged back one day (as stress was reported in the morning, reflecting stress experienced the previous day; see Figure 1 for a schematic). For analyses examining stress predicting subsequent nightmares, data were not lagged (as previous day’s stress and last night’s nightmares were reported simultaneously; Figure 1). For nightmare occurrence analyses, all observations were included; for nightmare severity analyses, only observations when individuals reported a nightmare were included in analyses. An example equation for between-person (level 2) baseline PTSD symptoms moderating the within-person (level 1) slope between daily stress on nightmare severity (i.e., Aim 2) is shown below:

$$\text{Level}1\left(\text{days}\right):{\text{Nightmare severity}}_{\text{ij}}={\beta }_{0\text{j}}+{\beta }_{\text{1j}}{\text{Daily stress}}_{\text{ij}}+r_{\text{ij}}$$

$$\begin{gathered} \text{Level}2\left(\text{people}\right):{\beta }_{0\text{j}}={\gamma }_{00}+{\gamma }_{01}{\text{Baseline PTSD}}_{\text{j}}+{\mu }_{0\text{j}} \\ {\beta }_{1\text{j}}={\gamma }_{10}+{\gamma }_{11}{\text{Baseline PTSD}}_{\text{j}} \end{gathered}$$

where: γ₀₀ is the overall mean of nightmare severity, γ₁₀ is the overall slope between stress and nightmare severity, γ₀₁ is the overall effect of between-person PTSD symptoms on the overall mean of nightmare severity, γ₁₁ is the cross-level moderation effect of between-person baseline PTSD symptoms on the within-person slope between stress and nightmare severity, and μ_0j is the random deviations of the j^th person’s mean nightmare severity from the overall mean.

Figure 1.

Schematic of Analytic Plan and Timing of Assessments

Results

Descriptive results

Table 1 displays descriptive results and participant characteristics. On average, nurses completed 13.29 (SD = 1.28) of 14 possible daily surveys for an average compliance rate of 95%. On average, nurses reported daily stress scores of 0.81 (SD = 0.93), out of possible scores of 0-4, and an average of 1.28 (SD = 2.15) total nightmares across the two weeks, with 47.2% of nurses reporting at least one nightmare across the two weeks. Approximately 26.5% of nurses reported =/>2 nightmares across the two weeks, and 12.0% reported =/>4 nightmares. At least one nightmare was reported on 8% (n = 437) of all daily observations (1 nightmare = 85% of nightmare observations; 2 nightmares = 12% of nightmare observations; 3+ nightmares = 4% of nightmare observations). Those with probable PTSD (10.5%) reported higher average daily stress and more frequent and severe nightmares (Table 1). For repeated measures variables, intraclass correlation coefficients (i.e., ratio of between-person variation to total variation) were 36% for daily stress, 12% for daily nightmare frequency, and 11% for nightmare severity, suggesting more variation at the within-person level (i.e., from day-to-day) than at the between-person level (i.e., from person-to-person) for each construct.

Aim1: Main effects of within-person, bidirectional associations between stress and nightmares

Days with greater stress severity were associated with higher odds of experiencing a nightmare that night (OR = 1.22, 95% CI [1.08 - 1.37], p = 0.001), as well as more severe nightmares among those reporting a nightmare (b = 0.09, 95% CI [0.01 - 0.16], p = 0.033; Table 2). Nightmare occurrence (b = 0.15, 95% CI [0.07 - 0.24], p < 0.001), but not nightmare severity (b = 0.01, 95% CI [−0.12 - 0.14], p = 0.856) was associated with greater next-day stress severity (Table 3).

Table 2.

Effects of daily stress severity on nightmare occurrence and severity

	Nightmare occurrence			Nightmare severity
Predictors	Odds Ratios	CI	p	Estimates	CI	p
(Intercept)	0.04	0.02 – 0.11	<0.001	1.61	1.16 – 2.05	<0.001
Daily stress severity	1.22	1.08 – 1.37	0.001	0.09	0.01 – 0.16	0.033
Age	1.00	0.98 – 1.01	0.676	−0.01	−0.02 – −0.00	0.006
Sex	0.99	0.51 – 1.92	0.981	0.08	−0.27 – 0.42	0.666
Random Effects
σ2	3.29			0.54
τ00	1.64			0.13
ICC	0.33			0.20
N	389 ID			189 ID
Observations	5434			431
Marginal R2 / Conditional R2	0.007 / 0.337			0.033 / 0.223

Note. Bold values represent p <.05 estimates. CI = 95% confidence interval. Sex was coded as 0 = male, 1 = female. Nightmare occurrence is a binary variable where 0 = no nightmare experienced, 1 = nightmare experienced. σ² represents level 1 variance (within-person). τ₀₀ represents level 2 variance (between-person). ICC = intraclass correlation coefficient. Nightmare severity analyses reflect only those observations when participants reported at least one nightmare. Odds ratios are used for dichotomous outcomes, and estimates (beta weights) are used for continuous outcomes.

Table 3.

Effects of nightmare occurrence and severity on next-day stress severity

	Next-day stress severity			Next-day stress severity
Predictors	Estimates	CI	p	Estimates	CI	p
(Intercept)	0.54	0.26 – 0.83	<0.001	0.71	0.04 – 1.37	0.038
Age	0.00	−0.00 – 0.01	0.392	0.00	−0.01 – 0.01	0.637
Sex	0.14	−0.08 – 0.36	0.199	0.23	−0.28 – 0.73	0.376
Nightmare occurrence	0.15	0.07 – 0.24	<0.001
Nightmare severity				0.01	−0.12 – 0.14	0.856
Random Effects
σ2	0.53			0.78
τ00	0.31			0.35
ICC	0.37			0.31
N	389 ID			181 ID
Observations	4848			385
Marginal R2 / Conditional R2	0.004 / 0.370			0.004 / 0.312

Note. Bold values represent p <.05 estimates. CI = 95% confidence interval. Sex was coded as 0 = male, 1 = female. Nightmare occurrence is a binary variable where 0 = no nightmare experienced, 1 = nightmare experienced. σ² represents level 1 variance (within-person). τ₀₀ represents level 2 variance (between-person). ICC = intraclass correlation coefficient. Nightmare severity analyses reflect only those observations when participants reported at least one nightmare. Odds ratios are used for dichotomous outcomes, and estimates (beta weights) are used for continuous outcomes.

Aim 2: Cross-level moderation by baseline PTSD symptoms

At the between-person level, baseline PTSD symptoms did not moderate within-person associations between daily stress and nightmare occurrence or severity, or vice versa (Supplementary Tables 1-2). Results were similar when the nightmare item was removed from the PCL-5 total score; therefore, only results with all items are displayed.

Discussion

This was the first study to examine daily bidirectional associations between stress and nightmares among nurses. Our results showed daily stress and nightmares were related in a bidirectional fashion. Specifically, days with greater stress were associated with higher odds of experiencing a nightmare and more severe nightmares that night, and nights with at least one nightmare reported were associated with greater next-day stress severity. These daily associations did not vary by an individual’s baseline levels of PTSD symptoms. Together, these results characterize the daily cycle of stress and nightmares in a sample vulnerable to high levels of occupational stress.

Our results map onto existing studies, which indicate stress and nightmares are robustly associated cross-sectionally (Schredl, 2003; Schredl et al., 2019; Zadra & Donderi, 2000), as well as bidirectionally in daily life (Lancee & Schrijnemaekers, 2013; Soffer-Dudek & Shahar, 2011). In contrast to previous studies and adding to existing scientific literature, we examined these associations over a longer period of time (14 days) in a larger sample, resulting in 5488 potential measurement occasions, and in a vulnerable sample of nurses. Nurses are a critical sample in which to investigate these links, as stress and nightmares may affect not only their own health and well-being, but also their ability to provide effective patient care and maintain patient safety.

It is possible the associations we observed are mediated by negative affect, cognitive arousal, and sympathetic nervous system activity. In this regard, studies have shown that days with higher stress are associated with increases in negative emotions and rumination, as well as dysregulated cortisol patterns the next day (Sladek et al., 2020). Higher stress is also associated with less nocturnal blood pressure dipping (Holt-Lunstad, Jones, & Birmingham, 2009; Smart Richman, Pek, Pascoe, & Bauer, 2010) and higher nocturnal heart rate (Paul, Alpers, Reinhard, & Schredl, 2019), which may indicate elevated nighttime arousal and increase risk for nightmares. Future studies should examine possible psychophysiological mediators of links between daily stress and nightmares.

Surprisingly, we did not find that baseline PTSD symptoms moderated associations between daily stress and nightmares. This suggests that, regardless of the severity of an individuals’ PTSD symptoms, the links between daily stress and nightmares were similar. Some studies have shown that daily processes such as negative emotions and rumination are more important for sleep than more trait-like characteristics (e.g., neuroticism) (Samson-Daoust, Julien, Beaulieu-Prévost, & Zadra, 2019; Schredl, 2003; Slavish et al., 2018). However, other research shows factors such as trait anxiety are important predictors of the emotional content of dreams (Samson-Daoust et al., 2019). The relative importance of trait- vs. state-like processes also may vary as a function of sample characteristics (college students vs. veterans vs. nurses) and participants’ level of chronic stress (Samson-Daoust et al., 2019). Together, this work highlights the importance of examining both trait- and state-like psychological processes that may influence nightmares and stress.

Limitations and Future Directions

Although this study has several notable strengths (e.g., large sample; 14 days of repeated measures; bidirectional, within-person analyses), there are some limitations warranting future research. We measured stress and nightmares concurrently and only once daily. Ideally, multiple assessments of stress exposure and reactivity would be taken daily to capture dynamics across a day. We also did not collect information on nurses’ specific work position or setting (e.g., critical care; emergency room; surgical; geriatric; family practitioner). These factors will be important to assess in future studies on nurses’ stress and sleep, as occupational conditions (e.g., role conflict, job security, task variety, autonomy, schedule flexibility, time pressure) are strong predictors of burnout among nurses (Dall’Ora, Ball, Reinius, & Griffiths, 2020). Night shift work also may be an important consideration for future research, as nurses who work nights report poorer sleep, more fatigue, and lower job satisfaction than day shift nurses (Ferri et al., 2016). Further, how perceived stress and nightmares interact with, or are mediated by, other processes (e.g., rumination, mood, substance use, blood pressure, cortisol, sleeping medication use, obstructive sleep apnea symptoms) will be fruitful for informing intervention targets. Finally, we were limited by our self-report measure of PTSD and the small number of nurses endorsing probable PTSD. Studies should examine these associations in a larger sample, including among those endorsing PTSD determined through a diagnostic interview (e.g., Clinician-Administered PTSD Scale for DSM-5).

Conclusion

Nightmares are increasingly recognized as an important transdiagnostic correlate of health and well-being (Blagrove et al., 2004; Zadra & Donderi, 2000). Our results suggest stress and nightmares may unfold in a bidirectional fashion in daily life among nurses. Targeting stress may be one means to promote more restorative sleep among nurses. Sufficient and high-quality sleep characterized by fewer nightmares may also help lessen nurses’ perceptions of next-day stress. Given the critical role that nurses play in the healthcare system, it is essential to recognize the unique challenges they face that may affect their sleep, well-being, and ability to provide effective patient care.

Data accessibility statement:

All R code used for analyses is available at: https://osf.io/gv53p/?view_only=41b684901e3a48e6a53f9cd6c0e9d040. Data will be made available upon request.