Panic Disorder and Agoraphobia
According to DSM-IV (APA, 1994) panic attacks are defined as sudden spells of unidentified feelings consisting of at least four out of 13 symptoms as palpitations, chest strains, sweating, shortness of breath, feelings of choking, trembling, nausea, dizziness, paresthesias, chills or hot flushes, depersonalisation or derealisation, fear of dying or losing control. Although having panic attacks does not imply that the diagnosis of panic disorder can be made and isolated panic attacks are not diagnosed as a disorder, they are often associated with substantial morbidity and do have some clinical significance (Klerman et al., 1991).
In order to make a diagnosis of panic disorder, additional criteria are that these attacks at least once have been unexpected, followed by at least one month of fearful expectation or concern about the consequences of an attack. In the DSM-IV criteria as to the frequency of the attacks (in DSM-III-R, APA 1987, three attacks in a period of three weeks) are abandoned.
Panic disorder is frequently followed (or accompanied) by agoraphobia. Agoraphobia in DSM-IV is defined as (a) fear of being in places or situations from which escape might be difficult or help might not be available; (b) these situations are avoided or endured with marked distress or the patient needs a companion; and (c) the fear is not better explained by another mental disorder.
As such, isolated panic attacks (as defined above) are quite frequent, estimated at 7 - 9% for lifetime prevalence rates, although rather heterogeneous figures come from different countries (Pe´lissolo and Le´pine, 1998). The ECA study on panic by Eaton reports 15% of all respondents have had a panic attack in a lifetime, 3% in the past month, while 1% of the subjects meet criteria for panic disorder in the past month (Eaton et al., 1994). In a survey among 1035 adolescents in Bremen, Germany, 18% of participants reported having had at least one panic attack (in a lifetime), with 0.5% of them meeting DSM-IV criteria for panic disorder (Essau et al., 1999). The prevalence of panic disorder, assessed with diagnostic criteria and structured interviews, has been found in the majority of surveys to have lifetime rates between 1.5% and 2.5%. Twelve-month prevalence rates are generally about 1% (Pe´lissolo and Le´pine, 1998).
Although panic disorder can be diagnosed apart from agoraphobia, in clinical practice it is rare to find a patient suffering from panic disorder who did not develop agoraphobia to a certain extent. This, however, is in contrast to the results from the NCS study, where it was found that 50% of panic disorder patients report no agoraphobia (Eaton et al., 1994). Either way, it is infrequent to find an agoraphobic without a history of a panic attack. Horwath et al. (1993) have shown that epidemiological studies that used the Diagnostic Interview Schedule and lay interviewers, such as the ECA study, may have over-estimated the prevalence of agoraphobia without panic. The investigators took a sample of 22 ECA subjects diagnosed as having agoraphobia without panic attacks, and had them blindly re-interviewed.
Re-analysis showed that only one of these subjects was left with the original diagnosis, one was assigned to having panic disorder with agoraphobia, one had agoraphobia with limited symptom attacks, and the vast majority were re-appraised as specific phobias (Horwath et al., 1993).
In addition, a recent study by Wittchen et al. (1998) in a community sample of 3021 young subjects (14 to 24 years old) in Munich, Germany, addresses the relationship between panic disorder and agoraphobia. They found that lifetime prevalence of panic disorder with agoraphobia was as high as for panic disorder without agoraphobia, both being 0.8%. Post hoc clinical review of the CIDI-positive agoraphobics revealed that many respondents did not have agoraphobia, but actually suffered from specific phobia, e.g. situational phobias. This resulted in a corrected agoraphobia prevalence of 3.5% instead of the original 8.5%. However, even after this correction, the majority of respondents with confirmed agoraphobia were found not to have a prior history of panic. However, in the NCS where agoraphobia was a separate diagnosis, it was found to have a lifetime prevalence rate of 6.7%, and a one-month prevalence of 2.3% (Magee et al., 1996).
Demographics and Risk Factors
Most studies reporting on panic disorder or panic attacks consistently show higher rates for women than for men. Panic attacks are almost twice as common in women, where panic disorder ranges from 1.5 to twice as much. The age at onset of panic disorder in general lies in the mid-twenties, with hazard rates for women ranging from 25 to 35 years, for men between 30 and 45 years (Wittchen and Essau, 1993).
Marital status is a significant risk factor for panic disorder: the highest lifetime prevalence rates are found in widowed, separated or divorced subjects (Wittchen and Essau, 1993). On educational level and risk of developing panic disorder there are no consistent findings, Eaton et al. (1994) report from the NCS data a tenfold higher risk for persons with less than 12 years of education. Several studies have suggested that life events such as early parental loss or childhood abuse may enhance the risk of panic disorder, but there appears to be no specificity since this also applies to other psychiatric disorders as post-traumatic stress disorder and affective disorders.
Other supposed risk factors are smoking habits, although there is no consensus about causality or just correlation because all psychiatric patients smoke more than the general population. Breslau and Klein (1999) find evidence for smoking leading to panic disorder. Others also indicate a role for smoking in developing panic disorder (Pohl et al., 1992; Amering et al., 1999; Biber and Alkin, 1999). Another factor associated with panic disorder is the existence of pulmonary disease. All conditions that have symptoms of shortness of breath can lead to anxiety in general and panic disorder in particular (Smoller et al., 1999). It is shown that respiratory disease in childhood (especially bronchitis and asthma) predisposes to panic disorder in later life (Zandbergen et al., 1991; Verburg et al., 1995; Perna et al., 1997). In addition, chronic obstructive pulmonary disorders can lead to panic disorder in primarily lung patients (Wingate and Hansen-Flaschen, 1997), although the role of psychological factors, such as cognitive misinterpretation of bodily symptoms (shortness of breath) must be taken into account (Moore and Zebb, 1999).
Natural Course
Panic disorder and agoraphobia seem to be a chronic condition (Pollack and Smoller, 1995; Hirschfeld, 1996; Liebowitz, 1997), mostly with a fluctuating course with periods of waxing and waning (Liebowitz, 1997; Pollack and Otto, 1997). In some studies longitudinal aspects have been addressed, e.g. in a three-year follow-up only 10% were shown to be symptom-free (Noyes et al., 1990). Faravelli et al. (1995) found that only 12% of panic disorder patients after five years were in full remission. Panic disorder can be a very disabling disorder with high impact on daily life and social, personal and professional functioning and can put a great strain on quality of life (Candilis and Pollack, 1997; Candilis et al., 1999).
Comorbidity
Comorbidity of panic disorder and agoraphobia is very common, as the panic attacks are often viewed as the precipitating cause for agoraphobia to develop (Klein and Klein, 1989). From this perspective, it is unnecessary to speak of true comorbidity, because the panic disorder with agoraphobia can be considered one disease entity.
The comorbidity of panic disorder with agoraphobia is reported as ranging from 29.5% to 58.2%. In the NCS (Eaton et al., 1994) 50% of the panic disorder patients had comorbid agoraphobia. Also, findings from the study of Wittchen et al. (1998), cited above, indicate that half of patients with panic disorder also developed agoraphobic avoidance. There is also a high degree of comorbidity with other anxiety disorders, such as social phobia (20 - 75%, see Pe´lissolo and Le´pine, 1998) and generalised anxiety disorder (20%), 14% for obsessive-compulsive disorder and 6% for post-traumatic stress disorder (Goisman et al., 1994).
The Munich Follow-up Study (Wittchen and Essau, 1993) also stresses the risk for panic disorder patients of developing comorbid other psychiatric disorders. In more than half of the cases, some comorbid disorder will develop over time. Major depression (Merikangas et al., 1996) is the most frequent comorbid diagnosis, 30 - 60% of panic disorder patients suffer from a depressive disorder (Weissman et al., 1997). Most studies report a concurrent prevalence rate of about 30%, with lifetime prevalence of depression occurring in about 60% (Wetzler and Sanderson, 1995).
Also, the NCS data have shown a lifetime prevalence of depression in panic disorder patients of 55.6% (Kessler et al., 1998a). There are divergent views about the order of onset in comorbid panic disorder and depression. In the ECA analysis it is shown that onset of panic is first in about 30% of cases, onset of depression precedes in another 30%, and simultaneous onset occurs in 40% (Regier et al., 1998). Also, some clinical studies have shown similar figures on the order of onset in comorbid panic disorder and depression (Lydiard, 1991; Stein et al., 1990). Other clinical studies report that the onset of panic disorder was first in about two-thirds of comorbid cases (Hunt and Andrews, 1995). The clinical significance of comorbidity lies in the severity of symptoms at the outset (Andrade et al., 1994).
Substance abuse (alcohol, drugs and medication) is also a common comorbid disorder, in 36% of cases, according to ECA data (Regier et al., 1990a). Most frequently, these abuse disorders are supposed to be secondary to the panic disorder, and can be interpreted as self-medication (Marshall, 1997; Swendsen et al., 1998; Merikangas et al., 1998). Some authors, however, dispute this; Katerndahl and Realini (1999), for example, found that the majority of drug and alcohol abusers report that the abuse started before the onset of the panic attacks.
The comorbidity of panic disorder with other general medical conditions is described in a review by Zaubler and Katon (1998). It is shown that panic disorder is common in cardiac, gastrointestinal, respiratory and neurologic disorders.
