Sleep, suicide behaviors, and the protective role of sleep medicine

Highlights

• Insomnia is associated with suicide attempt among veterans.

• Sleep medicine visits reduce risk of suicide attempt in sleep disordered patients.

• Treatment of sleep disorders may be a strategy to augment suicide prevention efforts.

Abstract

Objective/Background

Sleep disturbance is associated with suicidal thoughts and behaviors. The relationship of specific sleep disorders to suicide attempts is less well established. Whether treating sleep disorders reduces suicide attempts remains controversial.

Methods

Suicide attempts, treatment utilization, and psychiatric diagnoses were extracted from electronic medical records and a suicide attempt database from the U.S. Department of Veterans Affairs. The sample (N = 60,102) consisted of patients with any record of suicide attempt in FY13-14 and a 1:1 case-control of patients with no record of attempt, who were propensity score-matched based on age, gender, and prior year mental health treatment utilization. Associations among sleep disorders and suicide attempt were examined via logistic regression. Covariates included depression, anxiety, posttraumatic stress disorder (PTSD), bipolar, schizophrenia, substance use disorder (SUD), medical comorbidity, and obesity.

Results

Insomnia (OR = 5.62; 95% CI, 5.39–5.86), nightmares (odds ratio, OR = 2.49; 95% confidence interval, CI, 2.23–2.77), and sleep-related breathing disorders (OR = 1.37; 95% CI, 1.27–1.48) were positively associated with suicide attempt after accounting for age, gender, treatment utilization, and comorbid sleep disorders. Furthermore, when controlling for depression, anxiety, PTSD, bipolar, schizophrenia, substance use disorder (SUD), medical comorbidity, and obesity, insomnia (OR = 1.51, 95% CI, 1.43–1.59) remained positively associated with suicide attempt nightmares (OR = 0.96; 95% CI, 0.85–1.09) nor sleep-related breathing disorders (OR = 0.87, 95% CI = 0.79–0.94). Additionally, sleep medicine visits 180 days prior to index date were associated with decreased likelihood of suicide attempt for individuals with sleep disorders (OR = 0.86; 95% CI, 0.79–0.94).

Conclusion

Insomnia is associated with suicide attempt among veterans. Sleep medicine visits were associated with a reduced risk of suicide attempt in sleep disordered patients. The assessment and treatment of sleep disorders should be considered in context of strategies to augment suicide prevention efforts.

Introduction

Suicide is both insidious and ubiquitous, touching most segments of the population. In the United States, suicide sits conspicuously as the tenth most frequent cause of death [1]. Perhaps nowhere is the suicide rate more alarming than among military veterans, who after adjusting for age and gender have an approximately 1.5 times greater risk for suicide as compared to the civilian population [2]. While it is important to intervene with individuals following a non-fatal suicide attempt, the majority of first attempts are fatal [3]. Identifying populations at risk for suicide prior to a first attempt is an important, but difficult task of suicide prevention. Prevention efforts can be aimed at modifiable risk factors that arise early on a patient's trajectory toward a suicide attempt.

Sleep disturbance is positively associated with suicidal ideation (SI), suicide attempt (SA) [4], and suicide deaths [5]. For example, in a 2012 meta-analysis of 39 studies the presence of sleep disturbance was linked to increased SI (relative risk ratio = 2.95; 95% confidence interval [CI], 2.48–3.50), suicide attempt (3.13; 95% CI, 2.38–4.13), and suicide (1.95; 95% CI, 1.41–2.69) [5]. Sleep disturbance has also been linked to time to death among veteran suicide decedents such that following their final Veterans Health Administration (VHA) contact, veterans with a documented sleep disturbance had a 57% loss in survival time after controlling for psychiatric and substance use symptoms [6].

The sleep-suicide relationship is of particular importance given that sleep disorders are treatable, cut across patient populations, and are associated with the development and/or exacerbation of myriad medical and psychiatric conditions [7]. There are multiple pathways through which sleep may impact suicide risk. First, insomnia has been linked to the new onset of mental illness [8] including the development and exacerbation of depressive disorders, as well as a blunted response to depression treatments [9], [10], [11]. Second, the impact that sleep disturbance has on cognition and mood is posited to reduce one's ability to effectively cope with stressors, resulting in a poor response to stress and increased frequency of negative life events and/or decreased interpersonal functioning. Third, feelings of helplessness and hopelessness may also be influenced by the sense of loss of control sometimes associated with insomnia [12]. These pathways fit well with generalized sleep disturbance and insomnia. However, sleep is multidimensional, and insomnia, sleep-related breathing disorders, and nightmares are etiologically distinct. Therefore, research examining associations among suicide risk and more specific forms of sleep pathology are indicated.

While the preponderance of the literature focuses on insomnia, a small number of studies have also suggested relationships among suicide risk, nightmares [13], [14], [15], [16], and sleep apnea [17], [18], [19], [20]. Choi and colleagues [18] provided one of the first estimates of the prevalence of SI among individuals with untreated sleep apnea. The authors reported that among 117 subjects, 20.5% (n = 24) endorsed SI based on a single self-report item. Moreover, Edwards and colleagues [19] reported that among 426 patients assessed for sleep apnea, depression severity was associated with more severe baseline sleep apnea. Significant decreases in depression and apnea severity were observed following three-months of continuous positive airway pressure (CPAP) among treatment adherent participants (n = 228). In addition, among a subsample who reported baseline SI (n = 41), none endorsed SI at a three-month follow up. Most recently, findings from a national sample (N = 40,149) suggest significant associations among sleep apnea and past-year suicidal thought [17]. Individuals reporting that a physician had diagnosed them with sleep apnea were more likely to endorse SI (odds ratio [OR] = 1.50; 95% CI, 1.18–1.91) and suicide planning (OR = 1.56; 95% CI, 1.08–2.26) after controlling for key covariates (eg, depression, cerebrovascular disease, alcohol use disorder). Alternately, in this sample apnea had had not been significantly associated with suicide attempts (OR = 1.22; 95% CI, 0.66–2.26). A key limitation of these findings is that both the apnea and suicide variables were based on cross-sectional, self-report data.

Relationships among nightmares and suicidal thought and behavior are in many ways even more difficult to disentangle given that they are frequently observed in the context, or are part of the symptomatology, of mental health disorders such as posttraumatic stress disorder (PTSD). Further obscuring the impact of nightmares on suicide risk are reports that only a third of individuals experiencing nightmares report them to their healthcare provider, with many believing that their nightmares were not treatable [21], suggesting that they are under documented in the medical record. Extant research, however, has shown nightmares to be related to suicidal ideation among both college students [13], [22] and individuals with a prior attempt history [23], [24] while controlling comorbid psychopathology (eg, depression, anxiety, PTSD), further supporting the premise that sleep problems are associated with a range of suicidal thought and behavior.

There is also evidence demonstrating that insomnia interventions can reduce symptom severity in the context of co-occurring mental health conditions, particularly depression [25], [26] and PTSD [27]. Further, findings from a small number of studies suggest that cognitive behavioral therapy for insomnia may also lead to reductions in suicidal ideation [25], [26], [28]. We are unaware of any work that has examined the impact of sleep medicine clinic visits on the risk of suicide attempts. Therefore, while sleep represents a modifiable treatment target, it remains unclear what effect, if any, treating sleep has on suicide behavior. In addition, it has yet to be established how, and at what magnitude, etiologically distinct groups of sleep disorders (eg, nightmare, insomnia, and sleep-related breathing disorders, SRBD) are associated with suicide attempts. The research to date has focused on either generalized sleep disturbance (ie, sleep problems not rooted in a specific diagnosis) or the construct of insomnia. Further, the bulk of the sleep-suicide literature has relied on subjective measures to identify both problems with sleep and suicide risk [4], [29], [30], [31].

The present study addresses some of these knowledge gaps by utilizing objective markers for both sleep and suicide in a large sample. We provide risk estimates associated with specific sleep disorders as well as data examining the protective role of sleep medicine clinic visits.