Could improved recognition and treatment of anxiety disorders lead to a reduction in suicide?

Abstract

Suicide is a significant public health issue. Annually, approximately 800,000—1,000,000 people die by suicide: for each one, there are at least 20 others who attempt it. Traditionally, anxiety disorders were considered to be of little importance in suicide. Data indicating that patients with anxiety disorders can die by suicide may surprise clinicians even today—and certainly did during the 1980s when the first relevant reports emerged. This may be because anxious patients often express concerns about their physical health, or—especially in patients with panic attacks—fear imminent death and strive to avoid it through visiting emergency outpatient clinics, persistently seeking reassurance from medical professionals, and/or adopting dysfunctional safety-seeking behaviors (De La Vega et al., 2018).

Khan et al. (2002), using the United States Food and Drug Administration database, assessed suicide risk among patients participating in clinical trials evaluating novel medications for anxiety disorders, obsessive-compulsive disorder and post-traumatic disorder. They found that suicide risk among patients was high regardless of the type of disorder, and overall was higher than in the general population, by a factor of 10 or more. They considered this finding unexpected, especially since patients participating in clinical trials for new medications are selected strictly, so that suicide risk is anticipated to be minimal. Kanwar et al. (2013) reviewed 42 observational studies involving 309,974 patients with anxiety and anxiety-related disorders. Compared to those without anxiety, anxious patients were more likely to have suicidal ideation (OR = 2.89, 95% CI: 2.09, 4.00), attempted suicide (OR = 2.47, 95% CI: 1.96, 3.10), complete suicide (OR = 3.34, 95% CI: 2.13, 5.25), or have any suicidal behaviors (OR = 2.85, 95% CI: 2.35, 3.46): all anxiety disorders, but not obsessive-compulsive disorder, were associated with increased suicide risk.

However, regarding obsessive-compulsive disorder (OCD), more recent analyses show that around one out of every 10 patients attempts suicide during their life; about one-third have current suicidal ideation; and about half have had suicidal ideation in the past. Concerning adult separation anxiety disorder, it was significantly more frequent among outpatients with other anxiety or mood disorders who had suicidal thoughts compared to those without such thoughts (Pini et al., 2021). In ‘psychological autopsy’ studies on deaths by suicide, generalized anxiety disorder (GAD) was present three times more frequently than any other anxiety disorder. Furthermore, in subjects without suicide ideation during the last 12 months at baseline, only lifetime history of generalized anxiety disorder or depression significantly predicted the emergence of suicide ideation within a 6-year follow up. Compared to the lifetime prevalence of suicidal ideation (13.5%) and suicide attempt (4.6%) in the US general population, the overall prevalence of suicidal ideation and attempts in patients with panic disorder was 23% and 17%, respectively. In primary care settings, up to 16% of patients with social anxiety disorder reported suicidal ideation in the previous month, and 18% had a history of suicide attempts. In adolescents particularly, the relationship between separation anxiety disorder and suicidal ideation is mediated by feelings of loneliness and appears independent of depressive symptoms. Military personnel with post-traumatic stress disorder (PTSD), compared to those without, demonstrate an up to 5.4-fold increase of suicidal ideation, it being considered that stress relating to reintegration into civilian society contributes to this increase (Haller et al., 2016).

Older (≥60 years old) adults with anxiety disorders also have increased suicide risk: one third reported feeling like they would be better off dead. Among older adults with anxiety disorders, socially anxious individuals were more prone to develop suicidal ideation, so this subgroup needs special attention by clinicians (Petkus et al., 2018).

Controversy had existed over whether anxiety disorders are independently associated (i.e. after adjusting for comorbid mental disorders) with suicidality manifestations. However, there is clear evidence, both from cross-sectional and longitudinal data, that a pre-existing anxiety disorder is an independent risk factor for subsequent onset of suicidal ideation and attempts (Sareen, 2011). Comorbid anxiety disorders also amplify the risk for suicide attempts in persons with mood disorders.

Another uncertainty has been the potential association between anxiety and suicidality in non-clinical samples. Thus among 119,875 college students, a significant proportion reported moderate (17%) or severe (13%) anxiety, while over the prior year more than 14% reported suicidal ideation and over 1% reported a suicide attempt (Moskow et al., 2022). Screening only for depression would pick up to 23% of students with suicidal ideation, this increasing to 35% when also screening for anxiety. Those with ‘anxiety and no-to-minimal depression’ had the second highest likelihood of suicide attempt, following those with ‘anxiety and depression’. Students with co-occurring ‘depression and anxiety’ had approximately double the rates of suicide attempts of those with ‘depression alone’.

Potential clinical and biological correlates of increased suicidality in anxiety disorders have been reported. Anxious patients, especially treatment-resistant ones, may hold maladaptive beliefs as to the futility of therapeutic efforts and feelings of hopelessness and despair which may culminate in suicide. Avoidance behaviors—a hallmark of anxiety disorders—may have a deleterious effect on social relationships and often result in social isolation and loneliness which are important risk factors for suicide (Herres et al., 2019). Patients with panic disorder who had attempted suicide showed alterations in the white matter of the internal capsule and thalamic radiations (Kim et al., 2015). Suicidal patients with panic disorder had lower serum total cholesterol and LDL levels than both normal controls and non-suicidal PD patients (Ozer et al., 2004); such an association was not found however in GAD patients (De Berardis et al., 2017). Suicide attempters with anxiety disorders showed significantly higher levels of C-reactive protein and interleukin-6 compared to non-suicidal ones (Wiebenga et al., 2022). Evidence is accumulating for a link between chronic stress-induced neuroendocrine dysregulation and vulnerability to suicide (O’Connor et al., 2020).

Clinicians were initially skeptical concerning the association between suicidality and anxiety disorders. However, we now know that up to 11.5% of suicide attempts are attributable to anxiety disorders, irrespective of other risk factors for suicide. This is especially important as anxiety disorders often go undiagnosed and untreated or are considered ‘minor’ disorders—particularly when comorbid with ‘major’ ones, for example, schizophrenia—and receive inadequate treatment. Current data make it imperative for clinicians to be alert for signs of increased suicide risk in patients with major depression, since at least 45% of those who were depressed at the time of their suicide had some contact with a primary healthcare provider within the 4 weeks preceding their death (O’Rourke et al., 2023). This is particularly important when considering that effective pharmacological and psychological treatments of depression reduce suicide risk (Zalsman et al., 2016). Likewise, clinicians should be vigilant to assess for suicidality in individuals with anxiety disorders, OCD and PTSD—both at initial evaluation and at follow-up visits—and manage these disorders promptly. Lastly, assessing for specific symptoms of anxiety, even in non-clinical samples, may increase accuracy in identifying suicide risk.

Vasilios G. Masdrakis and Sofia Konstantopoulou report no conflict of interest. David S. Baldwin is Editor-in-Chief of Human Psychopharmacology.