Panic disorder occurs frequently in the general population. Four recent studies estimated that 1.6 to 2.9 percent of women and 0.4 to 1.7 percent of men have panic disorder (Crowe et al. 1983; Myers et al. 1984; Uhlenhuth et al. 1983; Weissman et al. 1978). These prevalence estimates are probably quite conservative because the panic disorder patient's tendency to somatize frequently leads to underdiagnosis (Katon et al. 1987a).
Several studies found that patients with panic disorder are overrepresented within the medical care system. Katon and colleagues (1986) randomly assessed 195 primary care patients, age 17 years and older, with a structured psychiatric interview, the NIMH Diagnostic Interview Schedule (DIS) (Robins et al. 1981). A total of 6.5 percent of patients met DSM-III criteria for panic disorder alone, and 6.5 percent met criteria for major depression and panic disorder. Finlay Jones and Brown (1981) found, using a structured psychiatric interview, that 17 percent of 164 female primary care patients suffered from anxiety neurosis, with 8 percent suffering from anxiety neurosis alone and 9 percent having anxiety neurosis and major depression.
Two studies examined the prevalence of severe anxiety in primary care patients on self-rating scales of anxiety (Linn and Yager 1984; Zung 1986). These scales are quite sensitive to panic disorder, but not very specific; for example, cases of major depression, alcohol abuse, and generalized anxiety may also score high on these scales. These studies determined that 20 percent of 739 primary care patients (Zung 1986) and 12 percent of 95 internal medicine patients (Linn and Yager 1984) scored in the moderate to severe anxiety range on the self-rating scales.
Lower estimates of the prevalence of panic disorder were found by Von Korff and colleagues (1987) using the DIS (Robins et al. 1981) in a large, primary care epidemiologic study of predominantly middle-age to geriatric-age internal medicine patients; they determined that panic disorder occurred in 1.4 percent of patients.
In clinical samples, panic disorder is significantly more common in females than males, with a ratio between 2.5 and 3.0 to 1 (Sheehan 1983). However, in the NIMH Epidemiologic Catchment Area study of the prevalence of mental illness in several large U.S. cities, panic disorder was consistently more common in females, but only in one of the three large cities did the difference reach statistical significance (Robins et al. 1984). The predominance of females in clinical samples of patients with panic disorder may, in part, reflect the tendency of females to seek health care more frequently than males do.
In clinical populations, the onset of panic disorder is generally between the ages of 17 and 30 years (mean age 22.5 years) (Sheehan et al. 1981). In the ECA study, the highest 6-month prevalence was in the 25 to 44 age group (Myers et al. 1984).
Panic disorder tends to be a relapsing, remitting illness. Wheeler and colleagues (1950) completed a 20-year followup of 171 patients diagnosed as having neurocirculatory asthenia. The criteria for neurocirculatory asthenia were roughly similar to the DSM-III criteria for panic disorder. Of the 171 patients, 73.3 percent had mild symptoms with no disability, 15 percent had moderate to severe symptoms, and 11.7 percent were well. Noyes and Clancy (1976) studied 57 patients with panic disorder and found that 16 percent were unimpaired, 51 percent were mildly impaired, and 32 percent were moderately to severely impaired at 5-year followup.
Followup studies of patients with agoraphobia have also been performed. Many agoraphobics’ avoidance behavior decreases with time or behavioral treatment, but these patients frequently continue to have panic attacks. Roberts (1964) followed 41 agoraphobic housewives for periods of 1 to 16 years and found that 55.3 percent had improved, although virtually all the improved still had some residual symptoms. Emmelkamp and Kuipers (1979) reported that 75 percent of agoraphobics treated with behavioral therapy continued to do well at 4-year followup. Marks and Herst (1970), however, in a large sample of agoraphobic patients, found that only 20 percent reported periods of complete remission after their phobias began.
Overall, in the followup studies of patients with panic disorder and agoraphobia, approximately 50 to 70 percent of patients show some degree of improvement (Reich 1986). Total remissions are much less common, with some of the studies suggesting that 50 percent of patients have some disability and 70 percent are still symptomatic 20 years after the initial diagnosis.
In a recent study (Katon et al.l987b), 25 primary care patients with past histories of panic attacks had significantly higher scores on psychologic tests of depression, anxiety, and phobic anxiety as well as significantly more phobias than 78 controls without any history of anxiety attacks. These data and studies suggest that untreated patients with panic disorder do improve without treatment; however, they are often left with residual symptoms such as avoidance behavior, multiple phobias, and higher psychologic distress, especially anxiety and depression.
Panic disorder frequently causes marked avoidance behavior and phobias; thus, it is important to review studies of patients after effective treatment in order to study and differentiate state and trait personality characteristics. Reich and Troughton (1988) found that approximately one-third of patients who had recovered from panic disorder met criteria for either a dependent, avoidant, or compulsive personality disorder. Avoidant personality disorder alone was present in about one-fifth of the patients with panic disorder. In clinical terms, patients with these personality characteristics are more socially insecure, easily hurt, and dependent on others in social situations.
Tyrer and colleagues (1983) found that the presence of a personality disorder was the best predictor of chronicity of anxiety neurosis (the British term for panic disorder). Patients with a personality disorder and panic attacks are less likely to respond well to psychopharmacologic or behavioral treatments (Mavissakalian and Hamann 1987). Specific personality traits associated with poor treatment response include impulsivity, interpersonal sensitivity, subordination of one's own needs, and social and occupational ineffectiveness (Cowley and Roy-Byrne 1988). Intensive early psychiatric intervention may be especially important in patients with personality disorders who develop panic disorder.