Current Knowledge/Study Rationale: Insomnia and nightmares confer increased risk for suicidal ideation in young people. However, the psychological processes driving this association are not yet known. This study tested a hypothesized multistep mediational pathway, derived from the Integrated Motivational Volitional Model of Suicidal Behaviour, to further understanding regarding the role of two robust suicide-related constructs, defeat and entrapment, as mediators within the sleep-suicidal ideation relationship in adolescents.
Study Impact: Taken together, analyses supported the hypothesized role of defeat and entrapment, as mediators, driving the psychological pathway from sleep disturbance to suicidal ideation in young people. These findings suggest that suicide risk assessment, prevention, and intervention efforts aimed at young people should consider insomnia, nightmares, defeat, and entrapment as important modifiable clinical targets.
During adolescence, both sleep problems and suicide represent major public health concerns.1,2 A converging body of research evidence, spanning different study designs, samples, and assessment techniques, suggests that sleep disturbances represent evidence-based risk factors for suicidal thoughts and behaviors.3–5 The transition from childhood to adolescence is accompanied by several biological and psychosocial changes that interact to create an increased vulnerability to sleep problems in young people.6 Two specific types of sleep disturbance that are frequently reported during adolescence are insomnia and nightmares. Insomnia is the most common sleep disorder in adolescence.7 When employing Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5) criteria for insomnia, a recent study reported a prevalence rate of 18.5%.8 Nightmares are also common during adolescence.9,10 One investigation, conducted recently within the United Kingdom, reported that 19.4% of adolescents experienced frequent nightmares.11 Both insomnia and nightmares confer increased risk for suicidal ideation in young people. These relationships have been demonstrated in both community and clinical samples.9,10,12–17
Although the relationship between clinically salient sleep problems and suicidality is well established, research examining the psychological mechanisms accounting for this association is in its infancy. Research has begun to uncover the psychological processes linking sleep and suicide in adults.18–21 For example, Littlewood and colleagues19 recruited adults, with experience of trauma and symptoms of posttraumatic stress disorder (PTSD), and tested a serial mediation pathway whereby nightmares were predicted to have an indirect effect on suicidal behaviors through perceptions of defeat, entrapment, and hopelessness (pessimism for the future). Defeat refers to a sense of failed struggle in relation to the loss or disruption of some valued status or internal hierarchical aims, while entrapment is characterized by an individual's perception of their circumstances as being inescapable.22 Defeat and entrapment are considered to be distinct but overlapping constructs, in the same way a depression and hopelessness are correlated but have discriminant validity.23 The authors derived this pathway from the Cry of Pain Model of suicidality and the predicted mediational model was supported by bootstrapped analysis. These findings have begun to help us better understand the mechanisms by which nightmares confer suicide risk in adults. However, investigations of this nature within adolescent samples are particularly limited. A recent systematic review24 highlighted only one study that had examined potential psychological mediators of the sleep-suicide relationship in this age group.25 That is, it remains unclear why sleep problems are associated with increased suicidal risk within this unique26 developmental period.
Further, it is not yet known whether the psychological constructs underlying these relationships are consistent across different types of sleep disturbances. While investigations consistently demonstrate a direct relationship between nightmares and suicidal thoughts and behaviors,21,27,28 the evidence regarding the association between insomnia and suicidality is mixed.19,28,29 It remains to be seen if similar or distinct psychological pathways lead to suicidal ideation from insomnia relative to nightmares.
Within the field of suicide research, moving beyond main-effect models to more complex theoretically informed models of suicide risk (involving mediators and moderators) has been highlighted as a priority for future investigations.30 Research of this nature is crucial in terms of understanding why certain factors are associated with enhanced or diminished suicide risk. The Integrated Motivational-Volitional Model of Suicidal Behaviour (IMV)23 is advantageous in this context as it provides a number of testable hypotheses, thereby highlighting candidate psychological mediators of the sleep-suicide ideation relationship.
The IMV: A Theoretical Context for Examining the Psychological Mechanism Linking Sleep Disturbance and Suicidal Ideation
The IMV proposes several variables that affect an individual's progression toward suicidal ideation, and the translation of these thoughts into suicidal acts. Intention to harm oneself is hypothesized to be determined by feelings of entrapment, the perception that there are no means of escape, and results in suicidal behavior being seen as the salient solution to life circumstances. Entrapment is triggered by defeat/humiliation, which is often associated with stress. There is a substantial body of evidence that supports the hypothesized role of defeat and entrapment in psychological pathways giving rise to suicidal thinking.31–33 On the basis of the IMV, it could be hypothesized that sleep disturbances may act as a stressor34–36 that can trigger feelings of defeat, which then contribute to a sense of entrapment.
Although sleep problems are not included within the IMV model, recent findings19,37 provide support for the pathway to suicidality proposed within this theoretical framework.38 Specifically, experience of nightmares acted as a vulnerability factor, and was associated with increased perceptions of defeat and entrapment. Research to date has employed adult samples recruited from clinical populations. Despite the fact that young people are at high risk of sleep disturbances and of the development of suicidal thoughts, the potential mediating role of defeat and entrapment on the sleep disturbance-suicidal ideation relationship has not yet been investigated in a sample of adolescents. Given that most adolescents experiencing these thoughts will not come into contact with mental health services, it is important to determine whether these findings extend to a community sample of adolescents. Research of this nature, examining the hypothesized multistep pathways linking sleep disturbance and suicidal ideation via perceptions of defeat and entrapment, is warranted. This is the case because identifying modifiable risk factors that trigger the start of this trajectory, in young people, will be helpful in developing effective suicide prevention strategies.
The Current Study
The overarching aim of the current study was to test predictions, derived from the IMV, in order to advance understanding of the psychological mechanisms linking sleep disturbance (ie, insomnia and nightmares) and suicidality in adolescents.
More specifically this study aimed to:
Determine whether insomnia and nightmares were associated with suicidal ideation independent of depressive symptoms.
Investigate differences in levels of defeat and entrapment between adolescents with and without clinically salient sleep disturbances (insomnia and nightmares).
Test hypothesized multistep pathways linking sleep disturbance and suicidal ideation, via perceptions of defeat and entrapment (Figure 1).
It was predicted that (1) insomnia symptoms and nightmares would be positively related to suicidal ideation independent of depressive symptomology, (2) adolescents experiencing clinically salient sleep disturbances would report elevated perceptions of both defeat and entrapment, and (3) the association between subjective sleep disturbance and suicidal ideation would be linked via an indirect pathway whereby sleep disturbance would be linked to defeat, defeat would be associated with entrapment, and entrapment would be related to suicidal ideation.
Predicted serial multiple mediation pathway (highlighted in nondashed lines) for association between insomnia/ nightmares and suicidal ideation, via defeat and entrapment.
Predicted multiple mediation pathway indicated by solid lines.
Predicted serial multiple mediation pathway (highlighted in nondashed lines) for association between insomnia/ nightmares and suicidal ideation, via defeat and entrapment.
In investigating the role of defeat and entrapment in the pathway to suicidality, the current study complements the important work of Littlewood and colleagues.19 However, the differences between both investigations should be considered to fully grasp the unique contribution that the current study intends to make to the literature. Littlewood et al. examined a clinical sample of 91 individuals who had experienced traumatic events and symptoms of PTSD. Comparatively, we were interested in examining a national community sample of adolescents. Littlewood and colleagues examined a serial mediational pathway, derived from the Cry of Pain Model of suicidality, whereby nightmares were predicted to have an indirect effect on suicidal behaviors through defeat, entrapment, and hopelessness (while controlling for depression and insomnia). The current investigation sought to test a serial mediational pathway, derived from the IMV, in two distinct sleep problems. Given that the IMV proposes that entrapment is a predictor of suicidal intention, the current study predicted that insomnia/nightmares would have an indirect effect on suicidal ideation through defeat and entrapment (while controlling for depression and nightmares/insomnia).
The overarching aim of the current study was to test theoretically derived predictions to advance understanding regarding the psychological mechanisms by which insomnia and nightmares confer an increased risk for suicidal ideation in young people. Results of this investigation reinforce previous evidence indicating that sleep disturbances, namely insomnia and nightmares, are associated with an increased risk of the development of suicidal thoughts during adolescence.9,10,13–17 Our findings highlighted that young people reporting symptoms of insomnia were three times more likely to report thoughts of suicide. This was also the case for those experiencing severe and intense nightmares. This association remained significant after controlling for the severity of depressive symptomology.
Moderate differences in feelings of defeat and entrapment were observed between individuals with clinically salient sleep complaints and those without. Specifically, young people experiencing insomnia or nightmares reported elevated perceptions of both defeat and entrapment. To the best of our knowledge, this is the first study to investigate these suicide-related psychological constructs in relation to insomnia, overall, and nightmares in young people. The current study provided support for the hypothesized multistep mediational pathway in that results of the mediation analysis suggested that nightmares are associated with defeat, defeat is linked to entrapment, and entrapment is related to thoughts of suicide. These findings reinforce recent research, from Littlewood and colleagues19, conducted with adults experiencing PTSD, which highlights the underlying role of defeat and entrapment within the pathway to suicidality. Further, this study demonstrates that these pathways partially account for this relationship in a community sample. Support for the hypothesized multistep mediational pathway was also provided when considering insomnia symptoms as a predictor. These findings offer novel insights regarding the link between sleep disturbance and suicidal thoughts, and provide preliminary evidence to suggest that these patterns are not specific to the experience of nightmares. This study reinforces the importance of defeat and entrapment as potentially transdiagnostic psychological constructs underlying suicidality.6,46,47
Given that this is the first study to examine the role of defeat and entrapment in the context of the sleep-suicide ideation relationship, within a community sample of adolescents, replication is important. Most research to date, examining the role of entrapment within the pathway to suicidality, has been conducted in relation to suicidal ideation, rather suicidal behavior. As a result, the evidence base is currently stronger for ideation than behavior.38 The authors highlight that this should be interpreted as absence of evidence rather than evidence of absence. Therefore, it is important that future research, examining sleep disturbance as a marker of risk, seeks to extend the current research to suicidal behavior.
Future research seeking to disentangle the nature of the link between insomnia/nightmares and defeat is warranted. With regard to insomnia, it is possible that perceptions of defeat could arise as a result of the effect of the sleep disorder on daytime functioning. Research conducted by Kyle et al.,48 employing focus groups and audio diaries, has demonstrated that individuals experiencing chronically disturbed sleep report daily difficulties with cognitive, emotional, interpersonal, and physical functioning. The impairments cumulatively were reported to be associated with compromised social functioning, feeling like an outsider, the notion of struggling through the day, and reduced quality of life. These experiences could be conceptualized as “defeating.” However, Kyle et al.'s study has been conducted solely in adult samples and so findings should be applied to young people with caution. Concerning nightmares, it has been suggested that defeat may occur because of the nightmare content, or alternatively the individuals' perception of their ability to cope or manage nightmares. These predictions should be scientifically tested in order to clarify mechanisms underlying the link between sleep disturbances and defeat.
Moderate between-group differences in entrapment were demonstrated when comparing young people with and without clinically salient sleep disturbance. This suggests that sleep disturbance during adolescence could lead young people to feel trapped by both their own thoughts and feelings in addition to external events or situational factors. Further, direct effects existed between both sleep disturbances and entrapment. Future research should seek to obtain a more fine-grained understanding of the cognitive mechanisms that contribute to entrapment within the context of sleep disorders. Littlewood et al.37 hypothesized, based on interviews with adults with experience of depression, that sleep may offer an alternative to suicide by providing a temporary escape from problems in daily life. As such, difficulties initiating and maintaining sleep associated with insomnia and disturbed sleep associated with nightmares may deny these young people such an escape, leading them to feel trapped.
It is important to note that perceptions of defeat and entrapment partly, but not fully, mediated the relationship between nightmares and suicidal ideation. This suggests that additional pathways may underlie this relationship and future research should focus on identifying other candidate psychological mechanisms (eg, emotion regulation, impulsivity, rumination, or neurocognitive deficits49). However, the direct link between insomnia and suicidal ideation was not maintained within the full mediational model, supporting the notion that this effect is fully accounted for via these pathways.
Taken together, these findings provide preliminary evidence to suggest that insomnia and nightmares may confer risk for suicidal ideation in young people via increased perceptions of defeat and entrapment. On the basis of the IMV we hypothesized that sleep disturbance, in the form of insomnia or nightmares, could act as a stressor that contributes to perceptions of defeat, defeat then is associated with entrapment, and entrapment is linked to suicidal ideation. Findings provided support for this hypothesized pathway in two distinct sleep disorders, providing preliminary evidence for this pathway in young people and ultimately suggesting that the IMV is a useful framework for examining the relationship between sleep problems and thoughts of suicide. Future research should seek to extend these findings to suicidal behavior, while employing the IMV as an overarching framework.
The current study demonstrated that sleep disturbances and suicidal ideation are common and represent a public health issue for adolescents in Scotland. Insomnia and nightmares are prevalent within this population, with 11.9% of young people reporting clinically salient insomnia symptoms and 19.5% endorsing frequent and severe nightmares. While research investigating insomnia in adolescents, using DSM-5 criteria is sparse, Hysing et al.8 reported an insomnia prevalence rate of 18.5% in their sample of Norwegian adolescents. Although this is higher than the prevalence rate reported in the current study, there were notable differences in recruitment strategies employed by both studies. Hysing et al. recruited both school attendees and nonattendees, whereas the current study excluded those that were absent from school during the data collection period. Given that insomnia symptoms are associated with higher odds of nonattendance at school50 these differences in recruitment could partially account for the discrepancies in prevalence rates. Although nightmares are reportedly common during adolescence, the exact prevalence rate is not yet clear. Comparisons between studies are complicated by large variations in operationalizations and measurement. Almost 20% of young people within the current study reported experiencing frequent and severe nightmares; this is in line with previous work conducted within the United Kingdom.11
Within the current study, 22.8% of Scottish adolescents reported suicidal ideation. A systematic review of 128 international school based and community studies51 demonstrated a higher proportion of adolescents (29.9%) reporting suicidal ideation (95% CI 26.1, 33.8). Nevertheless, the prevalence of suicidal ideation within the current investigation is notably higher (approximately 10%) than a separate recent study investigating suicidal and self-harming thoughts and behaviors in a community sample of Scottish adolescents.44 Diversity in prevalence rates may be accounted for by differences in sampling strategies. For example, Quigley et al. recruited participants aged 11 to 17 years, whereas the current study invited participation from classes in which most young people were aged 15 to 17 years. This age range was chosen as adolescent risk-taking behavior peaks at this time and suicide is the second leading cause of death for 15 to 24-year-olds.44
Examining the link between sleep and suicidal ideation within the context of a theoretical framework could have implications for risk assessment and prevention. For example, the current study supports the potential utility of defeat and entrapment as indicators of suicidality in young people with insomnia and/ or nightmares and could have implications for identifying those at risk of suicide. Further, sleep, as well as perceptions of defeat and entrapment, offer modifiable targets for interventions to target. Research suggests that cognitive behavioral therapy (CBT) for insomnia reduces suicidal ideation in adults. However, further investigation is required to determine if CBT for insomnia has comparable effects in young people, and to establish if sleep interventions lead to a reduction of defeat and entrapment.
It is important, however, to interpret these findings in the context of the following limitations. First, because of the cross-sectional nature of the data we were unable to investigate causal relationships between variables of interest. Temporal precedence is important in understanding mediational analyses52 and cross-sectional designs do not allow an understanding of how processes unfold over time. Future research employing prospective designs will be crucial in providing important temporal information regarding identified relationships and mediational pathways. Further, this cross-sectional data was collected during a 3-week period in June. Given that both sleep patterns and suicidality demonstrate variability across different seasons, it is possible that the time of year in which data were collected may confound the results. Longitudinal cohort designs (with multiple data collection time points) would allow researchers to examine and account for the effects of seasonal variation on this relationship.
Second, the investigation relied solely on self-report measures. As a result, it is possible that results could have been affected by memory related biases, socially desirable responding or demand characteristics. However, all scales employed, with the exception of the single item suicidal ideation measure, are recognized as reliable and valid. This is advantageous given that, until recently, most research examining the link between sleep and suicidal ideation in adolescent samples has employed sleep measures that are not standardized nor validated and that do not assess the full range of symptoms for sleep disturbances9,14,15 (eg, insomnia is a multidimensional disorder with a characteristic profile of daytime and nighttime symptoms). Given the size of the sample and the research setting of the current study, it was not possible to employ the gold standard clinical evaluation to validate insomnia symptoms and diagnosis. However, the SCI is reliable, valid, and is based on the most up-to-date (DSM-5) diagnostic criteria for insomnia.23 and the Disturbing Dreams and Nightmare Severity Index20,28 has been employed frequently within suicide research. Future research would benefit from combining both objective and subjective measures of sleep. Employing daily diary or experience sampling methodologies would circumvent each of the aforementioned limitations and so should be considered in future research exploring the relationship between insomnia, suicidal ideation, and underlying psychological mechanisms.
Finally, lifetime suicidal ideation was measured using a single-item self-report measure. The use of single-item self-report measures to assess the presence of suicidal ideation, plans, or attempts represents a widespread limitation within this literature. Although researchers have established definitions for suicidal ideation, single-item questions do not inform participants of this definition and so they are required to rely on their own interpretation of the measure. This may lead to misclassification. Further, single-item measures fail to capture the breadth of the construct (eg, severity and frequency). It should be noted that we assessed sleep disturbances within the last month and measured suicidal ideation across the lifetime of the participants. This represents a limitation as it is possible that a young person's experience of suicidal ideation may have occurred in advance of the development of a sleep problem. Future research should seek to build on the understanding provided by this study using prospective designs as well as reliable and valid measures of suicidal ideation.
Despite these limitations, this study represents the first examination of the relationship between sleep disturbances and suicidal ideation, in young people, within the context of a theoretical framework. Findings provide preliminary evidence to suggest that prevention efforts and intervention efforts aimed at reducing suicide risk in young people should consider that sleep disturbance, and feelings of defeat and entrapment, could be important clinical targets to reduce risk among young people.