Case Vignette 2
N, a 23 year old woman, training to be a painter, came to the psychiatric Outpatient Department (OPD) in September, with complaints of sleep disturbance, primarily early insomnia. The MSE revealed that she had low mood, anhedonia with sleep and appetite disturbances since the last 4 months. She complained of a recurrent dream of falling into nothingness and would wake up screaming. Several times in a week, she could not sleep the entire night, in fear of this nightmare. The precipitating factor for the current depressive episode seemed to be an inability to cope with the curriculum of a prestigious Art School in which she had enrolled in June. She returned to her hometown after a month of joining the Art School. A sense of failure and guilt feelings about letting her parents down persisted within her. She was prescribed a selective serotonin reuptake inhibitor (SSRI) drug, Sertraline, and simultaneously started on Cognitive Behaviour Therapy, twice a week. Her father and paternal grandmother had a history of bipolar disorder.
She came regularly for psychotherapy and improved significantly. Her guilt, related to the sense of failure and letting her parents down, diminished remarkably. Her biological functions like sleep and appetite improved. As her mood became better she started painting again. She had worked as a Disc Jockey on part-time basis. This job required her to be witty and humorous. She started hosting the shows again although she reported that the fun was still missing. After three months of intensive psychotherapy she dropped out of the therapy without notice. Her mother called to inform that she was progressing satisfactorily. After three weeks, N came to the OPD to inform that she was helping a friend of hers who was becoming suicidal and this strain had again led to sleep disturbance and the dreams were reoccurring. The MSE did not elicit any psychopathology apart from sleep disturbance and nightmares. She was encouraged to practice sleep hygiene methods and to resume psychotherapeutic sessions. Her mother was asked to help her monitor her medication to regulate her sleep cycle again. Nonetheless, a week later she was dead by suicide. Her parents could not come up with any explanation for this act of hers. In fact, a day previous to her suicide she had expressed her satisfaction and happiness in successfully helping her friend overcome suicidal thoughts.
Discussion
In this case, a strong relationship between poor quality of sleep, nightmares and suicide was revealed. Therapeutic sessions substantiated the above finding as she had expressed her anguish mostly regarding her poor sleep patterns and a sense of failure. As therapy progressed, her distress was more strongly associated with disturbed sleep and nightmares than a sense of failure. Although depressive symptoms appeared to have remitted and the subjective distress of low mood was apparently absent, recurrence of depression cannot be ruled out.
The reversal of direction could be due to confounding variables. As hypersomnia was excluded from the predictor multivariate analyses due to multi-collinearity with hopelessness, which itself is a well-known predictor of later suicidal behaviour, only speculations are possible on this relationship as reported in this study. It appears that initial insomnia is an important bivariate associate and predictor of attempted suicide, regardless of age and time while the role of other sleep disturbances varied with age and time.
As the onset timing of each symptom was not analysed in this study, it is not possible to comment on the impact of the timing of onset and duration of these sleep disturbances as related to attempted suicide, although some sort of an influence, which may or may not be prodromal or residual or both in nature, seems to be indicated by the change of direction of initial insomnia for older adolescents and young adults. As this study was primarily not about sleep disturbances or insomnia, its ability to draw related conclusions are limited despite it raising interesting questions. It therefore does not answer the call for longitudinal studies that examine the occurrence of future suicidal behaviour in those with and without sleep disturbances at baseline by the authors of the latest review on the topic, viz., Liu and Buysse (2006, described later).
In an early study of depressive symptoms and suicidal behaviour among 64 consecutively hospitalized adolescents, insomnia was identified as being bivariately significantly related to suicidal tendencies (p<0.001) and multivariately to seriousness of intent (p<0.03) (Robbins and Alessi, 1985). All items were assessed by the SADS semi-structured clinical interview. However, this study also reported negatively significant relationships between initial insomnia and suicidal tendencies (p<0.001), somewhat similar to Nrugham and colleagues (2008). Although insignificant, negative relationships were also found between initial, middle and terminal insomnia and all the four aspects of suicidality probed for in the study: suicidal tendencies, number of gestures, seriousness of intent and medical lethality. All these four aspects were significantly and positively correlated to each other and all assessments were made after the suicidal behaviour. Multi-collinearity was not controlled for and therefore, the results must been regarded with due consideration for this absence.
Insomnia Among Suicidal Adolescents: Post Traumatic Stress Disorder (PTSD)
A study designed to describe and evaluate the clinical pattern of 14 youths (aged 10 to 18 years) with presenting suicidality, found a common pattern characterized by: (a) suicidality, (b) insomnia, (c) bodily reactions such as stiffness and pain, and (d) deranged mood regulation with involuntary intrusions of negative emotive states and images (Hogberg and Hallstrom, 2008). This condition, the authors cite, has been described earlier in a similar way in a post-rape study by Darves-Bornoz in 1997. The authors called this condition a suicidal trauma reaction. Choquet, Darves-Bornoz, Ledoux, Manfredi & Hassler (1997) have also reported that rape victims among a nationally representative sample of high school students [n = 8140, 48.7% males, 51.3% females, mean age = 16.2 (2.02) years] were associated with attempted suicide (males) and current sleep difficulties (both genders) along with a host of behaviour problems.
Case Vignette 3
K is a 14 year old girl, studying in Class IX, coming from an upper middle-class family. Both parents had been married before to different persons. K’s stepfather had a son from his previous marriage and one girl child was born from his marriage with K’s mother. However, soon after this second marriage, disputes arose between the parents regarding styles of parenting. The stepfather was a conservative and domineering man whereas the mother was more liberal with the children. As K also had primary dyslexia, her mother had been encouraging her to participate in sports and extra-curricular activities, which the step-father opposed. All three children witnessed repeated physical abuse of the mother who supported their interests and ambitions. The mother had lodged a police complaint against the father a month before she came to the Child and Adolescent Guidance Clinic. Consequently, the father was evicted out of the family’s house by police. He took away his son and repeatedly demanded that the younger daughter, K’s younger sister be given to him. When the mother refused to hand over the younger daughter to him he would create a scene outside the house. K was very embarrassed by these daily occurrences. One day the father threatened bodily harm to K if the mother persisted in her refusal to hand over the younger daughter to him. K developed fever after hearing this and was taken to a general physician. She was treated for fever that improved in 3 days time.
However, K developed sleep difficulties, intermittent awakening and hypervigilance. Her sleep patterns did not improve and she would wake up screaming several times in the night after experiencing nightmares of being harmed by her stepfather. She talked a few times about killing herself, to her mother, which prompted the anxious mother to bring K to the clinic. K and her mother refused to meet the psychiatrist for medication. She was then seen twice a week for psychotherapeutic sessions. Deep muscular relaxation technique and sleep hygiene methods were taught to her. Cognitive Behaviour Therapy, catharsis and Behaviour Therapy techniques were used to alleviate her fear and stress. Both parents were called in for marital therapy even though both of them had individually decided not to live together. Mutual agreement was reached where the father was allowed to meet the younger daughter. He profusely apologised to K and promised never to harm her. Gradually, after 15 sessions of psychotherapy, K’s mental health improved. Her sleep patterns became normal and her suicidal ideation disappeared.
Discussion
In this case, a fairly strong relationship was indicated between traumatic stress, sleep disturbance and suicidal ideation. The traumatic stress led to sleep disturbance and consequent suicidal ideation although a suicide attempt was not known to be made. The improvement in symptoms occurred as soon as the ongoing traumatic stress decreased leading to improvement in sleep patterns.
Insomnia Among Adults and Adolescents: Suicide
In order to investigate whether certain DSM-IV depressive symptoms were more prevalent among individuals who die in the context of a major depressive episode and those who do not, whether this increased prevalence was associated with proximal or distal suicide risk, and whether depressive symptoms cluster to indicate suicide risk, 156 suicides who died in the context of a major depressive episode were compared with 81 major depressive controls (McGirr et al, 2007). They found that independent of concomitant axis I and II psychopathology, depressive symptoms among the suicides were more likely to include insomnia [OR(95%CI) = 2.3(1.2-4.6)] and recurrent thoughts of death or suicidal ideation [OR (95%CI) = 12.5(4.8-32.4)], among other depressive symptoms. They also found that the concomitant presence of weight or appetite gain and hypersomnia was associated with decreased suicide risk [OR(95%CI) = 0.29(0.06-1.2)]. The authors of this study concluded that inter-episode symptom concordance seemed to suggest insomnia as an immediate indicator of suicide risk, while other depressive symptoms were not.
Case Vignette 4
S, male, aged 15 years, studying in Class IX in a public school and the only child of an upper socio-economic status family was brought to the Child Guidance Clinic. Presenting complaints were disturbed sleep, disruptive and quarrelsome behaviour in school for more than 6 months. His parents revealed that S’s school had been repeatedly complaining about his disruptive behaviour in class and his bullying behaviour with classmates and younger children in the school bus. As he was the only child of his parents, most of his demands had been promptly fulfilled since childhood. From infancy and throughout his childhood, his sleeping time had ranged between 5-6 hours. He would sleep intermittently and wake up several times during the night. At the age of 6 years he was diagnosed to be suffering from borderline attention-deficit hyperactivity. However, he was never put on psychopharmacological therapy. As he grew older his hyperactivity decreased but attention span remained short. Academically, he was at the average level in his class, obtaining between 50-60% marks. He was brought to the OPD for his continuing behavioural disturbance. The routine mental status examination did not reveal any gross pathology except poor adjustment. Cognitive functions were grossly normal and attention span was adequate. He was advised to come the following day for detailed psychological evaluation and a psychiatric consultation for sleep disturbance. Probes for a family history of any psychiatric disorder did not reveal positive findings.
However, the same evening, his father developed angina pain and was admitted to a hospital. Consequently, the patient was not brought to the Clinic. The patient behaved responsibly during this period and was a support to his mother. During this period his sleep was disturbed and he appeared to be restless. These symptoms were taken as a natural anxiety for his father’s health. On the 12th day after the father’s admission, the patient went to meet him in the hospital. On the way back he was happy that his father would be discharged after 2 days. He planned a surprise holiday for his father and appeared in good spirits, chatting with his mother and paternal grandfather. Soon after lunch, he suddenly locked his mother in her room. He found a rope, called out to his mother that he was ending his life and hung himself. No significant event/symptom could be elicited from the parents prior to the suicide except sleep disturbance and restlessness.
Discussion
In this case, hidden depression cannot be ruled out, keeping in mind the history of poor adjustment. Young boys are known to exhibit aggressive behaviour when distressed or depressed, even in dysthymic condition. Sleep disturbances are a part of the clinical profile of depression.
Global insomnia was found to be one of the six clinical features significantly associated with those who completed suicide (p = 0.011) within one year (n = 13) among 954 psychiatric patients with major affective disorders also assessed by SADS (Fawcett et al, 1990). Half of the patients were 36 years old or younger and the mean age was 38.1 years. The authors called clinical symptoms that may be responsive to early clinical intervention as ‘modifiable risk factors’ and one year as short-term risk period amenable to therapeutic intervention that can substantially reduce acute suicidal risk. Nearly 20 years later, sleep difficulties preceding death in a sample of 140 adolescent suicide completers as compared with a matched sample of 131community control adolescents were thoroughly investigated using a psychology autopsy protocol and semi-structured psychiatric interview in the only study of its kind among adolescents (Goldstein, Bridge & Brent, 2008). Their findings indicated that when compared with controls, suicide completers had higher rates of overall sleep disturbance, insomnia and hypersomnia than controls, within the last week and the current affective episode, even after adjustment for the differential rate of affective disorder between the two groups. When severity of depressive symptoms was accounted for, overall sleep disturbance (last week and present episode) and insomnia (last week) distinguished completers from controls. This paper concluded that the findings supported a significant and temporal relationship between sleep problems and completed suicide among adolescents, just as Fawcett and colleagues (1990) had done among adults.
Reviews: Adults and Adolescents
Reviewers examining the importance of sleep regulation and behaviour in pathways to adolescent health found that while substantial evidence for bi-directional effects between sleep and behavioral/emotional regulation existed, there is mounting evidence that sleep deprivation has its greatest negative effects on the control of behaviour, emotion, and attention, a regulatory interface that is critical in the development of social and academic competence, and psychiatric disorders (Dahl and Lewin, 2002). The reviewers pointed out that clinicians experienced with these problems have pointed out that in many cases, it is difficult to differentiate decreased motivation, school refusal/anxiety, delayed circadian phase, attention difficulties, and depressive symptomatology indicating the clear need for the careful assessment of sleep patterns and behavioural symptoms for prevention, accurate diagnosis, and/or treatment planning.
After examining the scientific literature (predominantly biological) on the relationship of sleep with suicide, two reviewers called for further studies to investigate the possible role of sleep disturbance in suicidal behaviour (Singareddy and Balon, 2001). They summarized that although sleep-related complaints and EEG (electroencephalographic) changes have been seen widely across the spectrum of psychiatric disorders, sleep complaints such as insomnia, hypersomnia, nightmares and sleep panic attacks are common in suicidal patients. The authors added that the subjective quality of sleep as measured by self-rated questionnaires also appeared to be more disturbed in depressed patients who were also suicidal. They hypothesized that one mechanism responsible for the possible association between suicide and sleep could be the role of serotonin (5HT) and that the intervening factor between serotonin and suicide could be the dysfunctional control of aggression. They supported their hypothesis with the observations that serotonergic function had been found to be low in patients who attempted and/or completed suicide, particularly those who used violent methods and that serotonin has been documented to play an important role in the onset and maintenance of slow wave sleep and REM (rapid eye movement) sleep which has been documented to be increased in suicidal patients with depression, schizoaffective disorder and schizophrenia while 5HT2 receptor antagonists have been reported to improve slow wave sleep and that agents that enhance serotonergic transmission decrease suicidal behaviour.
A recent review concluded that sleep loss or disturbances were likely to signal an increased risk of future suicidal action in adolescents and that the link between insomnia and suicidal behaviour appeared to be mediated by depression (Liu and Buysse, 2006). The conclusion of these reviewers gains additional relevance in the light of sleep disturbances being argued to be a core symptom of depression with emphasis on the early restoration of sleep in the management of major depression among adults (Kennedy, 2008). However, polysomnographic and neuro-endocrine studies in children and adolescents have not found consistent changes in sleep architecture paralleling adult major depression (Ivanenko, Crabtree & Gozal, 2005). Here it is pertinent to note that Conroy and colleagues (2006) have documented that perception of poor sleep is associated with significant distress and consequences even in the absence of objective polysomnography findings. This distress would need to be clinically addressed, regardless of mechanical/laboratory evidence, more so, among children and adolescents.
A review on pharmacologic treatment approaches for children and adolescents with posttraumatic stress disorder stated, as a part of its conclusions that even the reduction in one disabling symptom, such as insomnia or hyperarousal, may have a positive ripple effect on a child’s overall functioning (Donnelly, 2003).
Conclusion
It is heartening to note that concerted efforts are being made to define insomnia and that research interest is being shown on this topic. Such efforts will go a long way in providing the spurt required in insomnia research, which is required in order to reject/accept hypotheses regarding the association of insomnia with different aspects of suicidality.
We see that published research on the relationship between sleep disturbances and deliberate self-harm is absent, while publications of findings from research projects studying the relationship between sleep disturbances and completed suicide are rare, both in the adult and adolescent literature. The relationship between sleep disturbances and PTSD has been explored equally sparsely as with suicide while the bulk of existing literature is on the relationship between sleep disturbances and either suicidal ideation or attempted suicide or both these aspects of suicidal phenomena. This available literature is essentially cross-sectional and many of the research projects or the studies were not designed specifically for the purpose of examining these relationships, with the findings on sleep coming as a byproduct of research on some other issue, usually depression. Biological aspects of this relationship have hardly been researched and this gap needs to be filled.
Although it is noted that insomnia and other sleep disturbances are common, chronic and expensive to society when untreated, neither insomnia nor sleep disturbances have received the attention they deserve from researchers. Clinicians, on the other hand, have been vigilant to the distresses in their patients and use their skills to include sleep management as a core part of the case management, even in the absence of support from research studies. Among adolescents, the vulnerability added by the presence of insomnia/other sleep disturbances is increased due to the increased task mastery expected from them and the incomplete coping repertoire available to them. This vulnerability appears to be enhanced among females. It was also seen from the review of research findings that although there are some gender specific findings, there is not enough replication to draw conclusions or to validate generalizations.
In both epidemiological and clinical studies, it was seen that the relationship between insomnia and all aspects suicidality remained statistically significant even in multivariate models of adult samples, while the same was true in the adolescent samples only for suicidal attempts and was inconsistent for suicidal ideation. The role of initial insomnia as an associate and as a predictor of suicide attempts from adolescence to early adulthood has been documented only in a single longitudinal study with a sample that is in-between the general population and clinical sample and needs to be replicated among both samples. On the other hand, the negative relationships between types of insomnia, including initial insomnia and aspects of suicidality, documented in clinical studies among adolescents must also be thoroughly investigated for confounding factors so that the true relationship can emerge. In doing so, not only will the nature and direction of the relationship be revealed but also the mechanism of its functioning. It was also seen that the statistical relationship between insomnia/sleep disorders were impervious to age, race and other sociodemographic variables but remained sensitive to psychiatric/psychosomatic variables. Insomnia and other sleep disturbances must be addressed by the clinician during case management, regardless of whether these disturbances are prodromal, residual or core signs of the presenting suicidal phenomena. This cannot happen unless research steps in with evidence so urgently required by clinicians, who cannot be blamed for ignoring the importance of insomnia as a prodromal/core/residual sign of suicidal phenomena in the absence of research evidence. For clinicians, suicidality is a multi-faceted phenomenon of which insomnia is just one facet..
The close temporal relationship between insomnia and suicidality seen in studies with PTSD and completed suicide samples, among adolescents and adults, needs to be examined urgently, so that research can aid the clinician in case management with scientific evidence. Despite the findings of all the available studies concurring on the direction and nature of this relationship, the number of studies found are too few to draw a clear conclusion. However, the findings of these available studies cannot be ignored, either by clinicians or by researchers. Although clinical significance is noted and managed by clinicians on their own, the establishment of statistical significance of this relationship will make case management more secure and safe, for the patient and the clinician.
The case studies presented here were able to inform us about several important aspects of clinical work; viz, the importance of normalising sleep patterns and the clinical benefits of doing so; the difficulty of clinical distillation of sleep disturbances from other behaviour disturbances; the role of stress-reduction, sleep hygiene, medication and psychotherapy in the clinical management of suicidal ideation and how sleep disturbances, specially insomnia might be enough reasons to keep the clinical relationship alive and active, with careful monitoring of sleep patterns by the clinician and family members.
The case studies reveal exactly how important it is to ask the patient or elicit from the patient the most disturbing aspect of personal distress so that the management plan can address it first and be vigilant for its re-occurrence. The clinician’s observations also inform us about the sequence of the symptoms: sleep disturbances may be a trait seen to develop from infancy itself or a state responding to negative environmental stimuli/stress. However, the clinician’s documentation of observations also raise the issue of exploring the relationship between continuity into adulthood of sleep disturbance traits established in early life and later suicidality. In other words, do children/adolescents who have insomnia continuing into adulthood also become suicidal later on? Is insomnia a predictor or an associate or both, of suicidality? Or is it a warning sign? This question begs a longitudinal research study design, whether epidemiological or clinical.
A relationship between insomnia and suicidality appears to be present as revealed in research findings and clinical case vignettes. The relationship is seen most clearly with suicide, as all the studies found the same temporal relationship between insomnia and suicide: immediate, ranging from last week to last 12 months, regardless of age. However, far too little or next to nothing is known about the nature of the relationship and the mechanisms of how this relationship works. Isolating a single symptom has its own shortcoming in a clinical setting where the reliability and adequacy of clinical history, the sequence of events, duration of each symptom and temporal relationship between occurrences of symptoms, are all important determinants. Far too often, underlying pathology gets known retrospectively, particularly as symptoms of depression may not be manifested early or may be hidden/masked. If markers of suicidal behaviour are isolated by researchers with a clinical background, clinicians can be better equipped to deal with suicidal patients.
We repeat the call for longitudinal studies examining the relationship between insomnia and suicidality among adolescents made by Liu and Buysse (2006), supporting this call with the specificity of prospective designs, more questions and hypotheses. We also reiterate the statement made by Donnelly (2003) that it is possible that reduction of clinical distress of such a disabling symptom as insomnia may have a ripple effect on the overall functioning of the individual, especially during adolescence and more so among suicidal adolescents.
