Panic Disorder and Agoraphobia

DSM-IV-TR Criteria

Of all the anxiety disorders, panic disorder is set to undergo the most changes in the proposed DSM-5. In the DSM-IV, there are three separate diagnoses, Panic Disorder with Agoraphobia, Panic Disorder without Agoraphobia, and Agoraphobia without History of Panic Disorder, while the DSM-5 proposes to have only two: Panic Disorder and Agoraphobia. As such, this section will be a bit different from the other anxiety disorders, in that I will detail information about DSM-IV panic attacks, agoraphobia, panic disorder, and then discuss the etiology, treatments, and DSM-5 changes that are proposed across all three.

Panic Attack

DSM-IV-TR Criteria

NOTE: A panic attack is not a codeable disorder. Code the specific diagnosis in which the panic attack occurs (e.g., 300.21 Panic Disorder with Agoraphobia)

A discrete period of intense fear or discomfort, in which four (or more) of the following symptoms developed abruptly and reached a peak within 10 minutes:

  • 1. Palpitations, pounding heart, or accelerated heart rate
  • 2. Sweating
  • 3. Trembling or shaking
  • 4. Sensations of shortness of breath or smothering
  • 5. Feeling of choking
  • 6. Chest pain or discomfort
  • 7. Nausea or abdominal distress
  • 8. Feeling dizzy, unsteady, lightheaded, or faint
  • 9. De-realization (feelings of unreality) or depersonalization (being detached from oneself)
  • 10. Fear of losing control or going crazy
  • 11. Fear of dying
  • 12. Paresthesias (numbness or tingling sensation)
  • 13. Chills or hot flashes

Associated Features

Panic attacks (PA) are actually fairly common across all the anxiety disorders, but are especially prevalent in the phobias and post-traumatic stress disorder. They usually last several minutes and can mimic signs of a heart-attack to those not familiar with them. The most commonly reported PA symptoms are heart-pounding and dizziness, although there is great variability among PA, even in the same person (as indicated by the large number of possible symptoms). The least common symptoms (paresthesias, choking, and fear of dying) are indicative of more severe PA problems and likelihood of reoccurrence. Also, the higher number of symptoms, the more severe the PA will be. In fact, one’s risk for suicide attempts and emergency room use was elevated by 20% for each additional PA symptom above the four. If a person has less than four of the PA symptoms listed above, it is referred to as a “limited panic attack.”

Recent research has shown that, contrary to previous beliefs, there are not significant differences in people who are “early peakers” (symptom severity reaches highest level prior to 10 minutes) and “late peakers” (those who have highest severity after 10 minutes). This is reflected in the proposed changes for DSM-5, as discussed below. Having a PA actually puts one at an increased risk for developing other anxiety disorders, even though they are relatively common (see “Epidemiology” below).

There are three types of PA: a) unexpected or uncued panic attacks, b) situational or cued panic attacks, and c) situationally predisposed panic attacks. Unexpected or uncued are PA where the individual cannot link the onset to specific situational trigger. In contrast, a situational or cued PA occurs either in anticipation of or exposure to a specific trigger (internal or external). Finally, situationally predisposed PA are similar to a cued PA, but a person may be exposed to the triggering stimuli and not have a PA.

Child vs. Adult Presentations

While children can experience panic attacks, it is fairly rare. Instead, rates of reported PA begin to increase sharply during the middle teenage years and then decline rapidly starting again at age 50. Presentation does not appear to differ among age groups, although adolescents have been found to be more reluctant to discuss PA symptoms, worrying that they may represent some sort of severe medical problem.

Gender and Cultural Differences in Presentation

Studies show that more women than men experience panic attacks, at a ratio of 2:1. Culturally, panic attacks can be seen in every ethnicity and social class. However, some studies reveal that there are differences in how the symptoms are expressed compared to Caucasians and Europeans. For example, paresthesias and fear of dying is more common among African Americans, while trembling occurs to a higher degree in Caribbean Latinos. Dizziness is a predominant symptom among several East Asian groups, with fear of dying seen more in Arabs. Finally, depersonalization, derealization, and loss of control are more often reported by Puerto Ricans than Caucasians.


Almost a third of the U.S. population, 28.3%, will have at least a single panic attack at some point in their life. The overall population 12-month rate is much lower, at 11.2%, but much higher in the college population, where over 22% of students report having a PA in the past year. About 3 to 4% of adults suffer from chronic, repeated panic attacks but do not meet the DSM-IV criteria for panic disorder.


DSM-IV-TR Criteria

Agoraphobia is not a codable disorder. Code the specific disorder in which the Agoraphobia occurs (e.g., Panic Disorder With Agoraphobia or Agoraphobia Without History of Panic Disorder)

  1. Anxiety about being in different places or situations from which escape might be difficult (or embarrassing), or in which help may not be available in the event of having an unexpected or situationally predisposed panic attack or panic-like symptoms. Agoraphobia fears typically involve characteristic clusters of situations that include being outside the home alone; being in a crowd or standing in a line; being on a bridge; and traveling in a bus, train, or automobile. NOTE: Consider the diagnosis of a specific phobia if the avoidance limited to one or only a few specific situations, or a social phobia if the avoidance is limited to social situations.
  2. The situations are avoided (e.g. travel is restricted) or else are endured with marked distress or with anxiety about having a Panic Attack or panic like symptoms, or require the presence of a companion.
  3. The anxiety or phobic avoidance is not better accounted for by another mental disorder, such as social phobia (e.g. avoidance limited to social situations because of fear of embarrassment), specific phobia (e.g. avoidance limited to a single situation like elevators), obsessive-compulsive disorder (e.g. avoidance of dirt in someone with an obsession about contamination), post-traumatic stress disorder (e.g. avoidance of stimuli associated with a severe stressor), or separated anxiety disorder (e.g. avoidance of leaving home or relatives).
  4. At least four of the following symptoms developed during at least one of the attacks:
    1. Shortness of breath or a smothering sensation
    2. Dizziness, unsteady feelings, faintness
    3. Palpitations or accelerated heart rate
    4. Trembling or shaking
    5. Sweating
    6. Choking
    7. Nausea or abdominal distress
    8. Depersonalization or derealizational
    9. Flushes, hot flashes, or chills
    10. Chest pain or discomfort

Associated Features

As noted above, the DSM-IV does not classify Agoraphobia (AG) as its own, distinct disorder, instead seeing it in the context of Panic Disorder. This is in sharp contrast to the ICD-10, which classifies AG as a distinct disorder. Regardless of the nosology, many agoraphobic people have fears of leaving their homes, resulting in their ability to perform normal everyday activities being severely limited. The principal symptom of AG is a fear that a panic attack will occur when the individual is in some sort of inescapable situation (e.g., crowds, tunnels, open spaces) and leave them helpless or embarrassed, even if they have never had a PA. As a result, the individual will try to avoid these situations unless there are security measures, such as a spouse or friend with them.

While most people who have PA do not develop AG, the chance to do so tends to increase with the history and frequency of them. Intriguingly, population based studies show that between 46-85% of people with AG have not actually had a full-blown PA, although this number is much lower in clinical samples (0-31%). In addition, AG is not only seen with Panic Disorder, but can be comorbid with a number of Axis I conditions. Almost 78% of people with AG qualify for at least one other anxiety disorder (phobias and GAD being the most common), while 64% are diagnosed with comorbid mood disorders and over 31% have substance abuse or dependence problems. It is also not uncommon for people diagnosed with Axis II disorders, particularly avoidant and dependent personality disorders.

Child vs. Adult Presentation

Although AG usually has a first onset between 23 to 29 years, younger children and older adults can also develop it. When children develop AG, there tend to be more physical symptoms reported, so a diagnosis of an anxiety disorder may not be considered at first. Adults who are diagnosed with this disorder are commonly afraid of a future PA in public, and are therefore afraid of the attack itself occurring. Children, though, do not necessarily have the cognitive ability to project that far in the future, and instead may show avoidance of certain activities without a clear reason for doing so.

Gender and Cultural Differences in Presentation

There are approximately 50% more females than males that experience AG during their lifetime (1.6% vs. 1.1%), although 12-month rates are very similar (0.9% vs. 0.8%). There is some data to suggest that cultural perceptions of females is highly influential on AG, as cultures where females are viewed as more submissive and dependent on males show higher rates of AG.

Epidemiological study rates vary greatly across national studies, from a low of 0% in urban Chinese to a high of 4.8% in South Africans. In the U.S., Caucasians tend to show lower rates than minority groups, with Puerto Ricans displaying very high rates (6.0%). This is not consistent across all studies, though as some find similar rates for all groups. Minority groups do appear to have an earlier age of onset than Caucasians, as well as showing decreases in prevalence with age.


Despite not being an official diagnosis in DSM-IV, best estimates are that AG has a lifetime prevalence of 1.3% in the general population. Rates of 12 month prevalence were only slightly lower at 0.9%. Other studies have found a point prevalence rate of 0.8% for panic attacks occurring with AG. Rates do not tend to decrease steadily with age, but instead show a pattern of decreasing slightly from 18-29 year olds (1.0%) to 30-44 year olds (0.8%), the increasing until age 59 (1.2%), and finally greatly decreasing afterward (0.4%).

Panic Disorder (PD) with Agoraphobia (w/ AG) OR without Agoraphobia (w/o AG)

DSM-IV-TR Criteria

  • A. Both 1 and 2:
  1. Recurrent, unexpected panic attacks
  2. At least one of the following:
  • i. Persistent concern about having additional attacks
  • ii. Worry about the implications of the attack or its consequences (e.g. losing control, having a heart attack, “going crazy”)
  • iii. A significant change in behavior related to attacks.
  • B. Absence of agoraphobia (PD w/o AG) OR presence of agoraphobia (PD w/ AG)
  • C. The panic attacks are not due to the direct physiological effects of a substance (e.g. hyperthyroidism).
  • D. The panic attacks are not better accounted for by another mental disorder such as social phobia (e.g. occurring on exposure to a feared social situation), obsessive-compulsive disorder (e.g. on exposure to dirt in someone with an obsession about contamination), post-traumatic stress disorder (e.g. in response to stimuli associated with a severe stressor), or separation anxiety disorder (e.g. in response to being away from home or close relatives).

Associated Features

Many individuals with PD report having occasional or constant feelings of anxiety that are not focused on any specific event or situation, while others become apprehensive about what might happen during routine activities. The negative impacts of PD are myriad. First, demoralization is common as the person becomes discouraged, ashamed and unhappy about the difficulties of living everyday life. They blame themselves, thinking that they are lacking in “character” or “strength”. Missing school or work because of medical visits is common, and can lead to dropping out of school or job loss. People with PD have very high rates of medical visits, procedures, and laboratory tests, both compared to the general public and persons with other anxiety disorders. They consistently report dissatisfaction with their medical treatment, and physicians rate people with PD as more difficult to care for. Medical visits over a 12-month period are especially common to the ER (43.9%), urgent care (48.8%), cardiologist (46.3%), and family practitioners (46.3%).

Comorbidity is higher for people who have combined PD and AG, compared to those with PD alone. In PD w/ AG, over 93% meet criteria for another anxiety disorder, while the overlap is only 66% in PD w/o AG. Similar differences are seen in comorbid mood disorder (73% vs. 50%) and substance abuse problem (37% vs. 27%) rates. Depression is a very comorbid, but can either precede (a third of cases) or occur after PD (two thirds of cases).

Child vs. Adult Presentation

While both children and adults can have PD, it tends to be very rare before puberty, gradually increases until middle age, and then decreases again. Youth and adults experience similar symptoms (trembling, breaking out in a sweat, heart, palpitations, nausea, and so on), although adolescents report worrying about subsequent PA less than young adults. It is crucial to note that some researchers have found that children who are later highly prone to developing PD display much higher rates of separation anxiety than same-age peers. Such children also tend to show other anxious behaviors, such as behavioral inhibition and anxiety-sensitivity.

Gender and Cultural Differences in Presentation

PD w/o AG is two times more common in women than in men, while PD w/ AG is three times more common in women. This gender gap begins to be observable by early adolescence, and just continues to widen with age. It is important to note that some cultural or ethnic groups restrict women from being in the public life, and that this should be distinguished from agoraphobia. PD appears to be more debilitating to women than it is for men, as females tend to become more depressed, rate higher on fear tests, and spend more time avoiding social situations. Men are also more likely to hold down a steady job.

In the U.S., minorities tend to have lower rates of PD than Caucasians, although Native American groups have been found to have higher rates. Cross-culturally, lower rates of PD are seen outside the U.S., even in European samples. For instance, studies in the Ukraine have found rates of 1.27% and 1.94% for 12-month and lifetime, respectively. Germany had slightly higher 12-month rates (1.8%), but still lower than the U.S, while Australia was even lower (1.1%). In Japan ( 0.5% for 12-month), South Korea (0.2%), China (0.2%), PD is extremely rare, with similarly low rates in other non-Western countries (0.6% in Mexico, 0.8% in South Africa).

As noted earlier, certain symptoms of PA are more or less frequently seen in certain cultural groups. Directly related to PD are several culturally-bound disorders. For example, khyâl attacks in Cambodia are characterized by a mix of PA and culture-specific symptoms including tinnitus and neck soreness w/ dizziness. Ataque de nervios (“attack of nerves”) among Latin Americans and trunggio (“wind”)-related attacks in Vietnam also appear to be culturally-relevant variations on PD.


Panic disorder (with or without agoraphobia) has a lifetime prevalence rate of 4.7% in the U.S., with a 12-month rate of 2.7%. Both lifetime and one year rates show an upside down U curve of distribution, with lower rates for 18-29 year olds (4.2% and 2.8%) and those over 60 years old (2.1% and 0.8%) compared to age groups of 30-44 (5.9% and 3.7%) and 45-59 (5.9% and 3.1%). Rates for children and adolescents are very low, likely due to the lack of development of cognitive abilities such as self-monitoring and metacognition.

In treatment-seeking clinical settings, the prevalence rates for panic disorder are noticeably higher, with some studies finding as high as 30%. In general medical settings, almost 10% of people referred for a mental health consultation were diagnosed with panic disorder. In specialty medical settings such as vestibular, respiratory, and neurology clinics the prevalence rates vary from 10% and 30%, while in cardiology clinics rates as high as 60% have been found. In community samples, a third to a half of individuals diagnosed with PD have AG as well. There is a much higher rate of PD w/ AG encountered in clinical samples than without AG.


Genetic and family studies have found that both biology and environment are strong contributors to the development of both PD and AG. Twin studies have revealed that there is a genetic link to the development of PD. Individuals with a first degree relative suffering from panic disorder are eight times more likely to develop panic disorder than people without. If onset is before age 20, though, the individual’s risk increases to 20 times as likely to develop PD. Heritability for PD seems to be around 45%, with shared (10%) and unshared (45%) environments contributing significantly. For AG, heritability estimates are slightly higher, at around 60%. The temperament trait of behavioral inhibition (BI) is highly implicated in the development of both, and parents with PD or AG are more likely to have children who are behavioral inhibited. This, however, holds true across all anxiety disorders. For PD and AG specifically, anxiety sensitivity (believing anxiety is harmful and bad) is the key trait. Furthermore, we know that early trauma and maltreatment are risk factors for developing both later, and that development may be meditated by the presence of BI.

Neurologically, panic attacks are closely linked to amygdala function. The amygdala is the anxiety “way-station” that mediates incoming stimuli from the environment (thalamus and sensory cortex) and stored experience (frontal cortex and hippocampus). As such, it impacts the anxiety and panic response by stimulating various brain areas responsible for key panic symptoms based on both internal and external stimuli and past events. In particular, the periaqeductal gray in the midbrain could be especially important for mediating panic symptoms. Pharmacology and CBT can effectively treat PA, but they act on different systems. While pharmacology can target all areas of the above described system, effecting amygdala and frontal-lobe interpretation of stimuli or output effects, CBT impacts the frontal-lobe areas, especially in the medial prefrontal cortex, which is known to inhibit input to the amygdala.

Psychologically, the major factor in the development of PD and AG seems to be anxiety sensitivity. This is the belief, which could be acquired in any number of ways, that anxiety could cause severe physical, social, and psychological consequences that extend beyond any discomfort during a PA. Examples of means of acquisition are direct experience, vicarious observations, information transmission, and parental reinforcement. Essentially, a person develops a “fear of fear.” This model posits that an individual who has a PA or PA symptoms may, through the process of interoceptive conditioning, learn to fear any change in physiological state that could signal the onset of panic. As such, they pay more attention to physical and bodily changes than most individuals, which ironically puts them at a higher risk of having panic attacks. For example, if you take the stairs to the third floor of a building, you may notice that you are flushed, breathing more heavily than usual, and sweating. For a person prone to PA, these signs would be seen as indicative of an oncoming PA rather than just being a sign of tiredness or being out of shape. This would make them nervous about the chance of having a PA, which activates the sympathetic nervous system and in turn makes it more likely they will actually have a PA. This can lead both to the avoidance of situations likely to trigger such sensations (AG) as well as a high likelihood of having repeated PA (PD).

Empirically Supported Treatments

Pharmacology meta-analyses for PD and AG show similar medium to large effect sizes (0.48-0.55) for both the tricyclic antidepressants (TCA) and SSRI classes of drugs. Benzodiazipines (BZD) are also effective at reducing incidence of PA, but they and the TCA are prescribed less than SSRI due to side-effect reasons. In treatment-refractory patients, SSRIs can be supplemented with BZD, or MAOIs can be used. Again, these are not front-line treatments due to their larger side effect profile. Clients should be made aware that there is a substantial (25-50%) relapse rate within 6 months when medications are discontinued, though. This may be partially due to the high potential for withdraw symptoms (from any medication) to become interoceptive cues for a PA, thus reversing the progress made while on the medication.

Cognitive-behavioral therapy is the most well studied and validated treatment for PD, with effect sizes of 0.9-1.55. It has been found to be equally effective in individual or group format, as well as in standard (14-18 meetings) or brief (6-8) sessions. As with all CBT treatments for anxiety, though, there is a massive underutilization due to lack of properly trained mental health professionals. CBT for PD emphasizes psychoeducation about panic symptoms, cognitive restructuring focusing on reducing anxiety sensivity, interoceptive exposure to feared bodily sensations, and in vivo exposure to the previously avoided and feared situations. Similar to other treatments discussed above, retraining of breathing to help patients cope with their panic and anxiety has been found to be unnecessary. CBT for AG is very similar, but with a smaller focus on interoceptive exposures and greater emphasis on in vivo exposure to feared situations.

Although both medical and psychotherapeutic treatments are effective alone, CBT has a stronger initial effect size and yields larger long-term effect sizes (0.88-0.99 vs. 0.40-0.55). Research has found no benefit for combining the two, as controlled trials show that CBT alone is as effective as the combination. As with several other disorders, researchers have also examined self-guided therapies based on CBT, using both bibliotherapy and computer-mediated models. Results are generally supportive, with one study finding similar one-year effect sizes for 10 session live CBT (0.93) and 10 module internet self-help treatment (0.80).

Proposed DSM-5 Revisions

As mentioned earlier, there are major changes proposed for these disorders in DSM-5. First, Agoraphobia is recommended to be classified as a distinct disorder. There are three primary lines of evidence that have supported this change: psychometric evaluations supporting the construct of agoraphobia alone, epidemiological investigations of prevalence, and the impact AG has on clinical course and outcome. This change would also bring the DSM-5 and the ICD (International Classifications of Disease) more into alignment, as Agoraphobia is already a separate disorder in that system. As such, Panic Disorder would no longer have the “with or without Agoraphobia” included in the diagnosis. It is also proposed that a specifier be added to all the anxiety disorders that would allow “with panic attacks” to be noted in the diagnosis, given the high rate of PA across the class.

In addition the disorders reviewed above, there are three others included in the DSM-IV anxiety disorder section. The first is Anxiety Disorder Due to a General Medical Condition. As expected from the name, this is where a person experience anxiety problems as a direct result of a medical problem, such as as hyperthyroidism, hypothyroidism, vitamin deficiencies, or brain lesions. People with cardiovascular problems, endocrine disorders, neurologic conditions, peptic ulcers, diabetes, and respiratory conditions are also at risk of developing anxiety as a result of their condition.

The second is Substance-Induced Anxiety Disorder, which is the direct result of either intoxication or withdrawal from a psychoactive substance. Common substances causing such problems include alcohol, cocaine, sedatives, hypnotics, and anxiolytics. The final is Anxiety Disorder Not Otherwise Specified. This is a “catch-all” category, where a person displays prominent anxiety symptoms or avoidance, but does not meet full criteria for any of the other, specific disorders.

Key References

Bakker, A., van Balkom, A.J., Spinhoven, P. (2002). SSRIs vs. TCAs in the treatment of panic disorder: a meta-analysis. Acta Psychiatric Scandinavia, 106(3), 163-167.

Craske, M.G., Kircanski, K., Epstein, A., Wittchen, H-U., Pine, D.S. et al. (2010). Panic Disorder: A review of DSM-IV Panic Disorder and proposals for DSM-V. Depression and Anxiety, 27, 93-112.

Deacon, B., Lickel, J., & Abramowitz, J. S. (2008). Medical utilization across the anxiety disorders. Journal of Anxiety Disorders, 22(2), 344-350.

Kiropoulos, L.A., Kleina, B., Austina, D.W., Gilsona, K., Piera, C., et al. (2008). Is internet-based CBT for panic disorder and agoraphobia as effective as face-to-face CBT? Journal of Anxiety Disorder, 22, 1273-1284.

Schmidt, N.B., Keough, M.E. (2010). Treatment of panic. Annual Review of Clinical Psychology, 27(6), 241-256.

Roy-Byrne, P.P., Craske, M.G., & Stein, M.B. (2006). Panic disorder. Lancet, 368, 1023-1032.

Wittchen, H-U., Gloster, A.T., Beesdo-Baum, K., Fava, G.A., & Craske, M.G. (2010). Agoraphobia: A review of the diagnostic classificatory position and criteria. Depression and Anxiety, 27, 113-133.