The history of generalized anxiety disorder as a diagnostic category

Marc-Antoine Crocq (2017) The history of generalized anxiety disorder as a diagnostic category, Dialogues in Clinical Neuroscience, 19:2, 107-116, DOI: 10.31887/DCNS.2017.19.2/macrocq

Google Wavenet text to speech

Abstract

From the 19th century into the 20th century, the terms used to diagnose generalized anxiety included “pantophobia” and “anxiety neurosis.” Such terms designated paroxysmal manifestations (panic attacks) as well as interparoxysmal phenomenology (the apprehensive mental state). Also, generalized anxiety was considered one of numerous symptoms of neurasthenia, a vaguely defined illness. Generalized anxiety disorder (GAD) appeared as a diagnostic category in the third edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-III) in 1980, when anxiety neurosis was split into GAD and panic disorder. The distinct responses these two disorders had to imipramine therapy was one reason to distinguish between the two. Since the revised DSM-III (DSM-III-R), worry about a number of life circumstances has been gradually emphasized as the distinctive symptom of GAD. Thus, a cognitive aspect of anxiety has become the core criterion of GAD. The validity of GAD as an independent category has been questioned from DSM-III up to preparation of DSM-5. Areas of concern have included the difficulty to establish clear boundaries between GAD and (i) personality dimensions, (ii) other anxiety-spectrum disorders, and (iii) nonbipolar depression. The National Institute of Mental Health has recently proposed the Research Domain Criteria (RDoC), a framework destined to facilitate biological research into the etiology of mental symptoms. Within the RDoC framework, generalized anxiety might be studied as a dimension denominated “anxious apprehension” that would typically fit into the research domain called “negative valence systems” and the more specific construct termed “potential threat.”

Precursors: panophobia, neurasthenia, and anxiety neurosis

The phenomenology of generalized anxiety disorder (GAD), ie, chronic free-floating anxiety accompanied with anxious apprehension or worry about many circumstances of daily life, has been described by several authors since the dawn of modern psychiatry in the late 18th century. Early authors reported that this type of chronic anxiety could culminate in paroxysmal attacks. Thus, GAD and panic attacks were lumped together under the same illness.

In this section, we will first discuss the term “pantophobia,” a medical term in common usage until the early 20th century; we will then mention “neurasthenia,” a diagnostic term whose extreme popularity was matched by its vagueness; we will finally turn to “anxiety neurosis,” the immediate precursor of GAD and panic disorder (PD).

Panophobia

The term panophobia also appeared under related forms, such as pantaphobia, pantophobia, or panphobia. According to the Oxford English Dictionary, pan(t)ophobia was first attested to in English in 1781. The etymology is the post-classical Latin pantŏphŏbŏs, from the Greek παντοφόβος (all-fearing), whose literal sense is “anxiety about everything.” The term “pantophobic” was first employed by the Latin-speaking physician Caelius Aurelianus (fifth century), who wrote about the etymologies of medical terms that he translated from Greek into Latin. He noted that some physicians spoke of “pantophobic” patients, who were allegedly afraid of everything.^1,2 A frequent archetype of pan(t)ophobia or “vain fear” (Latin: inanis metus) quoted by classical authors is a saying by Aristotle in the Nicomachean Ethics, written in 350 bc (Book 7): “the man who is by nature apt to fear everything, even the squeak of a mouse.”³

In 1917, Devaux and Logre⁴ argued that the correct denomination should be pantophobic and not panophobic; etymologically, the latter could be understood as the “fear of the god Pan.” Interestingly, the word “panic” is derived from Pan, the Greek god of shepherds and cattle; the adjective panikos (πανικός) (“of Pan”) was used to designate panic fright because noises heard in mountains and valleys were attributed to Pan.

Boissier de Sauvages (1706-1767) wrote the first significant French medical nosology. His position at the transition from classical to modern medicine is exemplified by the fact that his work was the last major medical textbook to be written in Latin and that it was soon followed by a posthumous French translation. Boissier de Sauvages subdivided mental disorders, called vesaniae, into four Orders: (i) Hallucinations; (ii) Morositates, including pica, bulimia, polydipsia, antipathia, nostalgia, panophobia, satyriasis, nymphomania, tarantism (ie, immoderate craving for dance), and hydrophobia; (iii) Deliria; and (iv) Folies anomales. The disorder mainly concerned with anxiety is panophobia.⁵ In Boissier de Sauvages' nosology,⁶ the first form of panophobia is little more than nocturnal terror. However, other subtypes of panophobia are reminiscent of modern anxiety disorders. In panophobia hysterica, also called “panic terror caused by vapors,” hysterical and hypochondriac subjects experience sudden fright and react dramatically with heart racing or pallor when startled by innocuous noises or sights. This was attributed to a diathesis of exacerbated sensibility. It was reported that these subjects may additionally present with the complicating symptoms of grief or worries. In panophobia phrontis (from the Greek φpοντίς: care, worry, preoccupation), also called worry (French: souci), the patients present with features evocative of GAD. These individuals are constantly extremely worried, and for this reason they avoid company, preferring to keep to themselves. They complain of pain and bodily tension.

According to the historian of psychiatry German E. Berrios,⁷ in 1902, Albert Pitres and Emmanuel Régis⁸ gave the best description of panophobia. Pitres (1848-1928) trained with Jean Martin Charcot, studied cortical localizations, and later became professor of psychiatry in Bordeaux. He wrote the book Les Obsessions et les Impulsions with Régis (1855-1918). Régis later wrote a textbook of psychiatry (Précis de Psychiatrie), whose successive editions were a reference in the early decades of the 20th century in France. Régis introduced in French many concepts from Kraepelin and Freud. The two authors described panophobia at a congress of psychiatry in Moscow in 1897.⁹ They quoted Théodule Ribot,¹⁰ who in turn had attributed the term panophobia to George Beard, the creator of neurasthenia. Pitres and Regis described a disorder of emotions, “a state of vague but permanent anxiety or terror that was termed panophobia or pantaphobia (Beard); it is a condition where the patient is afraid of everything, where anxiety, instead of being fixated to the same object, floats as in a dream and gets fixated only for an instant, according to random circumstances, passing from an object to the next. The most prominent symptom of the panophobic state is what Freud very rightly termed 'anxious expectation'.” For Pitres and Regis, panophobia and Freud's anxiety neurosis are clearly synonyms.

Neurasthenia

Beard's neurasthenia was a most successful diagnostic category. It is a very broad concept, whose meaning has evolved since Beard's original description in 1869 until its retention as a diagnostic concept (F48.0) in the tenth revision of the International Statistical Classification of Diseases and Related Health Problems (ICD-10), where it must be excluded for the diagnosis of GAD (F41.1). In Beard's treatise,¹¹ “pantaphobia” [sic] is one of the many manifestations of neurasthenia (“There is a manifestation of morbid fear which is not uncommon, and to which we might perhaps give the term pantaphobia, or fear of everything”).

In parallel with works on the phenomenology of GAD, clinical descriptions of PD were also published. An important figure in the history of PD is Edouard Brissaud,¹² who in 1899, identified “pure paroxystic anxiety” (anxiete paroxystique pure, p 348), specifying that the condition may at times evolve toward agoraphobia.

Anxiety neurosis

Sigmund Freud first used the term “anxiety neurosis” (angstneurose) in German in 1895 in his article entitled “On the Grounds for Detaching a Particular Syndrome From Neurasthenia Under the Description Anxiety Neurosis'.”¹³ Freud had also used the term “névrose d'angoisse” in a paper published in French in the same year.¹⁴ Freud's etiological theory—now outdated—postulated that anxiety neurosis was caused by an accumulation of sexual excitation that could not find discharge in coitus. Interestingly, Freud listed as second symptom in his clinical description of anxiety neurosis the “anxious expectation,”—a core criterion for GAD in DSM-5—for which he gave the following example: “A woman, for instance, who suffers from anxious expectation will think of influenza pneumonia every time her husband coughs when he has a cold, and, in her mind's eye, will see his funeral go past.” The symptomatology of anxiety neurosis according to Freud is much broader that that of modern GAD. Freud lists the following classes of symptoms: (i) general irritability; (ii) anxious expectation, which, as mentioned above, approximates the worry criterion of DSM-5; (iii) anxiety attacks; (iv) equivalents of anxiety attacks (eg, attacks of sweating, ravenous hunger, tremor, vertigo, paresthesias); (v) pavor nocturnus; ( vi) vertigo; (vii) typical phobias or agoraphobia; (viii) nausea, diarrhea (in contrast to constipation in neurasthenia); (ix) paresthesias, rheumatic pains (Freud adds that several individuals known as rheumatic are in reality suffering from anxiety neurosis—is this relevant to nowadays fibromyalgia?).

In the early 20th century, French authors—eg, Francis Heckel¹⁵ in his renowned treaty on anxiety neurosis—adopted Freud's term of anxiety neurosis but refuted the sexual etiology. Heckel's treatise was completed in 1914, but the onset of World War I delayed its publication until 1917. Heckel describes in detail the paroxysmal manifestations (“crises d'angoisse”) and the interparoxysmal symptomatology (the “mental state” of anxiety neurosis). He dismisses the validity of the psychoanalytical investigation method: “it is obvious that two psychoanalysts working with the same patient will delineate complexes and emotional syntheses that are radically different” (p 210).

Generalized anxiety disorder in modern psychiatric classifications

DSM-I and DSM-II

DSM-I ¹⁶ (known as DSM when it was published) had a chapter termed “psychoneurotic disorders.” Anxiety was supposed to be a danger signal perceived by the “conscious portion” of the personality and produced by a threat from within the personality; the various ways in which the patient attempted to handle this anxiety resulted in various types of “reactions.” The concept of reaction patterns, developed by Adolf Meyer, was at that time a hallmark of American psychiatry. The category called “anxiety reaction” was diagnosed when the anxiety was diffuse and neither restricted to situations or objects (as was the case in phobic reactions) nor controlled by any specific psychological “defense mechanism” (as was the case in dissociative, conversion, obsessive-compulsive reactions). Anxiety reaction was characterized by anxious expectation and frequently associated with somatic symptomatology; it was to be differentiated from normal apprehensiveness or fear.

DSM-II ¹⁷ renamed the reactions of DSM-I as “neuroses.” DSM-II stated that anxiety was the chief characteristic of “neuroses”; anxiety could be felt directly or controlled unconsciously by various symptoms. In DSM-II, the diagnostic category “anxiety neurosis” was characterized by “anxious over-concern extending to panic and frequently associated with somatic symptoms”; thus, it encompassed both future categories denominated panic attacks and GAD.

The advent of generalized anxiety disorder in DSM-III and subsequent revision of criteria in DSM-III-R and DSM-IV

The splitting of DSM-II anxiety neurosis into (i) GAD and (ii) PD in the DSM-III ¹⁸ in 1980 was the official birth date of GAD as a diagnostic category. In reality, as discussed by Rickels and Rynn,¹⁹ GAD was conceived of a few years earlier. Whereas the so-called Feighner Criteria²⁰ developed at Washington University in St Louis still grouped free-floating anxiety and anxiety attacks under the single diagnosis of anxiety neurosis, the Research Diagnostic Criteria—developed by Spitzer et al soon thereafter—separated GAD from PD.²¹ One major argument for the separation of spontaneous panic from anticipatory anxiety was the finding that they could be distinguished on the basis of response to medication, an approach later termed “pharmacological dissection.”²² Indeed, a double-blind placebo-controlled trial by Klein²³ had shown that patients treated with imipramine had fewer panic attacks even though their levels of anticipatory anxiety remained high. However, more recent studies have disproved this claim that PD is a specific indication for antidepressant therapy, whereas GAD is not responsive; Kahn and colleagues were the first to show results with imipramine that contradicted Klein's observations.²⁴ It is now recognized that antidepressants are effective in the treatment of GAD.²⁵

Because of the high degree of comorbidity with other disorders, the autonomy of DSM-III GAD was soon questioned, and its elimination was regularly debated with each new revision of DSM. GAD is known for showing high comorbidity with other anxiety disorders and with depression. The criteria of GAD have been modified in successive editions of DSM in order to bolster the validity of this category. Table I shows the evolution of the definition of GAD over succeeding editions ot DSM and ICD.¹⁹

^{Table 1 was here.}

A study reported in 1984 by Anderson and colleagues²⁶ found that subjects with GAD had fewer autonomic symptoms and an earlier, more gradual onset than patients with PD, a finding that tended to support the distinction between the two categories. An important study reported in 1986 by Barlow and colleagues²⁷ produced findings that anticipated the increased emphasis on worry in the criteria of GAD. Barlow et al commented that DSM-III GAD was largely a residual diagnosis among the spectrum of anxiety disorders because patients diagnosed with other DSM-III anxiety disorders almost always met the criteria for GAD also. Notably, two forms of GAD were suggested by Barlow et al: a residual form representing the apprehensive expectations that accompany other anxiety disorders and a qualitatively different form of chronic worry that represents an independent syndrome. Early clinical studies of DSM-III GAD found that GAD seldom occurred in the absence of major depressive disorder (MDD). However, this comorbidity weakened as duration of GAD increased; thus, the duration requirement for GAD was increased to 6 months in DSM-III-R and later, DSM-PV.²⁸

During the transition from DSM-III-R to DSM-IV,²⁹ it was also suggested that, often, patients with GAD experience excessive anxiety from an early age and with chronicity, and that GAD might be better conceptualized as a vulnerability that should be located on axis II of the DSM multiaxial system for assessment. Akiskal³⁰ has suggested that GAD might be best considered as an anxious temperament. A set of somatic symptoms associated with GAD that differs substantially from those for other anxiety disorders was determined. These findings led to a reduction in the number of items in the symptom criterion, from 18 in DSM-III-R to six in DSM -IV (see Table I). Another substantial revision in DSM-IV was greater emphasis on the uncontrollability of worry.

Generalized anxiety disorder in DSM-5

The high comorbidity between GAD and depressive disorders remained problematic during the preparation of DSM-5. This high comorbidity has been explained in various ways, including genetic pleiotropy,³¹ which means that GAD and nonbipolar depression might represent different phenotypic expression of a common etiology. Accordingly, it was proposed to merge GAD and nonbipolar depression in a spectrum of mood/anxiety disorders, by creating three subclasses of emotional disorders: (i) bipolar disorders; (ii) distress disorders (MDD, dysthymic disorder, GAD, and posttraumatic stress disorder [PTSD]); and (iii) fear disorders (PD, agoraphobia, social phobia, and specific phobia).³²

However, other data still argued for some distinctions between GAD and MDD. For instance, epidemiological surveys showed differences in risk factors for GAD and MDD, arguing against the view that the two disorders are merely different manifestations of a single underlying internalizing syndrome or that GAD is merely a prodromal, residual, or severity marker of major depressive episode.³³ Another proposal³⁴ was to relabel GAD as “generalized worry disorder” in DSM-5, to reflect that worry is the core feature and to capture the behavioral consequences of anxious apprehension and worry by adding a criterion C (“marked avoidance of potentially negative events or activities”; “marked time and effort preparing for possible negative outcomes of events or activities”; “marked procrastination in behavior or decision-making due to worries”; and “repeatedly seeking reassurance due to worries”.) One rationale for this criterion C is the supposition that worry acts as a cognitive coping strategy that manifests in avoidant behaviors.

The final version of DSM-5 did not differ much from DSM-IV,³⁵ GAD being defined by the following diagnostic criteria:

A. Excessive anxiety and worry (apprehensive expectation), occurring more days than not for at least 6 months, about a number of events or activities (such as work or school performance).

B. The individual finds it difficult to control the worry.

C. The anxiety and worry are associated with three (or more) of the following six symptoms:

1. Restlessness or feeling keyed up or on edge.

2. Being easily fatigued.

3. Difficulty concentrating or mind going blank.

4. Irritability.

5. Muscle tension.

6. Sleep disturbance (difficulty falling or staying asleep, or restless, unsatisfying sleep).

Criterion F in DSM-5 makes it clear that GAD is largely an exclusion diagnosis. GAD cannot be diagnosed if the anxiety is better explained by other anxiety disorders (panic, phobic, social anxiety, or obsessivecompulsive disorder). Also, GAD cannot be caused directly by stressors or trauma, contrary to adjustment disorders and PTSD.

Generalized anxiety disorder in ICD-10 and ICD-11

In ICD-10, GAD (F41.1) includes anxiety neurosis, anxiety reaction, and anxiety state, but excludes neurasthenia. ICD-10 also proposes diagnostic criteria for research: (i) at least 6 months with prominent tension, worry, and feelings of apprehension about everyday events and problems; and (ii) at least four symptoms out of a list of 22 items, of which at least one item is from a list of four items of autonomic arousal (palpitations/accelerated heart rate, sweating, trembling/shaking, dry mouth).

The ICD-11 Working Group decided to retain a categorical approach to the anxiety disorders. GAD will be included in the category of “anxiety and fear-related disorders,” distinguished from “obsessive-compulsive or related disorders” and “disorders specifically associated with stress.”³⁶ In the draft of ICD-11,³⁷ GAD is redefined: worry about multiple areas of everyday life has been introduced as an alternative essential feature to generalized apprehension. GAD can now co-occur with other mental and behavioral disorders.

Overview of the evolution of the concept of GAD since the second half of the 20th century and future perspectives

Prevalence, sex ratio, and the importance of worry

According to DSM-III and DSM-III-R, GAD had a low prevalence and rarely caused more than mild impairment. Starting with DSM-IV, it was required that GAD cause clinically significant distress or impairment; this is because this requirement became part of the definition of a disorder in DSM-IV. Also in DSM-IV, GAD morphed from a rarely diagnosed condition into a disorder with a lifetime prevalence reaching up to 5% in a community sample. Also, the sex ratio changed dramatically. According to DSM-5, GAD is about twice as likely in females as in males, whereas in DSM-III it was supposed to be equally common in both sexes.

However, the most important evolution is that since DSM-III-R, general apprehensiveness or worry that are not restricted to any particular stimulus has been progressively established as the distinctive and core symptom of GAD. Worry about everyday matters is a relatively specific symptom, contrary to somatic symptoms of GAD, which are also found in other disorders. Worry is translated as “souci” and “sorge” in the French³⁸ and German³⁹ editions of DSM-5, respectively. The concept of worry puts the emphasis on the psychological symptomatology, specifically on the cognitive functioning. Thus, an anxious condition that, earlier, might have been considered as related to noncortical emotions and corporeal sensations is now grounded on cognitions based in the prefrontal cortex. The emphasis on the cognitive aspect of anxiety had already been made much earlier. Edouard Brissaud (1853-1909)⁴⁰ distinguished “angoisse” (anguish), the sensation of somatic- distress, and “anxiete” (anxiety), the intellectual processing of anguish; this distinction still exists to some extent with corresponding word pairs in other romance languages. Worry may provide short-term relief by means of the avoidance of threatening imagery.⁴¹ Another important cognitive aspect of GAD is intolerance of uncertainty.⁴² The long-term consequences of worry include the inhibition of emotional processing and the maintenance of anxiogenic conditions.⁴³ One effect of worrying may be to suppress somatic (eg, cardiovascular) symptoms of anxiety.^44,45 Over successive editions of DSM, worry displaced somatic symptoms, although the latter are still important in transcultural psychiatry (see the article by Khambaty and Parikh in this journal issue). Research has provided evidence for the dimensionality of worry. Thus, GAD, whose central feature is worry, may be quantitatively rather than qualitatively different from normal functioning. Ruscio et al⁴⁶ argued that a focus on normal and pathological extremes has constrained the study of worry phenomena and that dimensional conceptualization of worry may significantly enhance understanding of both worry and GAD. The debate between dimensional and categorical approaches is relevant to the Research Domain Criteria (RDoC) initiative.

Future perspectives—RDoC

The National Institute of Mental Health initiated the RDoC project,⁴⁷ a research framework designed for studying the full range of human behavior from normal to abnormal based on multiple levels of information (genes, molecules, cells, brain circuits, physiology, behavior, and self-report). Currently, the RDoC framework is organized around five basic domains of functioning: negative valence systems, positive valence systems, cognitive systems, social processes, and arousal and regulatory systems. Logically, anxiety might be explored within the negative valence system.^48,49 The negative valence system contains several constructs, one being “potential threat (anxiety)” described as activation of a brain system in which harm may potentially occur but is distant, ambiguous, or low/uncertain in probability, characterized by a pattern of responses such as enhanced risk assessment (vigilance). The construct of “anxiety” is distinguished from other constructs within the negative valence systems, namely “acute threat,” “sustained threat,” “loss,” and “frustrative reward.” The RDoC encourages the study of clinical dimensions rather than DSM clinical categories. Anxious apprehension is a dimension that could be more validly studied within the RDoC framework than a diagnostic category such as GAD. Anxious apprehension is defined by a persistent pattern of negative repetitive thinking about perceived threats. It might involve left-hemisphere activity, notably the left inferior frontal gyrus.⁵⁰ Sharp et al⁵¹ analyze how the RDoC matrix might allow a transdiagnostic dimension such as anxious apprehension to be studied in several research domains. While it could belong under the cognitive systems domain of RDoC—since it involves negative repetitive thinking and a breakdown in executive flexibility—the most obvious place for anxious apprehension in the RDoC matrix lies under the construct “potential threat” within the domain of negative valence.