39 | DIAGNOSIS OF ANXIETY DISORDERS

MEGHAN E. KEOUGH, MURRAY B. STEIN, AND PETER P. ROY-BYRNE

THE CURRENT NOSOLOGICAL MODEL

Over the past 60 years, the conceptualization and diagnosis of psychiatric disorders has experienced substantial growth and transformation. Unlike the recent DSMs (Diagnostic and Statistical Manuals), the first two editions in 1952 and 1968 did not provide explicit categorical criteria sets and were rooted in the psychodynamic perspective that psychiatric symptoms reflect disguised psychological conflicts rather than disease states (American Psychiatric Association, 1952, 1968). The lack of standardized classification resulted in professionals operating largely based on their own personal beliefs and training, which resulted in substantial variability across individuals and institutions (Mayes and Horowitz, 2005). The publication of the DSM-III (1980) drew from work initiated in 1972 by a group of Washington University researchers (Feighner et al., 1972) and expanded in 1978 as the Research Diagnostic Criteria (RDC) by Spitzer and colleagues (1978). It emphasized an atheoretically derived group of symptom criteria designed to yield reliable diagnoses that, whenever possible, would also predict course of illness, family history, and treatment response (i.e., “predictive validity”). At the time, these investigators also hoped that biological variables might someday be included as diagnostic validators.

While these changes were built upon expanding scientific knowledge, the DSM-III changes also reflected political viewpoints within the mental health field, increased government involvement in mental health research, and pressure from insurance and pharmaceutical companies (Mayes and Horowitz, 2005). Subsequent revisions to the DSM-III and DSM-IV have further attempted to enhance both the reliability and the validity of the diagnostic criteria. One major revision in the DSM-III-R was the removal of the DSM-III hierarchical structure, which among other things, dictated that an individual could not be diagnosed with an anxiety disorder if depressed (American Psychiatric Association, 1980, 1987).

The current nosological model of the modern DSMs, while not without fault, has facilitated significant research growth in the anxiety disorders field over the past three decades, spurring scientific progress in our understanding of pathophysiology and the identification of effective treatments. It has allowed researchers to more reliably conceptualize a population of interest and clinicians to more reliably identify their patients’ diagnoses. The DSM is now extensively used around the world by clinicians, researchers, health insurance companies, pharmaceutical companies, funding agencies, and policymakers. This has created a shared language regarding psychiatric illness within and across these different professional stakeholders.

While the modern DSMs have bolstered the conceptualization and treatment of anxiety disorders, the current system continues to fall short of its fundamental goals of establishing a reliable and valid classification of mental disorders. While reliability has been largely achieved with the DSM-III and DSM-IV, a number of issues suggest that the validity of the diagnostic system has yet to be achieved. Among these issues are criticisms that the high rate of comorbidity among the disorders indicates an unwarranted splitting of underlying entities. The polythetic criterion, the requirement that only some criteria be present for diagnostic threshold to be met, has been criticized as resulting in widely disparate symptom presentations for the same diagnosis (i.e., two patients could meet criteria for the same disorder while having few, if any, overlapping symptoms). Others note that many clinically distressed individuals do not meet diagnostic criteria for any of the specific diagnoses, and subsequently, this results in an overreliance on not-otherwise-specified (NOS) diagnoses. Finally, and perhaps most importantly, because the current diagnoses developed apart from genetic and neuroimaging research that seeks to clarify the causes and mechanisms of psychiatric illness, the results of this emerging research do not necessarily align with current diagnostic conceptualizations. The DSM-5, currently in its final stages of development, hopes to address some of these criticisms.

PROPOSED REVISIONS IN DSM-5

BACKGROUND

The American Psychiatric Association is currently in the process of revising the DSM-IV to yield the DSM-5. The revisions are intended to incorporate the wealth of research that the field has generated in the past two decades, to maintain continuity with DSM-IV where possible, and to place the highest priority on clinical utility (American Psychiatric Association 2012a). The process has included preplanning white papers, a series of 13 planning conferences, and appointment of chair, vice-chair, and Work Group members. The DSM-5 development process has made use of systematic literature reviews, several rounds of public feedback, and extensive input from professional and consumer stakeholders. The release and publication of the DSM-5 is projected to coincide with the APA’s annual conference in May 2013.

Whereas this chapter is focused on anxiety disorders and the specific changes to these disorders as outlined in what follows, there are a number of general changes that will affect this category. First, the diagnoses previously included in the chapter “Disorders Usually First Diagnosed in Infancy, Childhood, or Adolescence” are now incorporated into other chapters by noting their developmental continuity with adult disorders. Thus, separation anxiety disorder is now listed first in the anxiety disorders chapter. Second, the current revisions have also placed an emphasis on the additional dimensional assessment of psychiatric disorders. The categorical diagnostic system utilized in the previous DSMs is not being replaced by a dimensional system; however, there is an acknowledgment that measurement-based care that utilizes dimensional measures is feasible and would be of benefit to routine clinical care (Regier et al., 2012). Adding these measures will potentially allow for both a more nuanced conceptualization of patients’ psychiatric disorders and a more objective monitoring of patients’ progress throughout therapy. Thus, accompanying the proposed revisions of each anxiety disorder is a note of whether a psychometrically sound measure currently exists. In addition, each anxiety disorder includes two newly developed measures that are being tested for these purposes (the first designed to assess the severity of the specific anxiety disorder and the second designed to assess severity across all anxiety disorders).

Revisions to the anxiety disorders chapter were the res­ponsibility of the Anxiety, Obsessive-Compulsive Spectrum, Posttraumatic, and Dissociative Disorders Work Group (referred from this point on in the chapter as the Anxiety Work Group). The Anxiety Work Group commissioned a series of literature reviews that were published in the peer-reviewed journal Depression and Anxiety. Proposed recommendations were informed by these reviews as well as secondary data analyses of existing datasets, collection of new data, surveys of experts and input and data provided by advisors and liaisons, DSM-5 Task Force members, other members of the research community, and other stakeholders (Phillips et al., 2010). The overarching approach continues to emphasize a search for associated factors that might “validate” these diagnostic categories, but has expanded beyond symptom measures, course of illness, family history and treatment response, to include, wherever possible, medical and psychiatric illness comorbidities, genetic and environmental risk factors, temperamental and personality antecedents, cognitive and emotional processing response measures, and neurocircuitry.

ANXIETY-RELATED DISORDERS ON THE MOVE

OBSESSIVE-COMPULSIVE DISORDER

Obsessive-compulsive disorder (OCD) is characterized by anxiety-provoking obsessions and/or compulsions that are intended to ameliorate anxiety. This disorder is equally common in men and women, yet due to its earlier onset in males, is seen more often in boys than girls. It typically follows a chronic course, and studies indicate a familial risk, with first-degree relatives at higher risk for OCD (American Psychiatric Association 2000).

Much of the discussion regarding OCD and the DSM-5 has surrounded its appropriate placement within the manual. This discussion has considered whether to leave OCD as an anxiety disorder, place it in its own chapter with other disorders from the OC spectrum, or include both anxiety disorders and OCD under the same umbrella category, as is done in the ICD-10, which places both anxiety disorders and OCD under neurotic, stress-related, and somatoform disorders (World Health Organization, 1993). This last option was the favorite of the majority of authors in both the Phillips et al. (2010) and Stein et al. (2010) reviews. However, OCD researchers were mixed regarding their approval (60%) of the removal of OCD from the supraordinate anxiety category (Mataix-Cols et al., 2007), and clinical psychiatrists and “other professionals,” largely psychologists, showed a significant difference in opinion (75% agreed versus 40%–45%, respectively). Stein et al. (2010) outlined the research from a series of validators (e.g., neurocircuitry; course of illness; treatment response; and genetic, familial, and environmental risk factors) regarding OCD’s appropriate placement. While there are some uniquely distinctive aspects to OCD (involvement of fronto-striatal neurocircuitry and related deficits, a narrower range of effective medications and a clear dose–response pattern to SRI medication response, and a unique association with basal ganglia neurologic disorders), there is still much overlap with anxiety disorders in multiple other domains (e.g., comorbidity, family history). Generally, these results present a mixed picture providing both support for the inclusion of OCD within the anxiety disorders and its removal. Nevertheless, DSM-5 will move OCD to the Obsessive-Compulsive and Related Disorders category along with several other disorders including body dysmorphic disorder, hoarding, hair pulling (trichotillo­mania), and skin picking. These disorders are all characterized by repetitive, anxiety- or dysphoria-inducing thoughts and/or behaviors (designed at reducing the discomfort). While genetic, neural, and biomarker validators that support this new category are few, several other validators link these disorders with OCD or one another, including high comorbidity, similar psychopharmacology and psychotherapy response, and evidence of familial transmission (Phillips et al., 2010).

Proposed changes to the OCD criteria have been made and include moving the two items from the obsessions definition that sought to distinguish obsessions from GAD (generalized anxiety disorder; obsessions are not simply excessive worries regarding life problems) and psychosis (the individual recognized that the obsessional thoughts, impulses, or images are not created by an external force but rather are a product of his or her own mind) and placing them in Criterion D among the other diagnostic hierarchy issues (Leckman et al., 2010). Additionally, the requirement that individuals recognize that their symptoms are excessive was removed, and the poor insight specifier was replaced with specifiers indicting a range of insight (i.e., good or fair insight, poor insight, or absent insight). Finally, suggestions are made for dimensional measures. Previously validated measures include the Yale-Brown Obsessive-Compulsive Scale (Y-BOCS; Goodman et al., 1989), the Florida Obsessive-Compulsive Inventory (FOCI; Storch et al., 2007), and the Brown Assessment of Beliefs Scale (BABS; Eisen et al., 1998). A newly developed five-item questionnaire is also included and currently being tested.

POSTTRAUMATIC STRESS DISORDER

Posttraumatic stress disorder (PTSD) can develop following exposure to a traumatic event and is characterized by reexperiencing the trauma, avoidance of traumatic cues, and increased symptomatic arousal. Evidence suggests that the development of PTSD following trauma is affected by social supports, childhood experiences, personality variables, type of trauma, and preexisting mental disorders (American Psychiatric Association, 2000). Additionally, genetic factors appear to affect the likelihood of developing the disorder (Stein et al., 2002).

Like OCD, much of the DSM-5 focus on PTSD has surrounded its most appropriate placement within the DSM. Stress-related fear circuitry findings from neuroimaging and fear conditioning studies, among other findings, support the retention of PTSD within the anxiety disorders (Friedman et al., 2011a). But PTSD frequently encompasses a broader range of emotions (e.g., numbing, guilt, alienation) than other anxiety disorders and shares a unique precipitating event (all cases are trauma induced). Emphasizing common etiology over symptom similarity, it has been concluded that PTSD should be placed in a newly formed category that includes disorders that are precipitated by a “serious life event” (the Trauma-and Stressor-Related Disorders). PTSD’s anticipated departure from the anxiety disorders has met strong criticism. Zoellner et al. (2011) note that the “rationale for this shift is unclear, underdeveloped, and unsupported” (p. 853) and delineate four main points that support retaining PTSD as an anxiety disorder: (1) fear is a critical component of PTSD, (2) the treatment of fear and avoidance is central to PTSD as it is with other anxiety disorders, (3) the evidence base does not support a separate category, and (4) reclassification moves the PTSD field away from its well-developed knowledge base.

The diagnostic criteria for PTSD are lengthier than most other diagnoses, and efforts at their improvement have been a frequent focus in previous revisions of the DSM. DSM-5, alas, will do nothing to simplify or shorten the diagnostic criteria. However, several of the proposed revisions to the PTSD criteria within the DSM-5 reflect substantial changes. Criterion A1, defining the nature of the traumatic event, has been retained, but the description of what constitutes an index traumatic event has been articulated in greater detail. Criterion A2, requiring an emotional response of fear, helplessness, or horror in response to the trauma, has been eliminated. One reason for its removal is that some individuals (e.g., military personnel, who in deployment situations are trained not to have a response like this) lack this type of emotional response at the time of the event but, nonetheless, can develop PTSD (Friedman et al., 2011b). Because of confirmatory factor analyses that indicate strongest support for a four-factor model (Friedman et al., 2011b), the DSM-5 is proposing splitting the current three-factor model (Criterion B—intrusion, C—avoidance, and D—heightened arousal) into a four-factor model by splitting Criterion C and adding symptoms to the newly formed Criterion D (negative cognitions and mood). There are also a number of other specific changes to the individual criteria. In an effort to enhance cross-cultural applicability of the criteria, the intrusion symptom regarding dreams about the trauma has been revised to include both trauma content or affect about the trauma (Hinton and Lewis-Fernandez, 2011). As the previous C category has been split in two with only two types of avoidance retained, the criteria now require only one to be endorsed. The new category D has two new symptoms, blame and negative emotional state (which, on the face of it, will overlap tremendously with major depressive symptoms), and the previous symptom of a foreshortened future has been expanded to encompass exaggerated negative beliefs and expectations to help expand the applicability to other cultures. Category E also has a new symptom that focuses on reckless and destructive behavior.

Friedman et al. (2011b) also conclude that there is insufficient research to support the acute versus chronic specifier and thus it has been eliminated. However, they conclude that there is support for both a dissociative subtype and a preschool subtype. There is evidence to suggest that individuals whose PTSD is characterized by dissociative symptoms demonstrate distinct prefrontal responses and a decrease in heart rate and skin conductance to trauma cues/memories (Friedman et al., 2011b). Traumatized preschool age children do develop PTSD but at lower rates than other age groups, which likely reflect, at least in part, that the current diagnostic criteria do not accurately capture the disorder among preschoolers. Thus, the revised diagnostic criteria for the preschool subtype is meant to more readily identify PTSD in these young people by being more developmentally sensitive and behaviorally anchored (Scheeringa et al., 2011). For a dimensional measure of PTSD, the Web site suggests the unpublished nine-item National Stressful Events Survey PTSD Short Scale (American Psychiatric Association, 2011).

MIXED ANXIETY/DEPRESSION DISORDER

Mixed anxiety/depression disorder (MADD) is characterized by dysphoric mood that is accompanied by symptoms of both anxiety and depression and is considered to be subordinate to anxiety and mood disorders (not considered if diagnostic criteria for one of these disorders has been met). Empirical work on this disorder and its validators is sparse.

While included as an official diagnosis in the ICD-10, MADD was relegated in DSM-IV to Criteria Sets and Axes Provided for Further Study (American Psychiatric Association 2000). Initial versions of the DSM-5 revisions moved MADD to mood disorders. This proposed move was met with intense criticism from a variety of standpoints, including MADD’s remarkably poor reliability as well as the paucity of validating data (e.g., First, 2011; Frances, 2012; Wakefield, 2012). On the heels of this criticism, the most recent DSM-5 revisions have moved MADD back to the section reserved for conditions requiring further study (where it had been placed in DSM-IV). This is, admittedly, an unorthodox move on the part of the DSM-5 framers. If the past 20 years was not enough time to study MADD, it seems unlikely that more specific criteria will result in a boon of research. A more judicious approach would have been to delete MADD altogether.

Separation Anxiety Disorder

This disorder is expressed as developmentally excessive anxiety in the face of separation from the home or an attachment figure. While it is more common among females, representation among the sexes is equal in clinical samples. The risk of developing separation anxiety disorder is elevated for children of mothers with panic disorder.

Despite clinically significant impairment as children, those with this disorder often do not have a disproportionate rate of anxiety disorders as adults (American Psychiatric Association, 2000).

Because of the redistribution of disorders previously included in the Disorder Usually First Diagnosed in Infancy, Childhood or Adolescence, separation anxiety disorder will be included in the Anxiety Disorders section of the DSM-5. The literature that the Anxiety Work Group relied upon to make the DSM-5 proposed changes to separation anxiety disorder is not publicly available at this time. But the proposed changes to the diagnosis are available and include removal of the criterion that onset occurs prior to age 18, further differentiating the DSM from the ICD-10 criteria that requires an age of onset prior to 6 (World Health Organization, 1993). The specifier “early-onset” has also been removed.

ANXIETY DISORDERS

AGORAPHOBIA

Agoraphobia (AG) is characterized by a fear of being in situations in which escape may be difficult or in which help may not be readily available. While not on the move to a different chapter as are the diagnoses in the previous section, the revisions to the DSM-5 have moved AG from subordinate to panic (i.e., necessarily attributing the avoidance of situations to a fear of panic-related symptoms) to an independent disorder, a status it originally held in the DSM-III. This will be consistent with how agoraphobia is treated by DSM’s international counterpart, the ICD (International Classification of Diseases; World Health Organization, 1993).

This decision was made despite remaining questions and controversy, as the authors of the literature review indicate that they reached a consensus in their recommendations but that the consensus was not unanimous (Wittchen et al., 2010). However, the decision was supported by familial genetic data, psychiatric history, patterns of comorbidity, course of illness, treatment response, and reliability (American Psychiatric Association, 2012b). This evidence, reviewed by Wittchen et al. (2010), highlights that AG without panic-like symptoms does occur (in epidemiologic studies) and is associated with significant disability and a persistent course with low rates of spontaneous recovery. Thus, while the majority of AG cases do occur in the presence of panic disorder, panic attacks, or panic-like symptoms, that does not uniformly demonstrate that AG is a function of panic.

The proposed criteria for AG are more specific than in DSM-IV to improve differential diagnosis. Criteria A lists five different groups of agoraphobic situations, provides several examples of each group, and requires the endorsement of one situation from two or more groups. Additionally, the new criteria include a six-month duration requirement to avoid unnecessarily pathologizing transient fear. In addition to the two newly developed dimensional measures, The Fear Questionnaire-Agoraphobia Subscale (Marks and Mathews, 1979) is listed as a previously validated dimensional measure, but specific recommendations regarding its use are not included at this time.

PANIC DISORDER

Panic disorder (PD) is characterized by discrete periods of sudden and intense physiological symptoms and fear (i.e., panic attacks [PAs]) that result in persistent concern about additional PAs or a change in behavior (avoidance, emergency department visits, seeking medical diagnostic tests). Age of onset for this disorder varies but is generally between late adolescence and mid-30s. Twin studies suggest a genetic contribution to the disorder, and studies indicated that familial risk is particularly elevated for first-degree relatives of those who developed PD prior to age 20 (American Psychiatric Association, 2000). Early suggestions that PD exhibited a unique pharmacological treatment response are untrue, and notions of unique psychosocial treatment response have been called into question by transdiagnostic cognitive behavioral therapy approaches (e.g., Norton and Philipp, 2008).

As PAs are a crucial piece of PD, the Anxiety Work Group–commissioned review of evidence focused on questions related both to PAs and PD (Craske et al., 2010). It suggested that the symptom “hot flushes” be changed to “heat sensations” to better capture cultural variants in this experience and recommended that culture-specific symptoms be included in the description of PAs but not counted as one of the four required symptoms. The extant evidence did not suggest that a change in the number of symptoms required for a PA is warranted at this time. To reinforce the paroxysmal temporal profile of panic, the wording was changed to indicate that a panic attack reaches “a peak within minutes” and “can occur from a calm state or an anxious state.” There was no new evidence exploring validators of panic attack frequency to suggest that the criteria of “recurrent” PAs be modified. Due to the empirical evidence that panic attacks are associated with an increase in symptom severity, comorbidity, suicidality, and treatment resistance of comorbid disorders, a note was added indicating that PAs can serve as a specifier for both anxiety and nonanxiety disorders (e.g., schizophrenia, panic attack specifier). As noted in the previous section, agoraphobia is now a codable disorder, and thus, PD is no longer listed as PD with agoraphobia or PD without agoraphobia.

The DSM-5 Web site identifies the Panic Disorder Severity Scale-Self-Report (Houck et al., 2002) as a dimensional measure that can be utilized to assess PD severity.

SPECIFIC PHOBIA

Specific phobia (SP) is characterized by clinically significant distress or anxiety in response to a specific feared object or situation. While it is more commonly found among women, the sex difference varies by type (e.g., animal and situational). Evidence indicates that familial risk seems to aggregate by type and that blood injection injury–type fears have a particularly strong family link. The symptoms of SP usually first emerge in childhood or midadolescence, and for those that persist into adulthood, remission is infrequent (American Psychiatric Association, 2000). There is some research suggesting neural differences between SP types, but this research remains at a nascent stage (LeBeau et al., 2010).

The diagnostic criteria for SP, which was referred to as simple phobia prior to the DSM-IV, have seen several modifications in preparation for the DSM-5 revisions. The review by LeBeau and colleagues (2010) focused on four main areas: the accuracy and utility of the specific phobia type classification, the validity of test anxiety as a type of SP, the boundary between agoraphobia and SP, and the reliability and utility of the specific phobia criteria. The general conclusion of the review is that the extant literature does support the retention of the specific phobia types as a descriptive option; that little evidence exists either for or against the inclusion of test anxiety as a type of SP; and similarly, that there is insufficient evidence to reclassify agoraphobia as a type of SP. Finally, LeBeau and colleagues (2010) reviewed the SP criteria and made recommendations about rewording and reordering the specific criteria in order to enhance consistency across anxiety disorders and improve clinical utility and ease of use. The criterion that one recognize that his or her fear is excessive or unreasonable has been removed, as it is common for adults to deny that their fear is excessive or unreasonable (LeBeau et al., 2010). Due to research indicating that transient fears and phobias occur among adults (LeBeau et al., 2010), the duration criterion of six months for those under the age of 18 has been extended to all age groups. Currently, the Web site does not recommend any specific dimensional scales for measurement-based care, since most measures assess one specific type of SP; however, two newly developed alternative scales focusing on severity of SP and of anxiety in general are now being tested.

Social Anxiety Disorder

Social anxiety disorder (SAD) is characterized by a marked fear of social and performance situations that often results in avoidance of those situations. Epidemiological data suggest that it is more common among women than men but that the sexes are equally represented in clinical populations. Onset typically occurs by late teens and can be either insidious or abrupt. The course of SAD is typically continuous and lifelong. Increased familial risk among first-degree relatives is particularly strong for the generalized type (American Psychiatric Association, 2000).

Social phobia is currently the official name for this diagnosis, but the term social anxiety disorder was added in parentheses following this title in the DSM-IV-TR to reflect that this is a broader diagnosis rather than just a circumscribed phobia (Bogels et al., 2010). Much of the Work Group review focused on the SAD specifiers. The DSM-5 framers have elected to delete the “generalized” specifier for SAD, arguing that there was considerable heterogeneity within this category. Instead, a performance-only specifier was introduced, supported by evidence that there are qualitative differences these individuals demonstrate from others with SAD (later onset, not characterized by childhood factors of shyness or behavioral inhibition, not familial, a stronger physiological response, and more likely to respond to beta blockers). In theory, if an individual with SAD does not meet the performance-only specifier, then he or she would be highly likely to have what had been called generalized SAD in DSM-IV. The merits of this change remain to be determined, whereas the disadvantages of disconnecting from two decades of research and over a thousand PubMed references to generalized SAD are obvious.

Selective mutism, listed in DSM-IV as a separate diagnosis in Disorders Usually First Diagnosed in Infancy, Childhood, or Adolescence, had tentatively been proposed as a SAD specifier since the majority of those with selective mutism have SAD, there are very high rates of SAD among parents of children with selective mutism, and preliminary treatment evidence suggests that psychological and pharmacological treatments that are effective for SAD are also effective for selective mutism. It was ultimately decided, however, to situate selective mutism as a separate diagnosis within the Anxiety Disorders section, apparently to encourage further attention to factors which may distinguish selective mutism from SAD (e.g., developmental language or speech problems).

Revisions to the SAD criteria also evaluated the comorbidity between SAD and avoidant personality disorder (AVPD), thought by some to be a more severe version of SAD rather than a qualitatively distinct disorder. Alden, Laposa et al. (2002) report that AVPD studies investigating the comorbidity of SAD note an average comorbidity of 42% with SAD, which is considerably lower than one would anticipate if AVPD were synonymous with severe SAD. Genetic studies indicate a link between SAD and AVPD, but this link does not appear to be specific to these two disorders (Bogels et al., 2010). Similarly, psychological and pharmacological treatment studies indicate a similar response style between the two disorders; however, numerous other disorders also respond to these treatments. In addition, there are some indications that AVPD shares links with the schizophrenia spectrum (Bogels et al., 2010). Based in part on these findings, it was concluded that while these two have a high degree of overlap, it is too simplistic to consider AVPD a severe form of SAD (Bogels et al., 2010). Thus, the DSM-5 revisions will not collapse these two disorders.

Additional revisions to the SAD criteria include the expansion of the types of social situations that are avoided to include social interaction, observation, and performance with examples of each provided. To increase the cultural sensitivity of the diagnosis, the feared consequence of offending others has been added to the previously listed fear of negative evaluation. The criteria no longer require the individual to recognize that his or her fear is excessive but does require the fear to be out of proportion to the actual threat posed. This acknowledges that some patients do not see their fear as excessive, but they are clearly not psychotic, and it is sufficient for the clinician to assess the excessiveness of the fear. Finally, the duration ­criterion of six months has been extended beyond those under the age of 18 to now apply across all age ranges, as it was noted that adults can also experience transient social anxiety, which might result in overdiagnosis.

In addition to the two newly developed dimensional measures, the Web site lists two previously published measures, the Social Phobia Inventory (Connor et al., 2000) and Mini-Social Phobia Inventory (Connor et al., 2001), along with a brief description of each.

GENERALIZED ANXIETY DISORDER

Generalized anxiety disorder (GAD) is characterized by clinically significant worry and anxiety that is disproportionate to the circumstances and persists for at least six months. While adult onset is not uncommon, approximately half of the individuals who present for treatment report childhood or adolescent onset. Course of this disorder is generally chronic but often worsens in times of stress. Twin studies indicate a genetic contribution to the development of GAD (American Psychiatric Association, 2000).

The diagnosis of GAD has been encumbered by poor reliability and substantial comorbidity with mood and anxiety disorders since its first inclusion in DSM-III. Revisions to subsequent versions have attempted to remedy this situation. Likewise, the proposed revisions by Andrews et al. (2010) are meant to further clarify the diagnosis and enhance its reliability. Included in these recommendations was reducing the duration criterion from six to three months as this would substantially increase test–retest reliability and not impact the type of patient included in terms of distress and impairment (Andrews et al., 2010). The criterion that worry is difficult to control has been removed because of a lack of validating support to indicate that it added anything to the criteria (Andrews et al., 2010). Four of the six associated symptoms of worry listed in Criterion C (being easily fatigued, difficulty concentrating or mind going blank, irritability, and sleep disturbance), are nonspecific to GAD and are being proposed for removal, retaining only the two associated symptoms that are specific to the diagnosis (restlessness or feeling keyed up or on edge and muscle tension) to enhance the GAD’s discriminant validity, especially from major depressive disorder, which greatly overlaps with these nonspecific symptoms. A list of four behaviors commonly employed by individuals with GAD in an attempt to decrease their worry or ameliorate distressing affect are now proposed for inclusion (e.g., repeatedly seeking reassurance due to worries), one of which needs to be endorsed to meet the diagnostic criteria. When the specific combination of the currently proposed changes was investigated, the prevalence of the disorder increased by 9%. However, these changes did not significantly affect the distress or impairment, indicating that the severity was retained and lending support to these changes (Andrews and Hobbs, 2010). At the time of this writing, it is unclear whether the preceding changes to GAD will be accepted or not. As with the other disorders, a previously published self-report measure of severity, the GAD-7 (Spitzer et al., 2006), is included among the DSM-5 revisions for GAD.

NEED FOR A NEW NOSOLOGICAL FRAMEWORK?

As noted, the central goal of the DSM has been to create a reliable and valid diagnostic system for mental illness that has concurrent and predictive validity. The past several decades have witnessed an increase in the reliability of psychiatric diagnoses and an accumulation of knowledge surrounding associated factors, both clinical and experimental, that might serve as validators. However, much work remains as diagnoses fail to align with recent findings from neuroscience and genetics, the diagnostic boundaries are not consistent with treatment response, and current validators remain relatively limited and lacking specificity.

A recent review nicely outlines the difficulty in identifying consistent validating data to support even the broad distinction of an “anxiety” disorder from other disorders of emotional distress and misery (Craske et al., 2009). Extant evidence does suggest there is some consistency in unique, anxiety-specific findings of distinct self-report measures, elevated sensitivity to threat (based on Pavlovian conditioning paradigms), cognitive bias to threat, and specific anxiety neurocircuitry. However, inconsistency in findings across developmental stages, lack of distinction among the individual anxiety disorders, and some overlap in findings with depressive disorders continue to confound attempts to nail down even the nosological validity of anxiety disorders as a distinct group! The DSM-5 revisions attempt to incorporate new research to bolster the diagnostic system and will likely result in incremental steps toward increased diagnostic validity and reliability, but addressing the evidence base honestly requires that we recognize we are still a long way off from a goal of diagnostic entities that have the kind of pathophysiologic validity of many recognized medical disorders.

The reviews commissioned by the Anxiety Work Group clearly indicate that much empirical work is left to be done. A common theme across these reviews was the call for additional research, and in numerous instances, the authors noted that there was simply insufficient empirical evidence for them to draw conclusions regarding central questions they had been asked to review. Whereas the DSM system is not perfect, it has become indispensable as it helps to bridge researchers and clinicians and is used extensively across settings such as insurance funding, treatment, and legal proceedings (Frances and Widiger, 2012). A system based on clinical description, such as we have, cannot match the validity (and even to some degree, the reliability) of a system based on objective medical tests; however, a mental disorder diagnostic system based on objective tests remains a long-term goal rather than a reality (Bernstein, 2011). The DSM was and is still meant to be a living document that serves as a framework for both research and clinical work that can expand and change based on accumulating knowledge. The mind is very complex, and the lofty goal of carving psychiatric nature at its joints requires more clearly elucidating how it functions and how missteps along its very complicated functioning can result in psychiatric illness.

Kendler has criticized commonly held explanatory models for psychiatric illness and puts forth the need to conceptualize the causes of psychiatric illnesses with “empirically based pluralism” (e.g., Kendler, 2008, 2012). He suggests that psychiatry has had the tendency to inaccurately dichotomize the etiology of disorders because of the influence of Descartes and computer functionalism. Descartes’ dualistic approach to psychiatry distinguishes between the mind (thinking) and the brain (physical) as two fundamentally separate entities and, following from this, that disorders are either organic (brain based) or functional (mind based). Similarly, Kendler points out that the advent of the computer era ushered in the functionalist perspective that our functioning runs parallel to that of computers, such that the computer hardware is synonymous with our brain and the software is synonymous with our mind. He suggests that this dichotomous thinking about mind/brain functioning continues to influence our conceptualization of the etiology of psychiatric illness, and beyond it simply being inaccurate, it has impaired our ability to integrate research from multiple domains into our understanding of psychiatric illness. In one of his articles (Kendler, 2012), he makes the case for the inaccuracy of this dualistic perspective by reviewing the biological, psychological, and higher order (social, political, and cultural) etiological effects on major depression, alcohol dependence, and schizophrenia. The presented research indicates that each of the three disorders is affected by factors from each domain and that these factors are not independent but, rather, mediate and moderate each other both within and across domains. Thus, he concludes that understanding the nature of psychiatric illnesses requires an appreciation for the complex interplay between factors from various domains that do not fit into the neat mind/brain divide. He further suggests that a pluralistic view is driven by research and not a priori theory.

On first glance, this view might seem at variance with the notion that psychiatric disorders could be mapped to definitive genetic etiologies tied to neuronal circuit pathophysiologies. However, recent meta-analyses of neuroimaging studies provide tantalizing clues that PTSD, GAD, and MDD share specific neurocircuitry characteristics (hyperactivity of the dorsal anterior cingulate cortex, which is responsible for monitoring and expression of anxiety, and hypoactivity of the ventral anterior cingulate cortex, responsible for extinguishing fear responses) that are distinct from more classic “fear disorders” like panic, social anxiety, and the phobic disorders (Etkin, 2012). Ironically, this does not accord with the proposed DSM-5 changes, as it does not suggest PTSD is unique from all anxiety and mood disorders, nor that GAD is similar to other anxiety disorders.

Related to Kendler’s call for an understanding of the complex interplay between factors is NIMH’s Research Domain Criteria (RDoC) project (National Institute of Mental Health, 2011). The NIMH has indicated that it will focus its funding priorities on projects that are developed within this framework. It is meant to guide research in a manner that bypasses current diagnostic systems by organizing the focus of research around the following five domains: negative affect (e.g., loss, fear, anxiety), positive affect (e.g., approach motivation, reward learning), cognitive systems (e.g., attention, perception, working memory), systems for social processes (e.g., attachment, imitation), and arousal/regulatory systems. Research into these domains is further structured around specific units of analysis: genes, molecules, cells, circuits, physiology, behavior, self-reports, and paradigms. It is noted that this initial structure is meant to be a starting point that reflects the current state of knowledge and that it will be modified and adjusted as warranted by new research findings (Sanislow et al., 2010). Within this framework, participants would not be selected based on current diagnostic categories, but rather, for example, a sample of all patients presenting to an anxiety clinic could be studied to investigate amygdala functioning in response to fearful stimuli (Sanislow et al., 2010). The hope is that by not restricting research to current diagnostic categories research knowledge will develop that is more apt to identify the underlying mechanisms of mental illness. RDoC efforts can be seen as providing a complementary path to that of DSM-5 for the study of mental disorders. It follows that this knowledge would eventually translate into practitioners being able to supplement clinical evaluation with functional or structural imaging, genomic sequencing, or lab-based evaluations of fear conditioning to more effectively determine prognosis and effective treatment (Insel et al., 2010).

As noted (Insel, 2009), the RDoC project is motivated by lofty long-term goals that may prove to dramatically affect the field. However, it also rests on three assumptions: mental illnesses are brain disorders, neural circuit dysfunction can be identified through tools of neuroscience, and genetic and clinical neuroscience research will provide biosignatures that will augment clinical management (Insel et al., 2010). If these assumptions turn out to be valid, then the RDoC project may lead to a revolution in our understanding of anxiety and other forms of mental illness. Until that day, DSM and its latest iteration (DSM-5) will continue to serve as the clinical guidepost for our field.

DISCLOSURES

Dr. Keough has no conflicts of interest to disclose. She currently receives NIMH funding through a National Research Service Award Institutional Training Grant (T32MH082709-01A2).

Dr. Stein is paid as co-editor-in-chief for UpToDate in Psychiatry, and as associate editor for the journal Depression and Anxiety. His research is funded by NIMH, the VA, and the Department of Defense.

Dr. Roy-Byrne has grant funding from NIDA and NIMH, receives salary for editor-in-chief duties for Depression and Anxiety, UpToDate in Psychiatry, and Journal Watch Psychiatry, and has received stock options as a consultant and advisor to Valant Medical Solutions, a behavioral health EMR company.

REFERENCES

Alden, L.E., Laposa, J.M., et al. (2002). Avoidant personality disorder: current status and future directions. J. Pers. Disord. 16:1–29.

American Psychiatric Association. (1952). Diagnostic and Statistical Manual of Mental Disorders, 1st Edition. Washington, DC: Author.

American Psychiatric Association. (1968). Diagnostic and Statistical Manual of Mental Disorders, 2nd Edition. Washington, DC: Author.

American Psychiatric Association. (1980). Diagnostic and Statistical Manual of Mental Disorders, 3rd Edition. Washington, DC: Author.

American Psychiatric Association. (1987). Diagnostic and Statistical Manual of Mental Disorders, 3rd Edition, Revised. Washington, DC: Author.

American Psychiatric Association. (2000). Diagnostic and Statistical Manual of Mental Disorders, 4th Edition, Text Revision. Washington, DC: Author.

American Psychiatric Association. (2011). Home / proposed revisions / trauma and stressor related disorders // G 03 posttraumatic stress disorder. http://www.dsm5.org/ProposedRevision/Pages/proposedrevision.aspx?rid=165#

American Psychiatric Association. (2012a). Frequently asked questions: why is DSM being revised? http://www.dsm5.org/about/Pages/faq.aspx

American Psychiatric Association. (2012b). Proposed revisions: agoraphobia. http://www.dsm5.org/ProposedRevision/Pages/proposedrevision.aspx? rid=405#

Andrews, G., and Hobbs, M.J. (2010). The effect of the draft DSM-5 criteria for GAD on prevalence and severity. Aust. N. Z. J. Psychiatry 44(9):784–790.

Andrews, G., Hobbs, M.J., et al. (2010). Generalized worry disorder: a review of DSM-IV generalized anxiety disorder and options for DSM-V. Depress. Anxiety 27(2):134–147.

Bernstein, C.A. (2011). Meta-structure in DSM-5 process. Psychiatric News 46(5):7.

Bogels, S.M., Alden, L., et al. (2010). Social anxiety disorder: questions and answers for the DSM-V. Depress. Anxiety 27(2):168–189.

Connor, K.M., Davidson, J.R., et al. (2000). Psychometric properties of the Social Phobia Inventory (SPIN): new self-rating scale. Br. J. Psychiatry 176:379–386.

Connor, K.M., Kobak, K.A., et al. (2001). Mini-SPIN: a brief screening assessment for generalized social anxiety disorder. Depress. Anxiety 14(2):137–140.

Craske, M.G., Kircanski, K., et al. (2010). Panic disorder: a review of DSM-IV panic disorder and proposals for DSM-V. Depress. Anxiety 27(2):93–112.

Craske, M.G., Rauch, S.L., et al. (2009). What is an anxiety disorder? Depress. Anxiety 26(12):1066–1085.

Eisen, J.L., Phillips, K.A., et al. (1998). The Brown Assessment of Beliefs Scale: reliability and validity. Am. J. Psychiatry 155(1):102–108.

Etkin, A. (2012). Neurobiology of anxiety: from neural circuits to novel solutions? Depress. Anxiety 29(5):355–358.

Feighner, J.P., Robins, E., et al. (1972). Diagnostic criteria for use in psychiatric research. Arch. Gen. Psychiatry 26(1):57–63.

First, M.B. (2011). DSM-5 proposals for mood disorders: a cost-benefit analysis. Curr. Opin. Psychiatry 24(1):1–9.

Frances, A. (2012). Newsflash from APA meeting: DSM 5 has flunked its reliability tests. Psychol. Today.

Frances, A.J., and Widiger, T. (2012). Psychiatric diagnosis: lessons from the DSM-IV past and cautions for the DSM-5 future. Annu. Rev. Clin. Psychol. 8:109–130.

Friedman, M.J., Resick, P.A., et al. (2011a). Classification of trauma and stressor-related disorders in DSM-5. Depress. Anxiety 28(9):737–749.

Friedman, M.J., Resick, P.A., et al. (2011b). Considering PTSD for DSM-5. Depress. Anxiety 28(9):750–769.

Goodman, W.K., Price, L.H., et al. (1989). The Yale-Brown Obsessive Compulsive Scale: I. development, use, and reliability. Arch. Gen. Psychiatry 46(11):1006–1011.

Hinton, D.E., and Lewis-Fernandez, R. (2011). The cross-cultural validity of posttraumatic stress disorder: implications for DSM-5. Depress. Anxiety 28(9):783–801.

Houck, P.R., Spiegel, D.A., et al. (2002). Reliability of the self-report version of the panic disorder severity scale. Depress. Anxiety 15(4):183–185.

Insel, T., Cuthbert, B., et al. (2010). Research domain criteria (RDoC): toward a new classification framework for research on mental disorders. Am. J. Psychiatry 167(7):748–751.

Insel, T.R. (2009). Translating scientific opportunity into public health impact: a strategic plan for research on mental illness. Arch. Gen. Psychiatry 66(2):128–133.

Kendler, K.S. (2008). Explanatory models for psychiatric illness. Am. J. Psychiatry 165(6):695–702.

Kendler, K.S. (2012). The dappled nature of causes of psychiatric illness: replacing the organic-functional/hardware-software dichotomy with empirically based pluralism. Mol. Psychiatry 17(4):377–388.

LeBeau, R.T., Glenn, D., et al. (2010). Specific phobia: a review of DSM-IV specific phobia and preliminary recommendations for DSM-V. Depress. Anxiety 27(2):148–167.

Leckman, J.F., Denys, D., et al. (2010). Obsessive-compulsive disorder: a review of the diagnostic criteria and possible subtypes and dimensional specifiers for DSM-V. Depress. Anxiety 27(6):507–527.

Marks, I.M., and Mathews, A.M. (1979). Brief standard self-rating for phobic patients. Behav. Res. Ther. 17(3):263–267.

Mataix-Cols, D., Pertusa, A., et al. (2007). Issues for DSM-V: how should obsessive-compulsive and related disorders be classified? Am. J. Psychiatry 164(9):1313–1314.

Mayes, R., and Horowitz, A.V. (2005). DSM-III and the revolution in the classification of mental illness. J. Hist. Behav. Sci. 41(3):249–267.

National Institute of Mental Health. (2011). NIMH Research Domain Criteria (RDoC). URL http://www.nimh.nih.gov/research-funding/rdoc/nimh-research-domain-criteria-rdoc.shtml

Norton, P.J., and Philipp, L.M. (2008). Transdiagnostic approaches to the treatment of anxiety disorders: a quantitative review. Psychother. Theor. Res. Pract. Train. 45(2):214–226.

Phillips, K.A., Friedman, M.J., et al. (2010). Special DSM-V issues on anxiety, obsessive-compulsive spectrum, posttraumatic, and dissociative disorders. Depress. Anxiety 27(2):91–92.

Phillips, K.A., Stein, D.J., et al. (2010). Should an obsessive-compulsive spectrum grouping of disorders be included in DSM-V? Depress. Anxiety 27(6):528–555.

Regier, D.A., Kuhl, E.A., et al. (2012). Research planning for the future of psychiatric diagnosis. Eur. Psychiatry 27(7):553–556.

Sanislow, C.A., Pine, D.S., et al. (2010). Developing constructs for psychopathology research: research domain criteria. J. Abnorm. Psychol. 119(4):631–639.

Scheeringa, M.S., Zeanah, C.H., et al. (2011). PTSD in children and adolescents: toward an empirically based algorithma. Depress. Anxiety 28(9):770–782.

Spitzer, R.L., Endicott, J., et al. (1978). Research diagnostic criteria. Arch. Gen. Psychiatry 35:773–782.

Spitzer, R.L., Kroenke, K., et al. (2006). A brief measure for assessing generalized anxiety disorder: the GAD-7. Arch. Intern. Med. 166(10):1092–1097.

Stein, D.J., Fineberg, N.A., et al. (2010). Should OCD be classified as an anxiety disorder in DSM-V? Depress. Anxiety 27(6):495–506.

Stein, M.B., Jang, K.L., et al. (2002). Genetic and environmental influences on trauma exposure and posttraumatic stress disorder symptoms: a twin study. Am. J. Psychiatry 159:1675–1681.

Storch, E.A., Kaufman, D.A., et al. (2007). Florida obsessive-compulsive inventory: development, reliability, and validity. J. Clin. Psychology 63(9):851–859.

Wakefield, J.C. (2012). DSM-5: proposed changes to depressive disorders. Curr. Med. Res. Opin. 28(3):335–343.

Wittchen, H.U., Gloster, A.T., et al. (2010). Agoraphobia: a review of the diagnostic classificatory position and criteria. Depress. Anxiety 27(2):113–133.

World Health Organization. (1993). The ICD-10 Classification of Mental and Behavioural Disorders. Geneva: World Health Organization.

Zoellner, L.A., Rothbaum, B.O., et al. (2011). PTSD not an anxiety disorder? DSM committee proposal turns back the hands of time. Depress. Anxiety 28(10):853–856.