Influence of poor emotion regulation on disrupted sleep and subsequent psychiatric symptoms in university students

Abstract

Previous studies in the neuroscience and psychology literature works suggest that poor sleep quality is associated with emotion dysregulation, and that poor sleep quality and emotion dysregulation are, independently, associated with the presence and severity of psychiatric symptoms. However, no previous study has examined simultaneous relations among multiple different emotion regulation strategies, sleep quality, and mental health outcomes. Such investigations are particularly important given the extensive literature describing the prevalence and manifestation of poor mental health outcomes in university students. This study investigated the influence of both maladaptive (avoidance and impulsivity) and adaptive (cognitive reappraisal) emotion regulation strategies on sleep quality and, subsequently, on the degree of depressive and posttraumatic symptomatology in a sample of South African university students (N = 336). Participants completed self-report instruments measuring their tendency to use avoidance, cognitive reappraisal, and impulsivity; their sleep quality; their accessibility to social support; and their number of depressive and posttraumatic symptoms. Structural equation modelling showed that more use of avoidance and impulsivity and less use of cognitive reappraisal negatively affected sleep quality, which, in turn, was associated with the presence of more depressive and more posttraumatic symptoms. Hence, our findings suggest that emotion regulation indirectly exerted its influence on the manifestation of psychiatric symptoms through sleep. We conclude that interventions targeted at improving sleep quality may prove beneficial in lessening the burden of depressive and posttraumatic symptoms in university students.

Keywords

Depression emotion regulation sleep structural equation model: trauma students

Although a rich literature describes the prevalence and manifestation of poor mental health outcomes in university students (see, for example, Bantjes et al., 2019), relatively few studies investigate specific information processing strategies and neurobiological factors affecting the presence and severity of psychiatric symptoms in these populations. In this study, we examined ways in which emotion dysregulation and sleep disruption (independently and in relation to each other) affect depressive and posttraumatic symptoms in a sample of South African undergraduates.

Emotion regulation refers to ways in which people manage and modify their experience of an emotional response to an affect-eliciting stimulus (e.g., regulate their anger in response to verbal confrontations; McRae & Gross, 2020; Thompson, 2019). Maladaptive emotion regulation strategies include avoidance (i.e., unwillingness to stay in contact with aversive internal experiences and to take action to alter them or the contexts created by them) and impulsivity (i.e., inability to control an immediate response to a stimulus). Adaptive strategies include cognitive reappraisal (i.e., cognitively reframing the perspective one takes towards a situation that typically generates negative emotion).

Emotion dysregulation impacts sleep quality negatively (for reviews, see Palmer & Alfano, 2017; Vandekerckhove & Wang, 2018). Studies that included large numbers of university students have shown that difficulties in regulating negative emotion reduce sleep efficiency, increase the number of night-time awakenings, and increase symptoms of insomnia (see, for example, Semplonius & Willoughby, 2018; Vandekerckhove et al., 2011). More specifically, the use of particular maladaptive emotion regulation strategies might precipitate distinctive patterns of sleep disruption. For instance, Hoyt et al. (2009) reported that frequent use of an avoidance-based strategy was associated with difficulties in falling asleep and with increased instability in morning awakening time. They speculated that their participants may have been more susceptible to ruminate on unhelpful sleep-related thoughts elicited by negative emotional experiences that had not been regulated adaptively.

Many psychiatric conditions are characterised by both emotion dysregulation and sleep disruption. For instance, a feature of the clinical presentation associated with depression is failure to use cognitive reappraisal strategies effectively. That is, depressed individuals struggle to attenuate negative emotional states and tend to experience emotion-eliciting stimuli more negatively than do healthy controls (Ong & Thompson, 2019). Existing evidence shows that this pattern is evident in studies recruiting university students, that is, those who tend not to use cognitive reappraisal have more depressive symptoms (see, for example, Gong et al., 2020). Separately, sleep in depressed individuals relative to healthy controls is characterised by longer sleep latency, shorter rapid eye movement (REM) sleep latency, longer periods of REM sleep, and more awakenings after sleep onset (Hein et al., 2017).

Similarly, a feature of the clinical presentation in individuals with posttraumatic symptoms is the maladaptive use of avoidance-based emotional regulation strategies. Specifically, these individuals often develop a conditioned fear response related to their trauma experience and consequently use experiential avoidance to avoid any potential fear-provoking stimuli (Seligowski et al., 2015). Persistent use of such avoidance strategies intensifies and prolongs the durability of those symptoms (Basharpoor et al., 2015). In students, one study found that students with probable posttraumatic stress disorder (PTSD) had greater emotional dysregulation, including greater avoidance than students without probable PTSD (Hannan & Orcutt, 2020). Separately, individuals with posttraumatic symptoms show, relative to healthy controls, increased sleep latency, reduced sleep efficiency, reduced sleep depth (less slow-wave sleep and more Stage 1 non-REM sleep), and REM abnormalities (for a review, see Zhang et al., 2019).

As noted earlier, emotion dysregulation (a) impacts sleep quality negatively in healthy adults, and (b) is associated with particular behavioural characteristics of individuals with depressive and posttraumatic symptoms. However, to our knowledge, only two previous studies have investigated whether associations between emotion dysregulation and clinical symptoms are mediated by sleep disruption.¹ Liu et al. (2020; N = 755 university students) found that less frequent use of cognitive reappraisal strategies was indirectly (via poor sleep quality) related to presence of depressive symptoms. Warnke et al. (2017; N = 300 community-dwelling volunteers) found that the association between experiential avoidance and depressive symptoms was mediated by sleep quality.

Another important factor affecting mental health outcomes in university students is social support. Undergraduates with higher levels of such support tend to perform better academically, to experience reduced levels of academic stress, and to adjust easier to social and emotional situations (Awang et al., 2014; J. Li et al., 2018). More generally, among young adults a high degree of social support is associated with healthier sleep and fewer psychiatric symptoms (Doré et al., 2017; Kim & Suh, 2017). Furthermore, emotion regulation processes appear to be sensitive to the influence of social support; hence, effects of social support on depressive symptomatology might be accounted for, at least partially, by their effects on emotion regulation (Marroquín, 2011).

This study attempted to add to literature investigating whether associations between emotion dysregulation and clinical symptoms are mediated by sleep disruption and to fill some of the remaining knowledge gaps in this field. Based on the literature reviewed above, we hypothesise that (after controlling for the effects of varying levels of social support) emotion regulation strategies influence sleep quality, which, in turn, influences degree of (a) depressive symptomatology and (b) posttraumatic symptomatology. We used survey-based data collection methods and structural equation modelling (SEM) to investigate relations between emotion regulation strategies, sleep disruption, and mental health outcomes (specifically, the presence of depressive and/or posttraumatic symptoms) in South African university students.

Method

Participants

Using convenience sampling, we recruited 336 undergraduate volunteers (93 men, 242 women; 1 participant self-described as queer) between June and August 2017. Individuals were eligible for participation if, at the time of recruitment, they were registered students at the University of Cape Town (UCT). The sample’s mean age was 21.13 ± 3.5 years (range = 18–45 years).

Instruments

The 15-item Brief Experiential Avoidance Questionnaire (BEAQ; Gámez et al., 2014) measures the degree to which individuals use avoidance strategies (behavioural avoidance, distress aversion, procrastination, distraction, repression, distress endurance) to regulate emotion. Respondents rate each item on a six-point Likert-type scale (1 = strongly disagree to 6 = strongly agree) so that higher item scores indicate greater use of avoidance strategies. The instrument has demonstrated excellent test–retest reliability (r = .80), internal consistency (Cronbach’s α = .86), and construct validity (Gámez et al., 2014; Valencia, 2018).

The six-item Impulsive Control Difficulties subscale of the 36-item Difficulties in Emotion Regulation Scale (DERS; Gratz & Roemer, 2004) measures the tendency to respond impulsively when distressed. Respondents rate, on a five-point Likert-type scale (1 = almost never to 5 = almost always), the extent to which the statement in each item applies to them. Hence, higher item scores indicate a greater degree of impulsivity. The instrument has demonstrated excellent test–retest reliability (r = .80), internal consistency (Cronbach’s α = .89), and construct validity (Weiss et al., 2015).

The 10-item Emotion Regulation Questionnaire (ERQ; Gross & John, 2003) measures the tendency to use cognitive reappraisal to regulate emotions. Respondents answer each item using a seven-point Likert-type scale (1 = strongly disagree to 7 = strongly agree) so that higher item scores indicate greater use of cognitive reappraisal strategies. The instrument has demonstrated excellent test–retest reliability (r = .82), internal consistency (Cronbach’s α = .85), and construct validity (W. Li et al., 2007).

The 19-item Pittsburgh Sleep Quality Index (PSQI; Buysse et al., 1989) measures self-reported sleep quality over the month prior to responding. Higher scores on items enquiring about sleep latency, sleep duration, sleep disturbances, use of sleep medication, daytime dysfunction, subjective sleep quality, and habitual sleep efficiency indicate poorer sleep quality. The instrument has demonstrated good test–retest reliability (r = .70), internal consistency (Cronbach’s α = .83), and construct validity (Mollayeva et al., 2016). It has been used extensively in research on South African samples (see, for example, Lipinska & Thomas, 2017, 2019; van Wyk et al., 2016).

The 19-item Medical Outcomes Study Social Support Survey (MOSSSS; Sherbourne & Stewart, 1991) comprises five subscales: emotional/informational support, tangible support, affectionate support, positive social interaction, and positive physical interaction. Respondents rate, on a five-point Likert-type scale (1 = none of the time to 5 = all of the time), how often each type of social support is available to them. Hence, higher item scores indicate more frequent availability of social support. The instrument has demonstrated excellent test–retest reliability (r = .88), excellent internal consistency (Cronbach’s α = .92), and stable construct validity (King et al., 2017). It has been used successfully in research on South Africa samples (Ncama et al., 2008).

The nine-item Patient Health Questionnaire for Depression–9 (PHQ-9; Kroenke et al., 2001) assesses the presence and severity of depressive symptoms over the most recent 2-week period (Blackwell & McDermott, 2014). Respondents rate each item on a 0–3 scale, with higher scores representing greater depression severity. The developers report that the instrument has excellent test–retest reliability (r = .84), internal consistency (Cronbach’s α = .89), and construct validity. It has been used extensively in research on South African samples (see, for example, Cholera et al., 2017; Rane et al., 2018).

The Primary Care PTSD Screen (PC-PTSD; Prins et al., 2003) assesses the presence of posttraumatic symptoms. It comprises four yes/no items, each referring to a core Diagnostic and Statistical Manual of Mental Disorders (4th ed., text rev.; DSM-IV-TR) PTSD diagnostic criterion: re-experiencing, avoidance, hyperarousal, and numbing. The instrument has demonstrated excellent test–retest reliability (r = .83), internal consistency (Cronbach’s α = .85), and construct validity (Ouimette et al., 2008; Prins et al., 2003). It has been used successfully in research on South African samples (see, for example, Peltzer & Louw, 2013).

Procedure

We used two platforms to recruit participants. First, we sent an invitation email via the UCT Department of Psychology’s Student Research Participation Programme (SRPP) intranet site. Second, we sent a similar email to a UCT-wide email listserv. Students who responded positively to the invitation were directed to an online survey, created using Google Forms. The survey presented the PHQ-9, PC-PTSD, PSQI, BEAQ, DERS, ERQ, and MOSSSS, in that order.

Ethical considerations

All study procedures were approved by our institution’s relevant Research Ethics Committee. All participants gave written informed consent (on the first page of the online survey), and were debriefed comprehensively (the final page of the online survey was a debriefing form that included contact details for local counselling services). Students were compensated with ZAR30.00 for their participation in the study.

Data analysis

We used R Studio (Verzani, 2018) to conduct all analyses. Initial procedures included screening the data for normality, multicollinearity, and missing values. Normality analyses indicated that item skewness and kurtosis were a cause for concern. Hence, when fitting models with continuous data (i.e., those involving depressive symptoms), we used maximum likelihood estimation with robust standard errors, whereas when fitting models with ordinal data (i.e., those involving posttraumatic symptoms), we used diagonally weighted least square estimation (Lai, 2018; Mîndrilă, 2010). Multicollinearity analyses indicated that all included items were correlated at <.85, indicating no substantial concerns in this regard (Kline, 2015). Some items had missing values of <5%, which were handled using the full information maximum likelihood method when fitting models with continuous data and pairwise deletion when fitting ordinal data.

SEM (the Lavaan package in R; Rosseel, 2012) assessed the hypothesised relationship between emotion regulation, sleep, and psychiatric symptoms (depressive and posttraumatic symptoms independently). To generate models, we followed a two-step procedure. First, a confirmatory factor analysis assessed the measurement model and optimised the indicators of each latent variable. Each model’s goodness-of-fit was examined using the comparative fit index (CFI) and the root mean square error of approximation (RMSEA). A good fit is indicated by CFI > .95 and RMSEA < .05 (Browne & Cudeck, 1992; Hu & Bentler, 1999). If the implied measurement model did not have a good fit, we dropped items whose squared multiple correlations were <.3 (Burnett & Dart, 1997). Furthermore, modification indices regarding allowing items to co-vary were examined and those that fitted theoretically were included (McDonald & Ho, 2002). A final measurement model was reached when the goodness-of-fit indices were within the acceptable range.

Second, for each of the two psychiatric outcomes, we tested a structural model in which emotion regulatory strategies and social support were conceptualised as earliest in a chain of relationships, having (a) direct effects on sleep and psychiatric symptoms and (b) indirect effects on psychiatric symptoms via sleep. In these models, sleep was then conceptualised as having a direct effect on psychiatric symptoms. Structural effects in the model were observed by specifying the covariance between the exogenous variables when fitting the model.

Results

Sample mean scores on the PHQ-9, PSQI, ERQ, DERS, and BEAQ generally fell close to the scale midpoints (Table 1). For the MOSSSS, the sample mean was considerably higher than the scale’s midpoint (2.50), indicating that these students had relatively good social support. Bivariate correlational analyses indicated that (a) avoidance and impulsivity were significantly positively associated with poor sleep and more depressive symptoms, (b) cognitive reappraisal was significantly negatively associated with poor sleep and more depressive symptoms, and (c) poor sleep was significantly positively associated with more depressive symptoms.

Table 1.

Continuous study variables: descriptive statistics and bivariate correlations (N = 336).

	α	M	SD	1	2	3	4	5	6
PHQ-9^a	.87	1.19	0.68	–
PSQI^b	.72	1.00	0.53	.64***	–
ERQ	.85	4.60	1.21	–.27***	–.28***	–
DERS	.90	2.37	1.00	.49***	.45***	–.31***	–
BEAQ	.86	3.50	0.86	.46***	.34***	–.18***	.47***	–
MOSSSS^a	.95	3.98	0.81	–.31***	–.30***	.22***	–.30***	–.33***	–

PHQ-9: Patient Health Questionnaire for Depression–9 (measured depressive symptoms); PSQI: Pittsburgh Sleep Quality Index (sleep quality); ERQ: Emotion Regulation Questionnaire (cognitive reappraisal); DERS: Difficulties in Emotion Regulation Scale (impulsivity); BEAQ: Brief Experiential Avoidance Questionnaire (avoidance); MOSSSS: Medical Outcomes Study Social Support Survey (social support).

The first column presents data for the internal consistency reliability (Cronbach’s α) of each instrument. M and SD values are unit-weighted composites. Correlations computed using Pearson method with pairwise deletion.

N = 335.

N = 321.

***

p < .001.

The PC-PTSD data indicated that more than one-third of participants reported experiencing at least one posttraumatic symptom (Table 2).

Table 2.

Primary care PTSD screen (PC-PTSD): descriptive statistics (N = 336).

Item	f (%)
In your life, have you ever had any experience that was so frightening, horrible, or upsetting that, in the past month, you:	f (%)
Have had nightmares about it or thought about it when you did not want to?	129 (39)^a
Tried hard not to think about it or went out of your way to avoid situations that reminded you of it?	152 (45)
Were constantly on guard, watchful, or easily startled?	109 (33)^a
Felt numb or detached from others, activities, or your surroundings?	138 (41)

PTSD: posttraumatic stress disorder.

Data are the number (percentage) of participants who answered ‘yes’ to this question.

N = 335.

Depressive symptoms, emotion regulation, and sleep

The hypothesised measurement model consisted of six latent variables: avoidance (six subscales), impulsivity (six items), cognitive reappraisal (six items), social support (five subscales), sleep (seven subscales), and depressive symptoms (nine items). The goodness-of-fit indices indicated that this measurement model did not adequately fit the data, CFI = 0.873, RMSEA = 0.058 (90% CI = [0.053, 0.062]). Using the methods outlined above, we systematically improved the degree of fit for the measurement model until it was high enough to warrant continuing with the analysis. Ultimately, the goodness-of-fit indices were CFI = 0.954, RMSEA = 0.041 (90% CI = [0.035, 0.047]).

The structural model showed there was a direct relationship between sleep and depressive symptoms, with the coefficient’s direction indicating that poorer sleep was correlated with more depressive symptoms (Table 3; Figure 1). There was also a direct pathway from avoidance to depressive symptoms, with the coefficient’s direction indicating that greater use of this strategy was directly associated with more depressive symptoms.

Table 3.

Structural equation model predicting depressive symptoms (N = 336).

Effects	β	p	95% CI
Effects	β	p	LL	UL
Direct
Social support → Sleep	–0.079	.288	–0.225	0.067
Cognitive reappraisal → Sleep	–0.143	.045*	–0.282	–0.003
Impulsivity → Sleep	0.322	<.001***	0.179	0.465
Avoidance → Sleep	0.204	.011*	0.046	0.362
Sleep → Depressive symptoms	0.823	<.001***	0.712	0.933
Social support → Depressive symptoms	–0.033	.474	–0.125	0.058
Cognitive reappraisal → Depressive symptoms	–0.023	.640	–0.121	0.075
Impulsivity → Depressive symptoms	0.024	.682	–0.092	0.141
Avoidance → Depressive symptoms	0.121	.022*	0.017	0.225
Indirect^a
Social support → Sleep → Depressive symptoms	–0.065	.290	–0.186	0.056
Cognitive reappraisal → Sleep → Depressive symptoms	–0.117	.051	–0.235	0.001
Impulsivity → Sleep → Depressive symptoms	0.265	<.001***	0.136	0.394
Avoidance → Sleep → Depressive symptoms	0.168	.014*	0.035	0.302
Total^b
Social support → Depressive symptoms	–0.099	.146	–0.231	0.034
Cognitive reappraisal → Depressive symptoms	–0.141	.027*	–0.265	–0.016
Impulsivity → Depressive symptoms	0.289	<.001***	0.164	0.414
Avoidance → Depressive symptoms	0.289	<.001***	0.157	0.422
Total effect^c	1.162	<.001***	0.877	1.447

β: standardised path coefficient; CI: confidence interval; LL: lower limit; UL: upper limit.

Indirect effect of social support and emotion regulation strategies on depressive symptoms through sleep.

Total effects of social support and emotion regulation strategies, including direct and indirect effects, on depressive symptoms.

Combined effect of all direct and indirect effects on depressive symptoms.

p < .05. ***p < .001.

Figure 1.

Structural equation model predicting depressive symptoms: initial model.

There were direct pathways from each of avoidance, impulsivity, and cognitive reappraisal to sleep, and an indirect pathways from each of those emotion regulation strategies to depressive symptoms via sleep. Directions of coefficients indicated that greater use of avoidance, greater use of impulsivity, and less use of cognitive reappraisal were independently associated with poorer sleep and, therefore, with more depressive symptoms. No other statistically significant pathways were observed. Avoidance and impulsivity had large and near-identical total effects on depressive symptoms, while cognitive reappraisal had a smaller (but still significant) effect.

Overall, SEM indicated that emotion regulation strategies had a statistically significant impact on depressive symptoms partially through sleep. The model had a good global fit, with CFI > .95 and RMSEA < .05.

Posttraumatic symptoms, emotion regulation, and sleep

The structural model showed there was a direct relationship between sleep and posttraumatic symptoms, with the coefficient’s direction indicating that poorer sleep was correlated with more posttraumatic symptoms (Table 4; Figure 2).

Table 4.

Structural equation model predicting posttraumatic symptoms (N = 336).

Effects	β	p	95% CI
Effects	β	p	LL	UL
Direct
Social support → Sleep	–0.041	.196	–0.104	0.021
Cognitive reappraisal → Sleep	–0.189	<.001***	–0.248	–0.129
Impulsivity → Sleep	0.371	<.001***	0.274	0.468
Avoidance → Sleep	0.227	<.001***	0.128	0.326
Sleep → Posttraumatic symptoms	0.542	<.001***	0.414	0.670
Social support → Posttraumatic symptoms	–0.012	.740	–0.083	0.059
Cognitive reappraisal → Posttraumatic symptoms	–0.009	.805	–0.082	0.064
Impulsivity → Posttraumatic symptoms	0.080	.179	–0.037	0.197
Avoidance → Posttraumatic symptoms	0.080	.162	–0.032	0.192
Indirect^a
Social support → Sleep → Posttraumatic symptoms	–0.022	.202	–0.057	0.012
Cognitive reappraisal → Sleep → Posttraumatic symptoms	–0.102	<.001***	–0.145	–0.060
Impulsivity → Sleep → Posttraumatic symptoms	0.201	<.001***	0.126	0.276
Avoidance → Sleep → Posttraumatic symptoms	0.123	<.001***	0.061	0.185
Total^b
Social support → Posttraumatic symptoms	0.022	.202	–0.057	0.012
Cognitive reappraisal → Posttraumatic symptoms	–0.112	<.001***	–0.173	–0.050
Impulsivity → Posttraumatic symptoms	0.282	<.001***	0.188	0.375
Avoidance → Posttraumatic symptoms	0.203	<.001***	0.105	0.301
Total effect^c	0.881	<.001***	0.703	1.059

β: standardised path coefficient; CI: confidence interval; LL: lower limit; UL: upper limit.

Indirect effect of social support and emotion regulation strategies on posttraumatic symptoms through sleep.

Total effects of social support and emotion regulation strategies, including direct and indirect effects, on posttraumatic symptoms.

Combined effect of all direct and indirect effects on posttraumatic symptoms.

***

p < .001.

Figure 2.

Structural equation model predicting posttraumatic symptoms: initial model.

There were direct pathways from each of avoidance, impulsivity, and cognitive reappraisal to sleep, and is an indirect pathway from each of those emotion regulation strategies to posttraumatic symptoms via sleep. Directions of coefficients indicated that greater use of avoidance, greater use of impulsivity, and less use of cognitive reappraisal were independently associated with poorer sleep and, therefore, with more posttraumatic symptoms. No other statistically significant pathways were found. Impulsivity had the greatest total effect on posttraumatic symptoms, and cognitive reappraisal the smallest.

Overall, the SEM indicated that emotion regulatory strategies had a statistically significant impact on posttraumatic symptoms through sleep. The model had a good global fit, with CFI > .95 and RMSEA < .05.

Discussion

University students are at high risk for psychiatric difficulties (see, for example, Bantjes et al., 2019). However, to our knowledge, no published study has investigated ways in which specific information processing strategies (e.g., emotion regulation) and neurobiological factors (e.g., sleep quality) affect the presence and severity of psychiatric symptoms in students. Here, we used data from 336 undergraduates to create structural equation models (SEMs) that tested two hypotheses: emotion regulation strategies will influence sleep quality, which, in turn, will affect number of (a) depressive symptoms and (b) posttraumatic symptoms.

The SEM testing our first hypothesis confirmed the predicted relationships. More use of maladaptive emotion regulation strategies (avoidance and impulsivity) and less use of an adaptive strategy (cognitive reappraisal) negatively affected sleep quality, which, in turn, contributed to a higher number of depressive symptoms. Although a substantial body of literature already describes the role of sleep disruption in the development of depressive and posttraumatic symptoms (see, for example, the comprehensive meta-analyses by Baglioni et al., 2016 and Wang et al., 2019), the mediational role of sleep disruption in the relationship between other well-established characteristics of major depressive disorder and PTSD (such as emotional dysregulation) and the presentation of clinical symptoms is still in its infancy.

Regarding the literature described, our findings are consistent with, and extend, those reported by Liu et al. (2020) and Warnke et al. (2017). Like them, we found that emotion regulation was indirectly (via poor sleep quality) related to the presence of more depressive symptoms. However, where they each limited their investigation to a single emotion regulation strategy and to depressive symptoms only, we examined multiple emotion regulation strategies simultaneously and posttraumatic as well as depressive symptoms. Furthermore, where they used regression-based mediation analyses, we used SEM. This choice allowed our constructs to be represented as latent variables uncontaminated by measurement error (Nachtigall et al., 2003). SEM also allows for the specification of complex models that more closely match theory and provide relatively accurate evaluation of directional mechanisms (Bag, 2015).

The SEM testing our second hypothesis also confirmed the predicted relationships. More use of avoidance and impulsivity and less use of cognitive reappraisal negatively affected sleep quality, which, in turn, contributed to a higher number of depressive symptoms.

The models predicting depressive and posttraumatic symptoms showed the same direction and approximate magnitude of association between emotion regulation strategies and sleep quality. Impulsivity was the strongest predictor, with the analyses suggesting that increased use of that strategy was associated with poor sleep quality. Impulsive individuals may experience poor sleep quality because they are more likely to engage in behaviours detrimental to healthy sleep (e.g., unrestrained screen use, increased alcohol consumption before bed; Schmidt et al., 2010).

Our finding regarding impulsivity is consistent with Short et al. (2014), who showed that, out of several emotion regulation strategies, only in individuals with high rather than low impulsivity there was an association between poor sleep quality and posttraumatic symptoms. To our knowledge, no previous study reports a similar finding in individuals with depressive symptoms. Hence, in this regard, our study makes a novel contribution.

Cognitive reappraisal and avoidance also had a significant impact on sleep quality and, in turn, on the number of depressive and posttraumatic symptoms. Although there is evidence that these emotion regulation strategies influence sleep quality (Vandekerckhove & Wang, 2018), no previous study has described simultaneous relations between cognitive reappraisal, sleep quality, and depressive/posttraumatic symptoms. Indeed, only one study has described a mediating effect of sleep quality on relations between avoidance and depressive (but not posttraumatic) symptoms (Warnke et al., 2017).

An interesting note is that, across all analyses, there was only one instance of an emotion regulation strategy having a significant direct effect on psychiatric symptoms (more use of avoidance was directly associated with more depressive symptoms). This is surprising because previous literature shows that emotion dysregulation correlates with the presence of psychiatric symptoms (e.g., poor cognitive reappraisal is associated with depressive symptoms, high avoidance and impulsivity are associated with posttraumatic symptoms; Aldao et al., 2010). The observed pattern of data therefore highlights the importance of sleep quality in mediating associations between emotion regulation and psychiatric symptoms, and suggests that sleep is influential in determining the development of both depressive and posttraumatic symptoms.

Our data have several clinical implications for students presenting with depressive and posttraumatic symptoms. Because we find that students’ use of avoidance, impulsivity, and cognitive reappraisal influences their sleep quality, which, in turn, influences the degree to which they experience clinical symptoms, intervention at either the level of emotion regulation or sleep may promote amelioration of symptoms. Student wellness programmes should routinely evaluate the emotion regulation strategies and sleep habits of students presenting with depressive and posttraumatic symptoms so that clinicians can suggest appropriate interventions, such as emotion regulation therapy (Renna et al., 2017) or cognitive behavioural therapy for insomnia (Mitchell et al., 2012). Furthermore, student orientation programmes should include psychoeducational components emphasising advantageous emotion regulation strategies and healthy sleep habits that may prophylactically protect students from significant symptom burden.

Limitations of the study include, first, that the data were collected via self-report instruments. Although these instruments are well-validated and used frequently in clinical research, reliance on subjective reports of affect and behaviour means data must be interpreted cautiously (Althubaiti, 2016; Lipinska & Thomas, 2017). Second, although the SEM approach allowed us to better understand associations between emotion regulation, sleep, and psychiatric symptoms, the study’s cross-sectional design does not permit definitive causal statements. Third, another factor which constrains the strength of our conclusions is our relatively small sample size. SEMs that include numerous parameters, as we did, typically require a much larger sample size to achieve adequate power and stability. Fourth, data were collected during midyear examinations. Because psychosocial and physiological stress encountered during student exams is associated with sleep disruption, increased use of maladaptive emotion regulation strategies, and more depressive symptoms (Goldberg et al., 2020; Zunhammer et al., 2013), the timing of data collection represents a potential confound.

Future studies might address these limitations by (a) using clinician-rated and observational measures of emotion regulation and psychiatric symptoms, and objective measures of sleep quality; and (b) collecting data longitudinally from a large and diverse sample of students at various stages of the academic year.

Conclusion

We found, in a sample of 336 South African university students, that more use of maladaptive emotion regulation strategies (avoidance, impulsivity), and less use of an adaptive strategy (cognitive reappraisal), was associated with poor sleep quality, which, in turn, was associated with more depressive and posttraumatic symptoms. Therefore, emotion regulation indirectly exerted its influence on the manifestation of psychiatric symptoms through sleep. Because students are particularly vulnerable to depression and trauma, but often have limited access to mental health services (Friedrich & Schlarb, 2018), our data have implications for intervention. Improving sleep quality (e.g., via mobile sleep applications and wearable technology) may be an accessible and affordable alternative to more traditional psychological and pharmacological therapies. Our findings show preliminary evidence in favour of a hypothesis stating that these sleep intervention options may have favourable results in reducing the burden of depressive and posttraumatic symptoms among students.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Lee R Nicholson

Raphaella Lewis

Kevin GF Thomas

Notes

References

Aldao

Nolen-Hoeksema

Schweizer

(2010). Emotion-regulation strategies across psychopathology: A meta-analytic review. Clinical Psychology Review, 30(2), 217–237. https://doi.org/10.1016/j.cpr.2009.11.004

Althubaiti

(2016). Information bias in health research: Definition, pitfalls, and adjustment methods. Journal of Multidisciplinary Healthcare, 9(1), 211–217. https://doi.org/10.2147/JMDH.S104807

Awang

M. M.

Kutty

F. M.

Ahmad

A. R.

(2014). Perceived social support and well-being: First-year student experience in university. International Education Studies, 7(13), 261–270. https://doi.org/10.5539/ies.v7n13p261

Bag

(2015). A short review on structural equation modeling: Applications and future research directions. Journal of Supply Chain Management Systems, 4(3), 64–69. https://doi.org/10.21863/jscms/2015.4.3.014

Baglioni

Nanovska

Regen

Spiegelhalder

Feige

Nissen

Reynolds

Riemann

(2016). Sleep and mental disorders: A meta-analysis of polysomnographic research. Psychological Bulletin, 142(9), 969–990. https://doi.org/10.1037/bul0000053

Bantjes

Lochner

Saal

Roos

Taljaard

Page

Auerbach

Mortier

Bruffaerts

Kessler

Stein

(2019). Prevalence and sociodemographic correlates of common mental disorders among first-year university students in post-apartheid South Africa: Implications for a public mental health approach to student wellness. BMC Public Health, 19(1), Article 922. https://doi.org/10.1186/s12889-019-7218-y

Basharpoor

Shafiei

Daneshvar

(2015). The comparison of experiential avoidance, mindfulness and rumination in trauma-exposed individuals with and without posttraumatic stress disorder (PTSD) in an Iranian sample. Archives of Psychiatric Nursing, 29(5), 279–283. https://doi.org/10.1016/j.apnu.2015.05.004

Blackwell

McDermott

(2014). Test review: Patient Health Questionnaire-9 (PHQ-9). Rehabilitation Counselling Bulletin, 57(4), 246–248. https://doi.org/10.1177/0034355213515305

Browne

M. W.

Cudeck

(1992). Alternative ways of assessing model fit. Sociological Methods & Research, 21(2), 230–258. https://doi.org/10.1177/0049124192021002005

10.

Burnett

P. C.

Dart

B. C.

(1997). Conventional versus confirmatory factor analysis: Methods for validating the structure of existing scales. Journal of Research and Development in Education, 30(2), 126–131.

11.

Buysse

Reynolds

Monk

Berman

Kupfer

(1989). The Pittsburgh Sleep Quality Index: A new instrument for psychiatric practice and research. Psychiatry Research, 28(2), 193–213. https://doi.org/10.1016/0165-1781(89)90047-4

12.

Cholera

Pence

B. W.

Gaynes

B. N.

Bassett

Quangule

Pettifor

Macphail

Miller

W. C.

(2017). Depression and engagement in care among newly diagnosed HIV-infected adults in Johannesburg, South Africa. AIDS and Behaviour, 21(6), 1632–1640. https://doi.org/10.1007/s10461-016-1442-6

13.

Doré

B. P.

Morris

R. R.

Burr

D. A.

Picard

R. W.

Ochsner

K. N.

(2017). Helping others regulate emotion predicts increased regulation of one’s own emotions and decreased symptoms of depression. Personality and Social Psychology Bulletin, 43(5), 729–739. https://doi.org/10.1177/0146167217695558

14.

Friedrich

Schlarb

(2018). Let’s talk about sleep: A systematic review of psychological interventions to improve sleep in college students. Journal of Sleep Research, 27(1), 4–22. https://doi.org/10.1111/jsr.12568

15.

Gámez

Chmielewski

Kotov

Ruggero

Suzuki

Watson

(2014). The Brief Experiential Avoidance Questionnaire: Development and initial validation. Psychological Assessment, 26(1), 35–45. https://doi.org/10.1037/a0034473

16.

Goldberg

Thomas

K. G. F.

Lipinska

(2020). Bedtime stress increases sleep latency and impairs next-day prospective memory performance. Frontiers in Neuroscience, 14(1), Article 756. https://doi.org/10.3389/fnins.2020.00756

17.

Gong

Wang

Liu

(2020). Emotion regulation and depressive symptoms mediate the association between schizotypal personality traits and suicidality in Chinese college students. Archives of Suicide Research, 13, 1–12. https://doi.org/10.1080/13811118.2020.1818655

18.

Gratz

Roemer

(2004). Multidimensional assessment of emotion regulation and dysregulation: Development, factor structure, and initial validation of the difficulties in Emotion Regulation Scale. Journal of Psychopathology and Behavioral Assessment, 26(1), 41–54. https://doi.org/10.1023/B:JOBA.0000007455.08539.94

19.

Gross

John

(2003). Individual differences in two emotion regulation processes: Implications for affect, relationships, and well-being. Journal of Personality and Social Psychology, 85(2), 348–362. https://doi.org/10.1037/0022-3514.85.2.348

20.

Hannan

S. M.

Orcutt

H. K.

(2020). Emotion regulation in undergraduate students with posttraumatic stress symptoms: A multimethod study. Psychological Trauma: Theory, Research, Practice, and Policy, 12(6), 643–650. https://doi.org/10.1037/tra0000577

21.

Hein

Senterre

Lanquart

J.-P.

Montana

Loas

Linkowski

Hubain

(2017). Hyperarousal during sleep in untreated, major depressed subjects with prodromal insomnia: A polysomnographic study. Psychiatry Research, 258(1), 387–395. https://doi.org/10.1016/j.psychres.2017.08.061

22.

Hoyt

Thomas

K. S.

Epstein

Dirksen

(2009). Coping style and sleep quality in men with cancer. Annals of Behavioral Medicine, 37(1), 88–93. https://doi.org/10.1007/s12160-009-9079-6

23.

Bentler

P. M.

(1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling: A Multidisciplinary Journal, 6(1), 1–55. https://doi.org/10.1080/10705519909540118

24.

Kim

Suh

(2017). Social support as a mediator between insomnia and depression in female undergraduate students. Behavioral Sleep Medicine, 17(4), 379–387. https://doi.org/10.1080/15402002.2017.1363043

25.

King

McKenzie-McHarg

Horsch

(2017). Testing a cognitive model to predict posttraumatic stress disorder following childbirth. BMC Pregnancy and Childbirth, 17(1), Article 32. https://doi.org/10.1186/s12884-016-1194-3

26.

Kline

(2015). Principles and practice of structural equation modeling (4th ed.). Guilford Press.

27.

Kroenke

Spitzer

Williams

(2001). The PHQ-9: Validity of a brief depression severity measure. Journal of General Internal Medicine, 16(9), 606–613. https://doi.org/10.1046/j.1525-1497.2001.016009606.x

28.

Lai

(2018). Estimating standardized SEM parameters given nonnormal data and incorrect model: Methods and comparison. Structural Equation Modeling: A Multidisciplinary Journal, 25(4), 600–620. https://doi.org/10.1080/10705511.2017.1392248

29.

Han

Wang

Sun

Cheng

(2018). How social support influences university students’ academic achievement and emotional exhaustion: The mediating role of self-esteem. Learning and Individual Differences, 61(1), 120–126. https://doi.org/10.1016/j.lindif.2017.11.016

30.

Hengchao

Zhongquan

(2007). Reliability and validity of Emotion Regulation Questionnaire Chinese revised version. Chinese Journal of Health Psychology, 15(6), 503–505.

31.

Lipinska

Thomas

K. G. F.

(2017). Better sleep in a strange bed? Sleep quality in South African women with posttraumatic stress disorder. Frontiers in Psychology, 8, Article 1555. https://doi.org/10.3389/fpsyg.2017.01555

32.

Lipinska

Thomas

K. G. F.

(2019). The interaction of REM fragmentation and night-time arousal modulates sleep-dependent emotional memory consolidation. Frontiers in Psychology, 10, Article 1766. https://doi.org/10.3389/fpsyg.2019.01766

33.

Liu

Wang

Zhu

Zhang

Jiang

Wang

Huang

(2020). The relationship between insecure attachment to depression: Mediating role of sleep and cognitive reappraisal. Neural Plasticity, 2020, Article 1931737. https://doi.org/10.1155/2020/1931737

34.

Marroquín

(2011). Interpersonal emotion regulation as a mechanism of social support in depression. Clinical Psychology Review, 31(8), 1276–1290. https://doi.org/10.1016/j.cpr.2011.09.005

35.

McDonald

R. P.

M. H. R.

(2002). Principles and practice in reporting structural equation analyses. Psychological Methods, 7(1), 64–68. https://doi.org/10.1037//1082-989X.7.1.64

36.

McRae

Gross

J. J.

(2020). Emotion regulation. Emotion, 20(1), 1–9. https://doi.org/10.1037/emo0000703

37.

Mîndrilă

(2010). Maximum likelihood (ML) and diagonally weighted least squares (DWLS) estimation procedures: A comparison of estimation bias with ordinal and multivariate non-normal data. International Journal for Digital Society, 1(1), 60–66. https://doi.org/10.20533/ijds.2040.2570.2010.0010

38.

Mitchell

M. D.

Gehrman

Perlis

Umscheid

C. A.

(2012). Comparative effectiveness of cognitive behavioral therapy for insomnia: A systematic review. BMC Family Practice, 13, Article 40. https://doi.org/10.1186/1471-2296-13-40

39.

Mollayeva

Thurairajah

Burton

Mollayeva

Shapiro

C. M.

Colantonio

(2016). The Pittsburgh Sleep Quality Index as a screening tool for sleep dysfunction in clinical and non-clinical samples: A systematic review and meta-analysis. Sleep Medicine Reviews, 25, 52–73. https://doi.org/10.1016/j.smrv.2015.01.009

40.

Nachtigall

Ulf

Friedrich

Rolf

(2003). (Why) Should we use SEM? Pros and cons of structural equation modeling. Methods Psychological Research Online, 8(2), 1–22.

41.

Ncama

B. P.

McInerney

P. A.

Bhengu

B. R.

Corless

I. B.

Wantland

D. J.

Nicholas

P. K.

McGibbon

C. A.

Davis

S. M.

(2008). Social support and medication adherence in HIV disease in KwaZulu-Natal, South Africa. International Journal of Nursing Studies, 45(12), 1757–1763. https://doi.org/10.1016/j.ijnurstu.2008.06.006

42.

O’Leary

Bylsma

Rottenberg

(2016). Why might poor sleep quality lead to depression? A role for emotion regulation. Cognition and Emotion, 31(8), 1698–1706. https://doi.org/10.1080/02699931.2016.1247035

43.

Ong

Thompson

(2019). The importance of coping and emotion regulation in the occurrence of suicidal behavior. Psychological Reports, 122(4), 1192–1210. https://doi.org/10.1177/0033294118781855

44.

Ouimette

Wade

Prins

Schohn

(2008). Identifying PTSD in primary care: Comparison of the primary care-PTSD screen (PC-PTSD) and the General Health Questionnaire-12 (GHQ). Journal of Anxiety Disorders, 22(2), 337–343. https://doi.org/10.1016/j.janxdis.2007.02.010

45.

Palmer

C. A.

Alfano

C. A.

(2017). Sleep and emotion regulation: An organizing, integrative review. Sleep Medicine Reviews, 31, 6–16. https://doi.org/10.1016/j.smrv.2015.12.006

46.

Peltzer

Louw

(2013). Prevalence of suicidal behaviour & associated factors among tuberculosis patients in public primary care in South Africa. The Indian Journal of Medical Research, 138(2), 194–200.

47.

Prins

Ouimette

Kimerling

Cameron

Hugelshofer

Shaw-Hegwer

Thrailkill

Gusman

Sheikh

(2003). The Primary Care PTSD screen (PC-PTSD): Development and operating characteristics. Primary Care Psychiatry, 9(1), 9–14. https://doi.org/10.1185/135525703125002360

48.

Rane

M. S.

Hong

Govere

Thulare

Moosa

M.-Y.

Celum

Drain

P. K.

(2018). Depression and anxiety as risk factors for delayed care-seeking behavior in human immunodeficiency virus-infected individuals in South Africa. Clinical Infectious Diseases, 67(9), 1411–1418. https://doi.org/10.1093/cid/ciy309

49.

Renna

M. E.

Quintero

J. M.

Fresco

D. M.

Mennin

D. S.

(2017). Emotion regulation therapy: A mechanism-targeted treatment for disorders of distress. Frontiers in Psychology, 8, Article 98. https://doi.org/10.3389/fpsyg.2017.00098

50.

Rosseel

(2012). Lavaan: An R package for structural equation modeling. Journal of Statistical Software, 48(2), 1–36. https://doi.org/10.18637/jss.v048.i02

51.

Schmidt

Gay

Ghisletta

Van Der Linden

(2010). Linking impulsivity to dysfunctional thought control and insomnia: A structural equation model. Journal of Sleep Research, 19(1), 3–11. https://doi.org/10.1111/j.1365-2869.2009.00741.x

52.

Seligowski

Lee

Bardeen

Orcutt

(2015). Emotion regulation and posttraumatic stress symptoms: A meta-analysis. Cognitive Behaviour Therapy, 44(2), 87–102. https://doi.org/10.1080/16506073.2014.980753

53.

Semplonius

Willoughby

(2018). Psychosocial adjustment throughout university: A longitudinal investigation of the roles of sleep quality and emotion dysregulation. Journal of Youth and Adolescence, 47(6), 1267–1278. https://doi.org/10.1007/s10964-018-0826-5

54.

Sherbourne

Stewart

(1991). The MOS social support survey. Social Science & Medicine, 32(6), 705–714. https://doi.org/10.1016/0277-9536(91)90150-b

55.

Short

Raines

Oglesby

Zvolensky

Schmidt

(2014). Insomnia and emotion dysregulation: Independent and interactive associations with posttraumatic stress symptoms among trauma-exposed smokers. Journal of Affective Disorders, 165(1), 159–165. https://doi.org/10.1016/j.jad.2014.04.069

56.

Thompson

(2019). Emotion dysregulation: A theme in search of definition. Development and Psychopathology, 31(3), 805–815. https://doi.org/10.1017/S0954579419000282

57.

Valencia

(2018). The Brief Experiential Avoidance Questionnaire: A viable alternative to the AAQ-II? Pensando Psicología, 14, Article 24. https://doi.org/10.16925/pe.v14i24.2412

58.

Vandekerckhove

Wang

(2018). Emotion, emotion regulation and sleep: An intimate relationship. AIMS Neuroscience, 5(1), 1–17. https://doi.org/10.3934/Neuroscience.2018.5.1

59.

Vandekerckhove

Weiss

Schotte

Exadaktylos

Haex

Verbraecken

Cluydts

(2011). The role of presleep negative emotion in sleep physiology. Psychophysiology, 48(12), 1738–1744. https://doi.org/10.1111/j.1469-8986.2011.01281.x

60.

van Wyk

Thomas

K. G. F.

Solms

Lipinska

. (2016). Prominence of hyperarousal symptoms explains variability of sleep disruption in posttraumatic stress disorder. Psychological Trauma: Theory, Research, Practice, and Policy, 8(6), 688–696. https://doi.org/10.1037/tra0000115

61.

Verzani

(2018). Using R for introductory statistics. CRC Press.

62.

Wang

Cheng

(2019). Systematic review and meta-analysis of the relationship between sleep disorders and suicidal behaviour in patients with depression. BMC Psychiatry, 19, Article 303. https://doi.org/10.1186/s12888-019-2302-5

63.

Warnke

Markarian

Pickett

(2017). Experiential avoidance predicts depression through stress, pre-sleep arousal, and poor sleep quality. Sleep, 40(Suppl. 1), A408–A409. https://doi.org/10.1093/sleepj/zsx050.1095

64.

Weiss

N. H.

Gratz

K. L.

Lavender

J. M.

(2015). Factor structure and initial validation of a multidimensional measure of difficulties in the regulation of positive emotions. Behavior Modification, 39(3), 431–453. https://doi.org/10.1177/0145445514566504

65.

Zhang

Ren

Sanford

Yang

Zhou

Zhang

Wing

Y.-K.

Shi

Tang

(2019). Sleep in posttraumatic stress disorder: A systematic review and meta-analysis of polysomnographic findings. Sleep Medicine Reviews, 48(1), Article 101210. https://doi.org/10.1016/j.smrv.2019.08.004

66.

Zunhammer

Eberle

Eichhammer

Busch

(2013). Somatic symptoms evoked by exam stress in university students: The role of alexithymia, neuroticism, anxiety and depression. PLoS ONE, 8(12), Article e84911. https://doi.org/10.1371/journal.pone.0084911