Panic has not always been recognised as an exclusively psychiatric condition. Research in this area continued along separate medical and psychological axes until 1980, when the development of Diagnostic and Statistical Manual (DSM)-III criteria established the overall concept of panic disorder. The lifetime prevalence of DSM-III panic disorder and repeated panic attacks, defined as the average of individual estimates from six studies, are 2.7% and 7.1% of the general population, respectively. Females are almost twice as likely as males to suffer panic disorder, and about seven times as likely to suffer repeated panic attacks. Overall, panic disorder or panic attacks occur in up to one in ten of the general population. The prevalence of panic disorder and panic attacks, their associations with other conditions, and their time courses have been investigated in a prospective epidemiological study in Zurich, Switzerland, in which 591 individuals were followed for 15 years. The validity of panic disorder and panic attacks as genuine psychological phenomena are attested to by their positive associations with a family history of panic disorder, elevated risk of suicide, lifetime treatment for psychiatric disorders, and especially treatment with prescribed medication and substantial work and social impairment. Strong comorbidity exists between panic states and other psychiatric conditions, including depression (major depression, bipolar disorder and recurrent brief depression), agoraphobia, social phobia, specific phobia, and obsessive-compulsive disease. A lower degree of comorbidity is seen with alcohol and tobacco dependence. Comorbid conditions usually precede panic, except for alcohol abuse, which is usually secondary to episodes of panic. The prognosis of panic states is often optimistic, and chronic disease is present in less than half of sufferers. Both panic disorder and repeated panic attacks are common, serious and disabling conditions. Effective diagnosis and treatment of repeated panic attacks and panic disorder are of equal importance.

The different coping responses to three types of aversive events - future threats, acute events and chronic stress — may be modulated by 5HT projections. Dysfunction in these coping mechanisms could cause, respectively: generalised anxiety disorder, panic, and depression. This theory proposes that dorsal raphe nucleus projections to 5HT2 and 5HT1D receptors mediate anticipatory anxiety and normally motivate avoidance of threats. The brain aversion system may be held in check by dorsal raphe nucleus 5HT projections to mediate behavioural inhibition during anticipatory anxiety. Proximal aversive stimuli such as pain and asphyxia elicit the fight-flight reflex mediated by the amygdala-hypothalamic-periaqueductal grey brain aversion system. Panic attacks may thus be due to spontaneous activation of this system. Median raphe projections to 5HT1A receptors have been implicated in adaptation to chronic stress-resilence. There is good evidence that 5HT1A function breaks down in depression and causes the depressed state. Experimental tests of this theory suggest that viewing anxiety, panic, and depression as dysfunctions in neurochemically and anatomically specific coping systems is heuristically useful.

Panic disorder is a common condition. Epidemiological studies throughout the world consistently indicate that the lifetime prevalence of panic disorder (with or without agoraphobia) is between 1.5% and 3.5%. Panic disorder shows substantial comorbidity with other forms of mental illness. Major depressive disorder occurs in 50 to 65% of individuals with panic disorder and there is considerable cross-sectional and longitudinal comorbidity with recurrent brief depression and dysthymia. Phobic anxiety disorders, most notably social phobia and generalised anxiety disorder, commonly occur with panic disorder, especially in individuals with more severe agoraphobia. Approximately 35 to 50% of individuals with panic disorder in community settings also have agoraphobia. Panic disorder also shows significant comorbidity with physical illness. Compared with individuals without or with some other psychiatric diagnosis, patients with panic disorder have an increased risk of suffering from multiple medically unexplained symptoms and are associated with high use of medical services and increased mortality from both cardiovascular and cerebrovascular disease.

Panic disorder is widespread in Australia, often in combination with other psychiatric conditions such as agoraphobia or major depression. Pharmacotherapy for panic disorder in Australia commenced with benzodiazepines, and later progressed to tricyclic antidepressants (TCAs) and monoamine oxidase inhibitors (MAOIs). More recently, treatment has moved towards use of the selective serotonin reuptake inhibitors (SSRIs), which are effective and better tolerated. Paroxetine is the first drug of this class to receive approval for treatment of panic disorder in Australia.

Panic disorder is a chronic condition for many patients and can be socially, emotionally and occupationally disabling. Until recently, clomipramine and alprazolam were the only drugs approved for its treatment. While widely used in the US and Europe, both belong to drug classes (tricyclics and benzodiazepines) with well-recognised side effects that can be problematic and thus limit their use. Recently, paroxetine became the first selective serotonin reuptake inhibitor to receive approval and licensing for panic disorder.

The short- and long-term efficacy and tolerability of paroxetine in panic disorder has been established in clinical trials of almost 1,000 patients meeting Diagnostic and Statistical Manual (DSM)-IIIR criteria for panic disorder, with or without agoraphobia. In a 12-week double-blind study of 120 panic patients receiving standardised cognitive therapy, paroxetine was significantly more effective than placebo in reducing panic attack frequency. In a 12-week placebo-controlled comparison in 367 panic patients, paroxetine was at least as effective as clomipramine and better tolerated. There was also some evidence that paroxetine had an earlier onset of action than clomipramine.

A 9-month extension of the placebo-controlled comparison with clomipramine showed that the efficacy of paroxetine and clomipramine is maintained when treatment is continued into the longer term. In a relapse prevention study, 105 responders to 3 months' treatment with paroxetine or placebo were re-randomised, either to continue existing treatment or to receive placebo for 3 months. Only 5% of patients who continued to take paroxetine experienced a relapse compared with 30% of those who switched to placebo (P = 0.002). Paroxetine was generally well tolerated. In the short-term trials, the frequency of withdrawals due to adverse events (7.3%) was lower than that for placebo (11.4%) or clomipramine (14.9%). In the longer term, the dropout rate due to adverse events increased in the clomipramine group (19.0%) but was unchanged in the paroxetine group (7.4%). Since most patients with panic disorder will require prolonged treatment, the long-term tolerability of paroxetine and its lack of potential for dependence are important advantages that will encourage good compliance with treatment and improve the quality of life of patients.