Safety behaviors are coping tactics used by persons with anxiety disorders to prevent, escape from, or diminish the severity of a perceived threat. Safety behaviors are particularly common in persons with panic disorder, agoraphobia, and social anxiety disorder. These behaviors emerge in response to external (e.g., situations, persons, activities) and/or internal (e.g., thoughts, emotions, memories) foci of perceived threat and are either anticipatory (avoidant) or consequential (escape) [1]. While safety behaviors may provide some measure of immediate relief, they have no impact on the individual's propensity for anxiety, nor do they diminish the frequency and severity of future symptomatic events. However, assessment of safety behavior provides guidance for implementing targeted interventions, and accurate assessment and elimination of safety behaviors is often necessary to maximize treatment of clinical anxiety [351].

Another coping mechanism observed in persons with anxiety disorder is safety signals, defined as the people or objects used to diminish distress in situations that elicit anxiety. Safety signals are also potentially counterproductive; they offer immediate relief but facilitate persistence of the anxiety disorder over time by preventing direct confrontation of feared stimuli in the absence of "safe" objects/people and by maintaining perceptions of risk/harm and coping inability. A patient's continued use of safety signals impedes therapeutic progress, in particular the response to exposure therapy. However, safety behaviors may be helpful early in treatment by making exposure therapy more tolerable and less threatening [1].

Anxiety disorders with earlier median age of onset are phobias and separation anxiety disorder (15 to 17 years of age), and those with latest age of onset are panic disorder and generalized anxiety disorder (23 to 30 years of age). Lifetime morbid risk is considerably higher than lifetime prevalence for most anxiety disorders, with magnitude of difference much higher for disorders with later than earlier age of onset. Also, the ratio of 12-month to lifetime prevalence roughly reflects persistence but varies meaningfully in ways consistent with differential persistence of these disorders [7].

The odds for a lifetime diagnosis of any anxiety disorder were calculated, and the same pattern was found for past 12 month diagnosis [8]. These odds are organized according to sex, socioeconomic status, education level, and age. Overall, the risk of developing an anxiety disorder is greater for women/girls than men/boys.

In one study, children with SEPAD were 3.5 times more likely to later develop panic disorder and more than twice as likely to develop any anxiety disorder but did not significantly differ in later development of depression or substance use disorder. These findings were considered supportive of a developmental psychopathology model of anxiety disorders [65].

Altered limbic and prefrontal cortex functioning characterize anxiety disorders, with amygdala hyper-responsivity to threatening stimuli and impaired ventromedial prefrontal contex inhibitory control over limbic-generated, anxiety-inducing signals, associated with aberrant communication and functional connectivity between the amygdala and the prefrontal cortex [67].

In patients with panic disorder/agoraphobia, fear conditioning has shown enhanced activation of the bilateral dorsal inferior frontal gyrus. Simple conditioning, safety signal processing, and anxiety sensitivity correlate with the extent of midbrain activation. These findings suggest alterations in "top-down" and "bottom-up" processes during fear conditioning, interpreted within a neural framework of defensive reactions that mediate threat through distal (forebrain) versus proximal (midbrain) brain structures. This network may play a key role in panic disorder pathogenesis [83].

With agoraphobia, specific phobia, and SAD, the requirement that patients recognize their anxiety as excessive or unreasonable has been eliminated. This change was based on evidence that individuals with such disorders often overestimated the danger in "phobic" situations and that older individuals often misattributed "phobic" fears to aging. Instead, the anxiety must be out of proportion to the actual situational danger or threat, with consideration of cultural contextual factors [102].

The DSM-5 (and previous DSM editions) has been criticized for emphasis on reliability at the expense of diagnostic validity and for use of symptom-based diagnosis when symptoms alone may not best inform treatment selection. In response, the National Institute of Mental Health is developing the Research Domain Criteria, a new taxonomy for mental disorders that draws from genetics, neuroscience, and behavioral science [103]. Additionally, the DSM-5-TR, which was released in March 2022, includes the addition of prolonged grief disorder; the inclusion of symptom codes for suicidal behavior and nonsuicidal self-injury; refinement of criteria; and comprehensive literature-based updates to the text [104].

GAD is characterized by excessive and inappropriate worrying that is persistent and not restricted to circumstance or situation. Patients have physical anxiety symptoms and key psychologic symptoms. GAD is often comorbid with major depressive disorder, panic disorder, phobia, health anxiety, and OCD [3]. The DSM-5 diagnostic criteria for GAD remain unchanged from previous editions [2,102]:

Agoraphobia is defined as the fear of panic attacks occurring in places or situations from which escape might be difficult or embarrassing or where help may not be available. These situations can include crowds, going outside the home, or using public transportation and are either avoided or endured with significant personal distress [3]. Agoraphobia can become severely disabling, and more than 33% of patients diagnosed with agoraphobia cannot endure leaving their home environment. Roughly 66% of patients with panic disorder develop agoraphobia [2].

Patients with SAD highly inflate perceived social costs from committing hypothetical blunders. Accounting for much of this social cost inflation are concerns about revealing self-flaws and, in particular, concerns over appearing socially incompetent [117].

SEPAD is characterized by fear or anxiety concerning separation from those to whom an individual is attached. Common features include excessive distress when experiencing or anticipating separation from home, and persistent and excessive worries about potential harms to attachment figures or untoward events that might result in separation [3].

The GAD-7 scale is a brief, validated screening tool specifically designed to identify anxiety disorders. The patient with symptoms is asked to rate severity and frequency of occurrence over the previous two weeks. Each item is rated on a four-point scale (0 to 3), yielding a maximum score of 21. A score of 10 or greater is the cut-off point for diagnosis of anxiety disorder (sensitivity 87%; specificity 82%) [135]. The degree of distress/functional impairment varies directly with the total score. The GAD-7 scale is available online at https://adaa.org/sites/default/files/GAD-7_Anxiety-updated_0.pdf.

A diagnosis of SAD should rule out avoidant personality disorder. Some symptoms of avoidant personality disorder resemble SAD, such as a pervasive pattern of social inhibition, feelings of inadequacy, and hypersensitivity to negative evaluation. However, avoidant personality disorder is distinguished by non-social avoidance that extends to novel situations and positive affect. Roughly 36% of patients with SAD are comorbid for avoidant personality disorder, and some believe avoidant personality disorder is a more severe variant of SAD [130,131].

It is important to rule out medication side effects as the underlying cause of anxiety by obtaining a complete list of currently used prescribed, over-the-counter, and herbal medications. Examples of common medications with anxiety side effects are asthma medications (e.g., albuterol, theophylline), herbal medicines (St. John's wort, ginseng, ma huang), corticosteroids, and antidepressants [2].

Advances in anxiety disorder neuroscience have increasingly pointed to the necessary role of fear extinction learning (through exposure therapy) in addressing underlying pathophysiology. While efficacy is shown with CBT and exposure, patients can have difficulty with the demanding and exhausting therapy process, and many who do manage to complete therapy respond partially and relapse with time. Efforts to improve CBT/exposure outcomes have led to the investigation of augmenting agents. In contrast to standard anti-anxiety drugs, these agents are not anxiolytic but are used to promote and accelerate long-term adaptive changes in brain function initiated by successful exposures [137].

Antidepressants are generally recommended as first-line therapy for panic disorder because, unlike benzodiazepines, antidepressants treat comorbid depression and lack abuse risk and potential side effects of excessive sedation, cognitive impairment, and ataxia. All major antidepressant classes are comparably effective, but SSRIs and, increasingly, SNRIs are recommended over TCAs and MAOIs due to better safety and tolerability [172].

It is now recognized that SSRIs differ in pharmacologic effect, including activity beyond SERT inhibition [179,180]:

Benzodiazepine treatment of anxiety disorders is controversial. While effective in rapid anxiety reduction, the potential drawbacks with long-term use are substantial. These agents are indicated when potent, short-term anxiolytic effects are necessary to permit infrequent exposure to feared stimuli and potentially severe anxiety, such as airplane travel [121,129,136]. Clonazepam, lorazepam, and alprazolam are effective for short-term use in panic disorder, GAD, and SAD, but ineffective for, and potentially worsening, comorbid depression [28]. The rapid anxiolytic effects make benzodiazepines highly appealing to patients with anxiety, but aside from this specific context, benzodiazepine prescribing for as-needed use is discouraged [136,225,226]. Benzodiazepines can reinforce pill taking, serve as a safety signal that undermines self-efficacy, and become incorporated into conditioned fear responses; these concerns are heightened with as-needed use. On-demand dosing links pill taking to rapid anxiety reduction, powerfully reinforcing avoidance in anxiety-provoking situations and encouraging longer-term reliance on the drug. This iatrogenic effect also contributes to poor CBT response.

The current recommended prescribing is for time-dependent use, instead of panic response-dependent use, to minimize the risks [121]. This would also seem to maximize risk of withdrawal syndrome from uninterrupted versus intermittent drug exposure.

Benzodiazepines are also useful in the initial weeks of SSRI/SNRI initiation, to rapidly reduce anxiety and possible early anxiogenic medication side effects before the onset of SSRI/SNRI anxiolytic effects [121,129,136]. However, patients may discontinue the antidepressant when co-prescribed a rapidly effective benzodiazepine, believing the benzodiazepine's symptom relief makes the SSRI/SNRI unneeded. Supportive therapy with regular visits or phone contacts may also help patients remain adherent until the delayed onset of antidepressant benefits appears or early antidepressant side effects lessen [227].

In some patients with anxiety disorder, and especially those with depressive comorbidity, consideration of overdose fatality is necessary when deciding on therapy options. Lethality rates from intentional overdose involving single medication ingestions of agents commonly prescribed for depression and anxiety were calculated from data obtained during the period 2000–2014 [252]. TCAs remain the most lethal antidepressants, with fatality rates from 31–142 per 10,000 ingestions. Amitriptyline is associated with the greatest lethality in overdose and alone accounted for 37% of all deaths from antidepressants [252]. Lithium, bupropion, venlafaxine, quetiapine, and valproic acid had rates of 6.9–12.0 per 10,000 ingestions, while citalopram and fluvoxamine had rates of 3.9–4.1 per 10,000. Fluoxetine, sertraline, paroxetine, and escitalopram had the lowest overdose lethality rates at 0.79–1.34 per 10,000 ingestions [252].

Several options as second-line agents have efficacy in GAD comparable to first-line agents but possess potential side effects or other risks that preclude first-line use [120]. Benzodiazepines would be considered first in most cases, except where there is a risk of substance abuse, while bupropion XL would likely be reserved for later. Quetiapine XR remains a good choice in terms of efficacy, but given the metabolic concerns associated with this atypical antipsychotic, it should be reserved for patients who lack response or cannot tolerate antidepressants or benzodiazepines [120]. It is important to note that drugs such as beta-blockers (e.g., propranolol) prescribed to address the physical symptoms of anxiety are ineffective in the treatment of GAD [243].

The first-line drugs recommended for the treatment of panic disorder are SSRIs or venlafaxine XR [120]. Research suggests that the largest effect size is found with clonazepam, followed by venlafaxine and fluoxetine [224]. Despite a sizeable number of pharmacologic options, less than 50% of patients with panic disorder experience full and sustained remission to first-line medication therapy [287].

CBT is the criterion-standard psychologic treatment for SAD. Cognitive restructuring techniques are used to challenge assumptions and maladaptive thoughts (i.e., "I will embarrass myself"), while the behavioral component centers on patient exposure to anxiety-provoking situations. Over time and with repeated exposure sessions, the goal is to reduce anxiety symptoms and to demonstrate that the situational anxiety is tolerable and without substantial risk. The efficacy of CBT is supported by many randomized controlled trials, with outcomes that vary but are typically similar to pharmacotherapy. Some reports suggest that, after treatment discontinuation, gains achieved with CBT may persist longer than those achieved with pharmacotherapy. CBT for SAD can be administered in group or individual formats. Although some studies have reported that individual CBT is superior to group CBT, meta-analyses have failed to find significant differences in efficacy between the two modalities. There is evidence to support the effectiveness of exposure therapy alone, but efficacy compared with CBT is equivocal [120].

Exposure-based therapies are the treatments of choice and show a high degree of successful remission for all phobias. In vivo exposure and virtual reality exposure to phobic stimuli or situations are both effective, with in vivo exposure superior to imaginal and virtual reality exposure at post-treatment but not at follow-up [307]. The effectiveness of exposure-based therapy is enhanced when exposure sessions are grouped closely together; when exposure is prolonged, real (not imagined), and provided in different settings; and when there is some degree of therapist involvement instead of being entirely self-directed. A greater number of sessions have been shown to predict more favorable outcomes. There is no evidence that flooding is more effective, and patients usually find graded, progressive exposures more tolerable [307].

Several randomized controlled trials have been conducted using placebo or active controls. In 21 patients with panic disorder with or without agoraphobia given inositol 12 g/day or placebo for four weeks, significant decreases were found in the frequency and severity of panic attacks and agoraphobia severity with inositol relative to placebo [335]. Another randomized controlled trial compared myo-inositol 18 g/day with fluvoxamine 150 mg/day in patients with panic disorder with or without agoraphobia. Both drugs led to significant but comparable improvements in anxiety symptoms/severity, agoraphobia symptoms/severity, and global impression. In the first month, reduction in the number of panic attacks per week was significantly greater with inositol than fluvoxamine (4.0 vs. 2.4). Nausea and fatigue were significantly more common with fluvoxamine [336].

Other randomized controlled trials found that myo-inositol 12 g/day in 28 patients with major depression significantly reduced Hamilton Depression Rating Scale scores (vs. placebo) after four weeks; and 18 g/day given to patients with OCD for six weeks led to significant reductions in OCD symptom scores (vs. placebo) [335]. A review of supplements and herbal therapies with purported anxiolytic efficacy concluded myo-inositol was one of very few with demonstrated effectiveness [337]. The published research needs larger trials but is intriguing in light of a study in which patients with severe depression receiving treatment with rTMS showed significantly elevated prefrontal cortex myo-inositol levels, and this elevation correlated with extent of clinical improvement [338]. A meta-analysis of inositol for depression and anxiety disorders found that the agent may be particularly beneficial for treatment of depression [339].

Yoga is an ancient mind/body practice that involves different techniques in physical postures, controlled breathing, deep relaxation, and meditation that have positive and specific influences. Research on yoga has demonstrated significant improvements in emotional self-regulation with consequent reductions in depression, stress, and anxiety levels and improvements in mood, quality of life, and well-being [346]. Several studies have found significant anxiolytic effects with yoga in patients with GAD or panic disorder, and it is considered the complementary therapy with strongest evidence of safety and efficacy in anxiety disorders.