Panic Disorder

Foreword

Panic disorder (with or without agoraphobia) is one of the most common — and, for patients, disabling—psychiatric disorders encountered in general medical settings. From a recent study in a primary care practice setting, investigators estimated that the 1-month prevalence of panic disorder among primary care patients is 1.4 percent (Von Korffet al. 1987). The estimate from the community-based Epidemiological Catchment Area Study is that in any month 0.5 percent of the population will have this diagnosis (Regier et al. 1988). Because a number of medical and mental disorders may mimic symptoms similar to panic disorder, an accurate differential diagnosis is essential—but often difficult.

During the past two decades, enormous strides have been made in elucidating the dysfunctional brain mechanisms underlying panic disorder, specifying empirically based diagnostic criteria, and developing effective treatments. All of these developments have given knowledgeable primary care physicians and psychiatric specialists alike newfound power to recognize and ameliorate the course of this once-daunting disorder. However, many practitioners, especially those trained before these developments occurred, may have had little opportunity to stay abreast of all these advances. Further, only a limited amount of the recent psychiatric research literature related to panic disorder has been translated into practical advice for primary care physicians.

This volume was written to fill that knowledge gap by providing busy primary care practitioners (and their mental health consultants) with practical, state-of-the-art diagnostic and treatment techniques for panic disorder. It is intended to aid clinicians in recognizing and treating patients with this common mental disorder and in identifying when psychiatric consultation or referral is required.

This volume includes up-to-date information on the natural course, clinical characteristics, epidemiology, psychobiology, and treatment of panic disorder. Dr. Katon cogently discusses the challenge of diagnosing panic disorder, the medical illnesses associated with it, and the bases for making a differential diagnosis. In addition, he suggests ways to provide followup care for these patients in primary care settings and offers guidelines for psychiatric referral. He has also thoughtfully included a list of articles and books related to panic disorder that should prove particularly useful for patients and physicians.

For the National Institute of Mental Health, which sponsored the development of this volume, it represents an important aspect of our research mission: closing the gap between the research laboratory and the clinician’s office.

Because primary care practitioners are the main source of ambulatory care for the mentally ill, we are particularly eager to assure that clinically relevant results of mental health research reach such clinicians quickly, in a form they will find readily accessible and useful in day-to-day practice. I believe that these goals are admirably met in this volume.

Preface

The following case is presented to intrigue and interest the primary care physician and to demonstrate that panic disorder is associated with many common patient symptoms seen in clinics every day.

Mr. W was a 22-year-old graduate student who presented to his family physician with acute episodes of chest pain and shortness of breath Mr. W had recently moved to a new city, where he had few friends or family, to begin graduate school. His physician performed a careful physical examination and electrocardiogram, which were both negative. Then he brought up the subject of stress and anxiety, and the patient readily agreed that both the recent relocation to a new city and graduate school were quite stressful. He was referred to psychotherapy in the student health service. Mr W’s symptoms continued, and over the next 2 years, he made about 20 visits to his primary care physician for various somatic symptoms. Labortory tests and physical examinations were all negative. In his second year of graduate school, he developed acute episodes of epigastiric distress that often awakened him from sleep and were relieved by antacids. During this time, the patient lost 20 pounds. An upper gastrointestinal series proved negative, and the patient refused endoscopy.

Finally, after 2 years of symptoms, the patient was referred for psychiatric consultation. He reported that his episodes of chest pain and shortness of breath were often accompanied by tachycardia, sweating, tremulousness, and a feeling that he was going to die. His epigastric distress frequently followed these acute episodes. Since the episodes began, he had become increasingly fearful of social situations, and he now avoided going to parties, to the theater, or to restaurants. The patient was diagnosed as meeting DSM-III criteria for panic disorder, and he was started on imipramine 25 mg at night, which was gradually increased to 100 mg over 2 weeks. His acute episodes of autonomic symptoms decreased over 3 to 4 weeks, and his avoidance of social situations gradually decreased over 3 months. Over the next 3 years, the patient only visited his family physician twice for yearly physical exams.

The above case emphasizes many points that are covered in this monograph:

• The association between panic disorder and stressful life events
• The frequent association between panic disorder and hypochondriacal complaints and high utilization of medical care
• The frequent misdiagnosis of panic disorder in primary care
• The association of panic disorder with psychophysiological illness such as peptic ulcer disease
• The association of panic disorder with phobic behavior
• The rapid amelioration of panic disorder with specific psychopharmacologic and/or cognitive-behavioral treatment
• The decrease in hypochondriacal behavior and high medical utilization with effective treatment of panic disorder

Acknowledgments

This publication benefited from the critical review provided by the following experts in the mental health field: Donald F. Klein, M.D., New York State Psychiatric Institute; R. Bruce Lydiard, Ph.D., M.D., Department of Psychiatry and Behavioral Sciences, Medical University of South Carolina; David A. Katerndahl, M.D., Department of Family Practice, University of Texas; James T. Marron, M.D., Department of Family Practice, University of New York; Thomas W. Uhde, M.D., Division of Intramural Research, National Institute of Mental Health; David H. Barlow, Ph.D., Department of Psychology, State University of New York; and Jack D. Maser, Ph.D., Division of Clinical Research, National Institute of Mental Health. The Institute gratefully acknowledges their reviews and comments.

Overview

Panic disorder with and without agoraphobia is a debilitating condition that will afflict at least 1 out of every 75 people in this country and worldwide during their lifetime. Panic attacks are characterized by sudden and unexpected discrete periods of intense fear or discomfort associated with shortness of breath, dizziness, palpitations, nausea, or abdominal distress. During an attack people often believe that they are having a heart attack or, alternately, that they are losing their mind. Panic sufferers often develop agoraphobia secondary to the occurrence of these unexpected panic attacks. Consequently, they begin to avoid places where they fear a panic attack may occur or where help would be difficult to obtain. If the agoraphobia becomes severe enough, a person may become housebound.

A growing body of knowledge indicates that some medications and selected psychosocial treatments are effective for panic disorder, with and without agoraphobic avoidance. Two classes of antidepressants (i.e., tricyclics and monamine oxidase inhibitors), as well as certain high-potency benzodiazepines (e.g., alprazolam, lorazepam, and clonazepam), have been found to be effective in reducing or eliminating panic attacks associated with the various forms of panic disorder. Substantial research efforts continue the search for other medications useful in the treatment of these conditions. Initial indications are that some of these other agents, particularly the serotonin uptake blockers, may be effective panic medications. The pharmacological agents may present problems such as undesirable side effects, the risk of dependence, and a significant relapse rate once medication is discontinued.

Several variations and combinations of behavioral and cognitive treatment approaches also have demonstrated efficacy in the reduction and/or elimination of panic attacks and agoraphobia. Early reports of research specifically targeting panic attacks indicate that significant numbers of patients are panic-free at the end of cognitive-behavioral treatment and remain so at a 2-year followup.

Information is sparse on such issues as (1) the effectiveness of combined psychosocial and pharmacological treatments, (2) the mechanisms of therapeutic action, (3) demographic and other patient factors that may predict responsiveness to either class of treatment, (4) the long-term effectiveness of treatments for panic disorder once treatment stops, and (5) the value of these treatments for those patients who suffer from panic disorder in combination with other psychological and psychiatric disorders. The latter group represents a significant segment of those people suffering from panic disorder.

To help resolve questions surrounding these and other issues, the Office of Medical Applications of Research of the National Institutes of Health in conjunction with the National Institute of Mental Health convened a Consensus Development Conference on the Treatment of Panic Disorder on September 25-27, 1991. Following a day and a half of presentations by experts in the relevant fields and discussion from the audience, a consensus panel comprising experts in psychology, psychiatry, cardiology, internal medicine, and methodology, as well as members of the general public, considered the scientific evidence and formulated a consensus statement that addressed the following five questions:

• What are the epidemiology, natural history, and course of panic disorder with and without agoraphobia? How is it diagnosed?
• What are the current treatments? What are the short-term and long-term effects of acute and extended treatment of this disorder?
• What are the short-term and long-term adverse effects of these treatments? How should they be managed?
• What are considerations for treatment planning?
• What are the significant questions for future research?

What Are the Epidemiology, Natural History, And Course of Panic Disorder With and Without Agoraphobia? How Is It Diagnosed?

What Is Panic Disorder?

Beginning in the 1960’s, investigators and clinicians began to differentiate patients who had unexpected anxiety attacks from patients with other anxiety disorders. The diagnostic category of panic disorder was first officially recognized with the publication of Diagnostic and Statistical Manual of Mental Disorders (3rd edition) of the American Psychiatric Association in 1980 (DSM-III). These criteria were modified slightly with the 1987 publication of the revised version of the Diagnostic Manual, DSM-III-R.

Fundamental to the diagnosis of panic disorder is the occurrence of panic attacks. These attacks consist of discrete periods of intense fear or discomfort in which at least four of the symptoms noted below develop abruptly and reach a crescendo within 10 minutes, typically lasting 10 minutes or so. Attacks may recur repeatedly and rapidly however, once these symptoms abate, severe anxiety may last for many hours. The symptoms include:

• shortness of breath (or smothering sensations)
• dizziness, unsteady feelings, or faintness
• palpitations or accelerated heart rate (tachycardia)
• trembling or shaking
• sweating
• choking
• nausea or abdominal distress
• depersonalization or drealization
• numbness or tingling sensations (paresthesias)
• flushes (hot flashes) or chills
• chest pain or discomfort
• fear of dying
• fear of going crazy or doing something uncontrolled

Panic attacks may occur as rare isolated incidents that cause little or no sustained impact on the individual’s functioning or as clusters of attacks with adverse effects. They also occur during sleep.

To satisfy the diagnostic criteria for panic disorder, at least some of the panic attacks must occur unexpectedly or spontaneously, that is, in the absence of specific environmental or situational triggers such as elevators, public speaking, snakes, closed spaces, or other situations that evoke fearful avoidance in some people. Further, the diagnostic criteria require either a clustering of at least four attacks spread over a 4-week period or one or more attacks followed by at least 1 month of fearful anticipation of experiencing more such attacks.

Although research is under way to test and refine these criteria, there is a broad consensus that panic disorder, as currently defined, is a distinct condition with a specific presentation, course, positive family history, complications, and response to treatment.

Panic disorder must be differentiated from other disorders that may share similar clinical features. At this time, diagnosis is dependent on a detailed clinical assessment of the presenting complaints and history because there are no specific laboratory tests. A medical workup is recommended to rule out other conditions. At the same time, the risk of misdiagnosis leading to costly medical investigations and delays in treatment for panic disorder must also be guarded against.

Currently, two main subtypes of panic disorder are widely recognized and codified in DSM-III-R. These subtypes vary in the severity and extensiveness of phobic avoidance: panic disorder without agoraphobia and panic disorder with agoraphobia. In cases of panic disorder with agoraphobia, there is avoidance of places or situations from which escape might be difficult or embarrassing or in which help might not be available in the event of a panic attack. The degree of avoidance may vary from mild to moderate or, at the extreme, to a constricted lifestyle imposed by severe avoidance, resulting in the individual’s being nearly or completely housebound or otherwise severely dysfunctional.

Investigators are seeking to develop additional ways of sub typing panic disorder based on the phenomenology, age of onset, response to treatment, etc., which may have implications for etiology, diagnosis, and treatment.

Differential Diagnosis: Separating Panic Disorders From Other Disorders

There are many other disorders in which panic attacks may occur. The more common are simple phobia (in which the panic occurs immediately before or upon exposure to the feared situation and nowhere else) and social phobias in which they occur only when individuals feel they are the focus of others’ attention (e.g. while eating). Other disorders that should be considered in differential diagnosis include claustrophobia; severe depression; dissociative disorders; generalized anxiety without panic; alcohol or drug withdrawal; stimulant abuse (caffeine, cocaine, amphetamines); physical disorders such as cardiac, adrenal, vestibular, thyroid, or seizure disorders.

Epidemiology and Course

Panic disorder is relatively common; similar rates have been found in many countries in international studies. Approximately one third of the individuals with panic disorder also have agoraphobia, although in clinical settings, the majority present with some agoraphobia. Panic disorder with agoraphobia is diagnosed about twice as frequently in females as in males.

The most common age of onset is middle teens and early adulthood; however, panic disorder may onset at any time. A common pattern of onset is the occurrence of occasional unexpected panic attacks that then increase in frequency and are associated with mounting fears of having subsequent attacks. Over time there is often a pattern of spreading fearful avoidance.

Little is known about the long-term course of this disorder. The limited findings to date suggest that in most cases it is a chronic disorder that waxes and wanes in severity. However, some people may have a limited period of dysfunction that never recurs, while others may experience a severe chronic form of the disorder. Those with agoraphobia tend to have a more severe and complicated course. Treatment early in the development of this disorder may shorten the duration and may prevent complications, including agoraphobia and depression.

Comorbidity: Associated Disorders

Certain conditions have been found to be associated with panic disorder, particularly in those individuals with longstanding panic attacks and agoraphobia. These conditions include abuse of alcohol and drugs, depression, and other anxiety and personality disorders. Other medical disorders that occur more commonly in patients with panic disorder may include atypical chest pain, irritable bowel syndrome, asthma, and migraine.

What Are the Current Treatments? What Are The Short-Term and Long-Term Effects of Acute and Extended Treatment of This Disorder?

Panic disorder is a treatable condition. The effectiveness of treatment should be evaluated on a number of dimensions: (1) acceptance and tolerance by patients; (2) reduction or elimination of panic attacks, reduction of clinically significant anxiety and disability secondary to phobic avoidance, amelioration of other common comorbid conditions such as depression; and (3) long-term prevention of relapse.

Several different classes of treatment have been shown to be clinically effective, including cognitive and behavioral, pharmacologic, and combinations of the two. The most commonly used behavioral approach is graduated exposure, aimed primarily at reducing phobic avoidance and anticipatory anxiety. Cognitive-behavioral approaches, developed more recently, also treat panic attacks directly. These treatments involve cognitive restructuring, that is, changing of maladaptive thought processes and are generally used in combination with a variety of behavioral techniques, including breathing retraining and activities that target exposure to bodily sensations and external phobic situations. Ongoing assignments to practice the techniques are made by the therapist. These treatments seem to be well accepted by patients and typically involve weekly sessions for 8 to 12 weeks. Initial improvement is noted in many patients within 3 to 6 weeks of beginning treatment. Among the various psychotherapeutic approaches, combined treatments that include cognitive therapy in addition to other techniques appear to be most effective, especially in reducing panic attacks. Longer term followup of these interventions suggests a low relapse rate.

Pharmacologic treatments include tricyclic antidepressants, monoamine oxidase (MAO) inhibitors, and high-potency benzodiazepines. A significant proportion of patients do not easily tolerate certain of the tricyclics, whereas benzodiazepines are better accepted. Patients who tolerate tricyclics show significant improvement, with a reduced number of panic attacks during the period of treatment, ranging from 8 to 32 weeks in controlled trials. Benzodiazepines have a rapid onset of action with immediate reduction of panic symptoms, whereas antidepressants require 3 to 6 weeks to achieve therapeutic effect. In addition, the action of benzodiazepines in reducing anxiety between attacks is thought advantageous by some clinicians. Careful titration of medication to effective therapeutic doses with gradual increase in dosage is necessary. Very gradual increases may be particularly important with tricyclics in order to reduce attrition. Longer term duration of treatment probably increases clinical response. Gradual tapering of all medications when treatment ends is strongly indicated. The relapse rate following termination of medication for antidepressants is moderate but is probably higher for benzodiazepines. The relatively high response rate to the control conditions (placebo) needs further examination.

Few studies have examined combined behavioral and pharmacologic methods. There is some evidence that a combination of tricyclics and exposure therapy may have additive effects in the short term, but there is no evidence for long-term advantage over either method alone. Currently, there are few published studies available that assess the combined effect of cognitive and pharmacologic intervention, nor has the optimal sequence of combined methods been examined satisfactorily. Whether using a combination of two effective methods improves upon the effectiveness of either alone or is less effective than either alone is not a settled issue.

There are no controlled data on efficacy of treatment for panic disorder or other widely used approaches, such as psychodynamic psychotherapy.

What Are the Short-Term and Long-Term Adverse Effects of These Treatments? How Should They Be Managed?

Adverse effects can be classified in a number of categories, including drug-related disturbances and other physical effects, adverse psychological and behavioral side effects, rebound effects (i.e., worsening of the disorder when treatment is removed), and misplaced confidence in unproven treatments that may preclude other treatments with a better chance of effectiveness.

The adverse effects discussed in this section are based on clinical research studies of panic disorder. It is unclear how and on what dimensions research patients may differ from the general clinical population; thus, the research samples may not be representative of the group of patients that present for treatment in a nonresearch, clinical setting.

In programs offering pharmacotherapy, individuals are not admitted to studies if they have preexisting medical conditions (including pregnancy) that would contraindicate the use of the medications under study. In both pharmacotherapy and cognitive-behavioral studies, individuals are typically referred elsewhere if the individual meets criteria for substance abuse.

Cognitive and Behavioral Treatments

Cognitive and behavioral treatments are ordinarily well tolerated when applied by skilled therapists. Dropout rates in controlled studies range from 5 to 8 percent in the cognitive behavioral therapies and between 12 and 16 percent in the relaxation and in vivo exposure-based treatments. Therapies that include cognitive techniques may also address accompanying depression. Although very few adverse effects of these treatments have been reported, there have been some instances of panic attacks induced by relaxation. This can be counteracted by a more gradual approach to relaxation and teaching the patient techniques for controlling the relaxation procedure. No other adverse effects have been reported.

Other Psychotherapies

In the absence of any empirical studies examining the effectiveness of treatments other than cognitive and behavioral therapies, no conclusions can be drawn about adverse effects. However, given that recent research results have identified useful pharmacotherapy and psychotherapy approaches, one risk of maintaining individuals in nonvalidated treatments of panic disorder is that misplaced confidence in the therapy’s potential effectiveness may preclude application of more effective treatment. This can be particularly problematic with psychotherapy treatments if the nature of the therapeutic relationship makes it difficult for the patient to seek additional or alternate treatment. Psychotherapies without demonstrated effectiveness in panic, such as psychodynamic psychotherapy, however, may be helpful for other difficulties that the patient presents. Thus, when progress in the reduction of panic disorder is not apparent within 6 to 8 weeks, ancillary pharmacotherapy or cognitive behavioral treatment or a brief break in psychotherapy for these treatments should be considered.

Pharmacological Treatments

With three effective classes of pharmacological agents now available in the treatment of panic disorder, risks and benefits of each need to be considered.

Tricyclic antidepressants offer the benefit of once-a-day dosing, a low risk of dependence, and no dietary restrictions. They also have a concomitant antidepressant effect that is frequently helpful. Adverse effects include anticholinergic side effects, low blood pressure, overstimulation, and weight gain. Taken together, these effects may cause up to 35 percent of patients to discontinue treatment before therapeutic benefits occur.

The benefits of MAO inhibitors include, as with the tricyclics, an antidepressant effect and a low risk of dependence. However, the anticholinergic effects may be lower than for the tricyclics. Sexual difficulties, particularly problems in orgasm, may occur as do hypotension and weight gain. One added complication, which may be difficult for some patients, is the need to follow a low tyramine diet.

One benefit of the benzodiazepines, because they have a rapid onset of action, is that they can be used to treat surges of anticipatory anxiety or panic. This “as needed” use of benzodiazepines should not replace the use of sufficient daily doses when that is indicated. Risks include sedation and psychomotor impairment. Benzodiazepines will interact with alcohol if it is not restricted. Although some of these adverse side effects largely subside after 4 to 6 weeks of treatment, subjective cloudiness may remain. The most serious risk with this class of medication is that of physical dependence. Withdrawal symptoms or a recurrence of panic symptoms during drug tapering is a definite risk with long-term treatment.

The attrition rate in pharmacologic studies varies with the drug under investigation. It is approximately 25 percent for the tricyclics, slightly lower for the MAO inhibitors, and approximately 15 percent for the high-potency benzodiazepines. Many of these dropouts appear directly related to the drug side effects. With imipramine, starting with a low dose and building up slowly may significantly reduce the risk of premature treatment termination. Similarly, the potential excessive use of benzodiazepines requires caution in their use in individuals who have a history or risk of drug dependence. Care must be exercised in prescribing the tricyclic and the MAO inhibitor medications for individuals with cardiovascular disease; if acute relief is needed in such patients, high-potency benzodiazepines are the treatments of choice.

What Are Considerations for Treatment Planning?

The practicing clinician does not usually see panic disorder in its pure form. Further, because there are a number of different treatment strategies with similar treatment efficacy in the acute phase, the central question becomes not “What is the treatment of choice?” but, “What factors need to be considered in choosing optimum treatment?” Decisions need to be made regarding choice of single modality, concurrent, or sequential interventions.

Primary care physicians or other clinicians who identify patients with panic disorder will need to address the issue of potential referral for treatments specific to panic disorder with or without agoraphobia.

The factors that need to be considered by any clinician include degree of urgency, comorbid conditions, history, and patient fit and compliance issues.

Each of these groups of factors will be examined independently both in terms of the assessment data required and their implications for strategic interventions.

Degree of Urgency

There are cases of emergency such as medical complications secondary to the phobic fears (e.g., fear of swallowing leading to dehydration and weight loss), imminent loss of job or relationship, inability to undergo necessary medical procedures, children’s welfare at risk, or acute and rapid generalization of phobic behavior. In such cases, mobilization of family resources or high-potency benzodiazepines may be the starting point for treatment once the patient has received basic educational information. This may be accompanied by cognitive-behavioral treatment, alternative medications, and other followup care. The patient’s own subjective sense of urgency may or may not indicate a need for urgent intervention. A panic attack in and of itself is not an emergency. Common obsessive fears of losing control need to be carefully distinguished from actual imminent loss of control.

History

The history of the patient and his or her family will yield critical information for treatment planning. Is this the first episode or one in a lifetime series? Has the patient ever received the diagnosis before? What treatments have been tried in the past, and were they successful in some or any measure? Is there a family history of psychiatric disorder or substance abuse? Did the patient or the family engage in or respond to any treatment? Were there recent events that may have triggered the current onset of symptoms, such as surgery, illness, childbirth, miscarriage, trauma, loss, or external stressors? Are there any known developmental vulnerabilities such as a history of abuse or dysfunctional family? The need for and advisability of including family or significant others in the educational and/or treatment process should be assessed.

Comorbid Conditions

There are three kinds of medical conditions that may affect treatment planning and may need to be treated concurrently. These are (1) conditions that may affect the safety or efficacy of psychopharmacological treatments (such as some specific cardiovascular, pulmonary, gastrointestinal, or endocrine disorders; pregnancy; or lactation); (2) conditions with a prominent component of anxiety (such as thyroid disease, polycythemia, lupus, and pulmonary insufficiency); and (3) conditions requiring treatment with medications such as vasoconstrictors, bronchodilators, or steroids, which may cause or exacerbate anxiety.

The necessity for a complete psychological assessment in addition to the medical workup cannot be overemphasized. Up to 70 percent of patients with panic disorder may have a comorbid psychological or psychiatric condition that will need to be included in the treatment planning and perhaps addressed therapeutically concomitantly or at a later point. A high percentage are depressed or demoralized secondary to suffering panic attacks but should be treated for panic first. Other conditions such as major depression, posttraumatic stress disorder, bipolar mood disorder, dissociative disorders, other anxiety disorders such as obsessive compulsive disorder or social phobia, eating disorders, or complex personality disorders may require concurrent treatment.

Finally, individuals need to be assessed explicitly regarding substance abuse, including alcohol, marijuana, opiates, hallucinogens, cocaine, over-the-counter drugs such as nasal sprays and diet pills, caffeinism, or benzodiazepine abuse. Patients in current withdrawal or active abuse must be treated for substance abuse before or concurrent with specific panic disorder treatment.

Patient Fit and Compliance Issues

The clinician, in consultation with the patient, should select one of the treatments with demonstrated efficacy or a combination as the initial treatment. Selection should be based on patient preference in the context of a comprehensive assessment of urgency, history, and comorbidity. It may be the case that the selected treatment will require referral, consultation, or supervision.

The individual with panic disorder needs to be an active, fully informed participant in the treatment planning process. Education and demystification are frequently needed. This means advising the patient not only of the short-term benefits and risks but also of long-term benefits and risks where known and addressing the issue of long-term relapse prevention. The patient’s initial degree of relief and motivation following education may give direction to the next step. Attitudes and concerns regarding various treatment options must be explored and negotiated. The patient’s request in presenting for treatment must be kept in mind. Answering questions such as “why me?” or “why now?” or “what is this about?” may establish a better foundation for treatment.

Patients should be given education about the disorder and encouragement to re-enter phobic situations gradually when medication alone is chosen as the initial treatment. Current research suggests that an absence of any noticeable improvement after about 6 to 8 weeks of any treatment should suggest a reassessment, consultation, or change of modality.

Particularly for those patients for whom there has been a chronic course or a history of multiple episodes of acute symptomatology, recovery, and relapse, longer term strategies need to be considered following the acute phase of treatment, Unfortunately, at this time, little is known regarding the relative long-term efficacy of maintenance doses of medication, other psychotherapies, changes in lifestyle aimed at stress reduction, or participation in ongoing self-help groups. These current practices have been shown to be of value in other disorders and may in the future be shown to be so in panic disorder as well. As with many other treatable disorders, access to effective care is at times limited by regulatory decisions, lack of financial resources, inadequate third party coverage, and stigma.

What Are the Significant Questions for Future Research?

As would be expected in a relatively new field, many research questions remain, and each new finding is likely to stimulate further questions. Among the most important questions are the following:

Identifying Those at Risk

Although onset is known to be most frequent in adolescence and young adulthood, little is known about who is more likely to have an isolated attack, and, of those persons, who will go on to develop the full disorder, and what sequence of events may influence this. In this area, promising leads to follow are the investigation of temperament and personality; family and genetic patterns; developmental growth characteristics; and other biological, psychological, and environmental factors. Thus, both high-risk studies (e.g., children of high-risk families) and population studies are needed to answer these questions.

Course of Disorder

Much of the information currently available is derived from cross-sectional studies and from short-term followup. Also needed are long-term prospective studies that track episodes and the context in which they occur over time, assessment of the development of comorbid conditions, treatment-seeking behavior, medical care utilization and costs with and without treatment for panic disorder, as well as changes in functioning and the quality of life.

Methodological Studies

Currently, different measures, often idiosyncratic and some of undocumented quality, make comparison of subjects and results across studies difficult. More reliable, valid measures of all clinical features of panic disorder must be developed and standardized for general use. Similarly, there is a need for standardized methodologies for measuring all facets of outcome, including operational definitions of response, remission, recovery, and relapse.

Although field studies of diagnostic boundaries and criteria are ongoing, further research is required on the clinical definition of panic disorder, including the validity of the diagnostic criteria and possible subtyping or variations of the disorder, which may have different natural histories or responses to treatment. Sensitive screening diagnostic instruments will be needed for population and genetic studies, prevention programs, and general clinical use.

Treatment Research

It is essential that recruitment strategies, success rates, and inclusion and exclusion criteria be very carefully and fully documented in each clinical research study.

Current information does not permit satisfactory comparison of the effectiveness and value of cognitive-behavioral and pharmacological treatments. Not only are multisite studies and comparable control groups needed, but cross-disciplinary studies within sites will facilitate interprofessional exchange of knowledge and skills. Multisite studies should be done in which psychosocial and pharmacological therapies are compared with each other and to combinations of the two. Further research is needed on optimal duration of treatment and on strategies to maintain treatment response. Studies are also needed to ascertain the type and extent of training of clinicians necessary for effective intervention. Studies are needed to assess patient match with treatment methods, including the sequencing of treatments.

Patients who drop out of clinical trials should be carefully followed. Some clinical drug trials also have revealed a high placebo response, suggesting that there are nonspecific psychosocial, unsystematic exposure instructions, or other unspecified factors that may have a potential influence on therapeutic outcomes.

Finally, new emphasis should be placed on prevention research programs for individuals at risk.

Basic Research

Current evidence supports familial prevalence, but there is only preliminary evidence for genetic transmission. Larger studies are needed to separate the genetic from the environmental contribution and to identify the most salient milieu influences (life events, family functioning, etc.). In such studies, there should be a focus on identification of which diagnostic criteria are most likely to identify a genetic form of the disorder. Segregation studies should be done to determine likely patterns of transmission and to obtain estimates of genetic parameters necessary for the successful analysis of linkage studies.

Further basic studies of the biological and psychological underpinnings, as well as the influence of environmental factors associated with the disorder, are needed to understand its nature. Neurobiologic studies, including molecular approaches, and experimental studies of basic cognitive and behavioral processes will yield information and contribute to more effective treatment.

Conclusions and Recommendations

• Panic disorder is a distinct condition with a specific presentation, course, and positive family history and for which there are effective pharmacologic and cognitive behavioral treatments.
• Treatment that fails to produce benefit within 6-8 weeks should be reassessed.
• Patients with panic disorder often have one or more comorbid conditions that require careful assessment and treatment.
• The most critical research needs are:
• the development of reliable, valid, and standard measures of assessment and outcome;
• the identification of optimal choices and structuring of treatments designed to meet the varying individual needs of patients; and
• the implementation of basic research to define the nature of the disorder.
• Barriers to treatment include awareness, accessibility, and affordability.
• An aggressive educational campaign to increase awareness of these issues should be mounted for clinicians, patients and their families, the media, and the general public.

Consensus Development Panel

Layton McCurdy, M.D.
Panel and Conference
Chairperson
Vice President for Medical Affairs and Dean
Medical University of South Carolina
Charleston, South Carolina

Frank A. DeLeon-Jones, M.D.
Professor of Psychiatry University of California at
Los Angeles
Olive View Medical Center Los Angeles, California

Susan Dime-Meenan
Executive Director
National Depressive and
Manic Depressive Association Chicago, Illinois

Jean Endicott, Ph.D.
Chief
Department of Research
Assessment and Training New York State Psychiatric Institute
New York, New York

Raquel E. Gur, M.D., Ph.D.
Professor of Psychiatry and Neurology
Department of Psychiatry
University of Pennsylvania
Philadelphia, Pennsylvania

Helena Chmura Kraemer, Ph.D.
Professor of Biostatistics in Psychiatry
Department of Psychiatry and Behavioral Sciences
Stanford University
Stanford, California

Marsha M. Linehan, Ph.D.
Professor of Psychology
Psychology Department
University of Washington
Seattle, Washington

Carl I. Margolis, M.D.
Internal Medicine/Psychiatry
Private Practice
Rockville, Maryland

Charles R. Marmar, M.D.
Associate Professor of
Psychiatry
University of California at
San Francisco
Director
Post-Traumatic Stress
Disorder Program
San Francisco Veterans
Administration Medical Center
San Francisco, California

Susan Mineka, Ph.D.
Professor of Psychology
Department of Psychology
Northwestern University
Evanston, Illinois

Jeanne S. Phillips, Ph.D.
Professor of Psychology Department of Psychology University of Denver
Denver, Colorado

Ray H. Rosenman, M.D.
Director of Cardiovascular
Research (Ret.)
SRI International Menio Park, California Associate Chief of Medicine Mt. Zion Hospital and Medical Center
San Francisco, California

Peter C. Whybrow, M.D.
Professor and Chairman
Department of Psychiatry
University of Pennsylvania
School of Medicine
Philadelphia, Pennsylvania

Sally M. Winston, Psy.D.
Director
Anxiety Disorders Program
Sheppard and Enoch Pratt
Hospital
Baltimore, Maryland

Speakers

James C. Ballengei, M.D.
“Acute Pharmacological Treatment of Panic Disorder:
Standard Medications”

David H. Barlow, Ph.D.
“Behavioral Treatment of Panic Disorder”

Dianne L Chambless, Ph.D.
“Discussion of Psychotherapy Treatments”

David M. Clark, D.Phil.
“Cognitive Therapy for Panic Disorder”

Alien Frances, M.D.
“Psychodynamic Treatment of Panic Disorders”

Jack M. Gorman, M.D.
“New and Experimental
Pharmacological Treatments for Panic Disorder”

George R. Heningei, M.D.
“Mechanism of Action in the Phannacotherapy of Panic Disorder”

Wayne Katon, M.D.
“Primary Care Panic Disorder Management Model”

Heinz Katschnig, M.D.
“The Long-Term Course of Panic Disorder”

Gerald L. Klerman, M.D.
“A Critique of the Research Literature on Combined Treatment of Panic Disorder Discussion”

Michael R. Liebowitz, M.D.
“Diagnosis and Clinical Course of Panic Disorder With and Without Agoraphobia”

Larry K. Michelson, Ph.D.
“Risk-Benefit Issues in Psycho-social Treatment of Panic Disorders”

S. Rachman, Ph.D.
“Mechanisms of Action in Psychosocial Treatments of Panic Disorder”

Karl M. Pickets, M.D.
“Risk/Benefit Issues in Pharmacological Treatment of Panic Disorders”

M. Katherine Sheal; M.D.
“The Future”

Michael J. Telch, Ph.D.
“A Critique of the Research
Literature on Combined
Treatment of Panic Disorder”

Thomas W.Uhde, M.D.
“Discussion of Pharmaco-therapy”

Myrna M. Weissman, Ph.D.
“The Epidemiology and Genetics of Panic Disorder”

Planning Committee

Robert M.A. Hirschfeld, M.D.
Planning Committee
Chairperson Chairman Department of Psychiatry and
Behavioral Science University of Texas Medical
Branch at Galveston Galveston, Texas

James C. Ballenger, M.D.
Professor and Chairman Department of Psychiatry and
Behavioral Sciences Director
Institute of Psychiatry Medical University of South
Carolina Charleston, South Carolina

David H. Darlow, Ph.D.
Distinguished Professor of Psychology
Department of Psychology
University at Albany
State University of New York
Albany, New York

Lynn Cave
Information Office National Institute of Mental
Health Alcohol, Drug Abuse, and
Mental Health Administration Rockville, Maryland

Marsha Corbett
Director
Office of Scientific Information
National Institute of Mental
Health Alcohol, Drug Abuse, and
Mental Health Administration Rockville, Maryland

Jerry M.EIIiott
Program Analyst
Office of Medical Applications of Research
National Institutes of Health
Bethesda, Maryland

Paul Emmelkamp, Ph.D.
Professor of Clinical Psychology and Psychotherapy
Academic Hospital
Department of Clinical Psychology
Groningen, The Netherlands

John H. Ferguson, M.D.
Director
Office of Medical Applications of Research
National Institutes of Health Bethesda, Maryland

William H. Hall
Director of Communications
Office of Medical Applications of Research
National Institutes of Health Bethesda, Maryland

Lewis L. Judd, M.D.
Professor and Chairman
Department of Psychiatry
School of Medicine
University of California at San Diego
La Jolla, California

Martin B. Keller, M.D.
Professor and Chairman
Department of Psychiatry and Human Behavior
Brown University
Butler Hospital
Providence, Rhode Island

Donald F. Klein, M.D.
State University of New York
College of Medicine at New York City
Director of Psychiatric Research
New York State Psychiatric Institute
Professor of Psychiatry
College of Physicians and Surgeons of Columbia University
New York, New York

Gerald L. Klennan, M.D.
Professor of Psychiatry
Associate Chairman for Research
Cornell University Medical College
Payne Whitney Clinic
New York, New York

Jack D. Maser, Ph.D.
Acting Chief
Mood, Anxiety, and Personality Disorders Research Branch
Division of Clinical Research
National Institute of Mental Health
Alcohol, Drug Abuse, and Mental Health Administration
Rockville, Maryland

S. Rachman, Ph.D.
Professor
Psychology Department
University of British Columbia
Vancouver, British Columbia
Canada

Darrel A. Regiel, M.D.,M.P.H.
Director
Division of Clinical Research
National Institute of Mental Health
Alcohol, Drug Abuse, and Mental Health Administration
Rockville, Maryland

Morton Reiser, M.D.
School of Medicine
Yale University
New Haven, Connecticut

Thomas W.Uhde, M.D.
Chief
Section on Anxiety and Affective Disorders
Biological Psychiatry Branch
Intramural Research Program
National Institute of Mental Health
Alcohol, Drug Abuse, and Mental Health Administration
Bethesda, Maryland

Barry E. Wolfe, Ph.D.
Staff Psychologist
Mood, Anxiety, and Personality Disorders Research Branch
Division of Clinical Research National Institute of Mental
Health Alcohol, Drug Abuse, and Mental Health Administration
Rockville, Maryland

Conference Sponsors

National Institute of Mental Health
Alan Leshner, Ph.D.
Acting Director

Office of Medical Applications of Research, NIH
John H. Ferguson, M.D.
Director