Both cognitive-behavioral1-3 and pharmacological4-6 treatments for panic disorder have been found to be effective over the short term. Not all patients, however, can tolerate or fully respond to these approaches,1-3,7,8 and the effectiveness of these interventions over the long term remains unclear.1,9 The degree to which these treatments affect psychosocial and quality-of-life impairments associated with panic disorder is uncertain because of limited data.10,11 Given the high level of associated morbidity and health costs, it is important to continue to develop and study additional treatments for panic disorder.12-15
Psychodynamic psychotherapy is a widely used treatment for panic disorder, but it has received little systematic assessment.16 The authors have developed a manualized version of psychodynamic psychotherapy called panic-focused psychodynamic psychotherapy (PFPP).17 PFPP differs from nonspecific psychodynamic psychotherapy in its focus on symptoms and dynamics of panic disorder. In a recent randomized controlled trial,18 PFPP demonstrated efficacy for panic disorder in comparison to a less-active psychotherapy called applied relaxation training (ART).19 This study was the first randomized controlled trial of a manualized psychoanalytical treatment as a sole intervention for panic disorder. This article will describe this treatment and PFPP research studies.
PFPP is based on core psychoanalytical concepts, including the existence and centrality of the unconscious, the relation of defense mechanisms and conflicted wishes to symptom formation, differences between signal and traumatic anxiety, and the importance of transference phenomena. We will discuss these concepts specifically as they relate to the psychodynamic treatment of panic disorder.
From a psychodynamic perspective, symptoms develop in part from unconscious fantasies and conflicts.20 For example, patients with panic disorder often struggle with angry feelings and fantasies, which they experience as threats to important attachment figures.17,21,22 Although patients are usually aware of some angry feelings, which they commonly find uncomfortable, other unconscious angry or vengeful fantasies that are less tolerable emerge in the course of treatment.
Feelings and fantasies that are experienced as threatening or dangerous are often avoided by unconscious mental processes called defense mechanisms.23 Certain defense mechanisms are used more frequently by patients with panic disorders. These are reaction formation, undoing, and denial.24 These mechanisms permit the patient to avert recognition of angry feelings toward important love objects and intensify affiliative efforts, with the aim of reducing threats to attachments.
Many aspects of people’s lives, including fantasies, behavior, and symptoms, come about as the result of compromise formations, which are a com- promise between a conflicted wish, often unconscious, and the defense against that wish (also unconscious).20 Panic attacks can represent a compromise between angry and dependent wishes that are expressed through a coercive demand for help and a presentation of helplessness to avoid having to acknowledge rage.
As described by Freud, signal anxiety is a small dose of anxiety that alerts the ego to the presence of wishes and impulses that are felt to be dan-gerous, typically triggering defense mechanisms.25 The failure of defenses to modulate the threat leads to traumatic levels of anxiety, akin to what we call panic attacks. In PFPP, therapists work to help patients recognize and reappraise frightening wishes and fantasies.
In the course of psychotherapy, unconscious conflicts and expectations that the patient has developed with others in the past emerge in the relationship with the therapist (transferance).26 A focus on transference during the course of therapy facilitates articulation of overarching fantasies, permitting the patient to gain more understanding and control of them.
A psychodynamic formulation for panic disorder
A psychodynamic formulation for panic disorder was developed for the purpose of identifying core conflicts to be addressed in a panic-focused psychodynamic treatment. The construction of this formulation from psychoanalytic theory and concepts, including those described above, clinical observations, and more systematic psychological studies, is described in detail elsewhere.17,21,27,28
The formulation suggests that individuals prone to panic have a fearful dependency on significant others, which is derived from a biochemical vulnerability or traumatic and/or ambivalent relationships with early caregivers that broadly affects their psychological functioning.29 They have a sense of personal inadequacy, and their attachments feel insecure. They often feel they must depend on others to provide a sense of safety, leading separation to be experienced as traumatic. Many patients with panic disorder dread becoming angry at people they love and are frightened that this anger will damage these relationships. A vicious circle of fearful dependency, anger, and anxiety can develop, often triggered by stressors, such as interpersonal loss (Figure 1).30
Panic-focused psychodynamic psychotherapy
Based in part on this formulation, the authors developed a manualized treatment, PFPP.17 In the PFPP research studies, the treatment consists of 24 twice-weekly sessions. This manualized approach targets core conflicts about anger recognition, ambivalent feelings about autonomy, and fears
of loss or abandonment commonly found in panic disorder. Characteristic psychodynamic techniques of clarification, confrontation, and interpretation are used to elucidate and address these conflicts.
The treatment is divided into 3 phases (Figure 2). In the initial phase, the therapist focuses on identifying the meaning and content of panic symptoms, derived from exploring the circumstances, stressors, and feelings surrounding panic onset. Elucidation of a developmental history, including previous panic episodes, helps determine early life experiences and self and object representations that may play an active part in panic. The initial phase aims at relief of panic symptoms.
In the second phase, the therapist works with the patient to identify core conflicts underlying panic disorder. Conflicts surrounding anger and autonomy, as well as other contributing dynamics, are brought to the patient’s attention. Defense mechanisms, including reaction formation, undoing, and denial, are addressed as efforts—often unconscious—to avoid facing emotional contributions to panic symptoms. Emergence of the transference allows for exploration of these conflicts and defenses in the relationship between therapist and patient. This phase focuses on addressing vulnerability to panic and relapse.
In the third (termination) phase, mixed feelings surrounding anger, autonomy, and separation are addressed as they emerge in ending the therapeutic relationship. The therapist helps the patient articulate feelings about the loss of the therapist, allowing for further recognition of conflicts and a reduced risk of panic recurrence. Increased assertiveness, encouraged by this patient-directed approach, and an increased sense of safety in being more able to tolerate mixed feelings, helps improve psychosocial function.
Although indications for psychodynamic psychotherapy have been described as good verbal skills, psychological mindedness, and a curiosity about the origins of symptoms, panic patients tend to focus on bodily experiences as a way of avoiding frightening feelings and the verbal articulation of conflict. In studies of PFPP, patients have generally responded well despite often limited psychological mindedness.31 Patients become engaged in the treatment as the therapist identifies links between feelings and circumstances surrounding panic onset and patients’ emotional lives.
Research on PFPP
PFPP was initially studied in an open trial of 21 patients with primary DSM- IV panic disorder,32,33 diagnosed on the Anxiety Disorders Interview Schedule for DSM-IV, Lifetime Version (ADIS-IV-L).34
Four patients dropped out of the study. At the end of treatment, 16 of the remaining 17 patients met remission criteria established by the multicenter study of treatments for panic disorder.1 Patients experienced significant improvements in psychosocial function, depression, and nonpanic anxiety. These improvements were maintained at 6-months follow-up without intervening treatment. Notably, 8 patients who entered the study meeting DSM-IV criteria for comorbid major depression experienced resolution of depression as well as panic.
Our group recently completed a landmark study in which we demonstrated efficacy of PFPP for panic disorder with and without agoraphobia.18 This was a randomized controlled trial comparing PFPP to a less-active psychotherapy, ART.19 Inclusion criteria were a diagnosis of primary DSM-IV panic disorder with or without agoraphobia (ADIS-IV-L), with at least 1 panic attack per week. Patients entering the study agreed to stop all nonstudy psychotherapy and to hold their medication constant if they were taking medication (15% of the sample). Patients with comorbid severe agoraphobia, major depression, and DSM-IV personality disorders were included. Exclusion criteria were psychosis, bipolar disorder, and substance abuse (6-months remission necessary).
The Panic Disorder Severity Scale (PDSS) was the primary outcome measure.35 Response was defined as a 40% reduction from baseline on the PDSS.1 Patients were assessed with the Sheehan Disability Scale (SDS),36 the Hamilton Rating Scale for Anxiety (HAM-A),37 and the Hamilton Rating Scale for Depression (HAM-D).38 Both therapies were of 12 weeks’ duration, with twice-weekly (24 session) interventions. ART included a cognitive explanation about panic disorder, progressive muscle relaxation, cue-controlled relaxation, twice-daily homework, and an exposure protocol.
Psychotherapy sessions were videotaped for adherence monitoring. Both groups demonstrated high levels of adherence to the manualized treatments based on an assessment of 3 videotapes of each therapy using condition-specific adherence measures: the PFPP Adherence Rating Scale (available from the authors) and the ART Adherence Scale.39
No significant demographic or clinical differences were found between the 2 treatment groups at baseline besides sex—more men were in the ART group: 47% versus 15%; 2-tailed Fisher’s exact P < .05 (Table 1). There were no baseline differences between randomized groups in severity of panic disorder, PDSS score, or baseline severity ratings. PFPP had a significantly higher response rate than ART: 73% versus 39%; P = .016 (Table 2); and significantly greater improvement in symptoms of panic disorder: PDSS, P = .002 (Figure 3) and psychosocial function: SDS, P = .014. The 2 treatments did not differ significantly in changes on the HAM-D (P = .07) or in nonpanic anxiety (HAM-A, P = .58).
Only 2 patients in the PFPP group dropped out (7%), an unusually low attrition rate for a randomized controlled trial of panic disorder in the United States, compared with 8 dropouts in the ART group (34%). Patients had good response to treatment despite the inclusion of severe agoraphobia and comorbid depression, making this sample relatively more generalizable than subjects in some other panic disorder outcome studies.1,2,7,40,41 PFPP performed comparably to clinical trials of cognitive-behavioral therapy (CBT) and medication, but these treatments were not directly compared with one another. Currently, a 2-site study (Cornell, New York and University of Pennsylvania, Philadelphia) is under way comparing PFPP with CBT and ART.
Preliminary moderator analyses of the efficacy study showed that patients with primary DSM-IV panic disorder with comorbid SCID-II-diagnosed Axis II cluster C disorders responded better to PFPP than patients without this comorbidity.42 Because panic patients with cluster C comorbidity have not always responded well to CBT,43 PFPP may be of particular value for this population.
Although the results of the PFPP studies are promising, further studies are needed to determine which treatment interventions are most appropriate for any given patient with panic disorder.
Mr B had onset of severe panic disorder 2 days after his 39th birthday and presented with a score of 12 on the PDSS. In the first phase of treatment, the therapist explored the circumstances surrounding panic onset. Although Mr B acknowledged that he had been “stressed,” he was puzzled about the source of the panic attacks. He focused initially on problems at work: he was not comfortable with the pressure at work since being promoted to leader of his division. As therapy progressed, he started to realize that he had been having some of the symptoms of his subsequent panic attacks for several weeks before their onset when having to reprimand or fire employees. Specifically, he felt tremors in his arms and a sense of loss of control.
Mr B described a difficult background. His father was a temperamental man who was especially intolerant of his son’s early reading and writing difficulties. He described his mother as self-absorbed, often neglecting him. On 3 occasions, his parents sent him away from home, once to relatives and twice to boarding schools, for reasons he did not understand at the time. As a child, he assumed he was sent away because he was “bad,” or he was being punished for losing his temper with his mother. His parents often fought and ultimately divorced.
Although his relationship with his father improved over the years, his mother became a lonely embittered woman who viewed herself as a victim of unfair life circumstances. She pressured Mr B to take care of her, which he experienced as efforts to draw him away from his wife and job. Although he was angry at his mother’s refusal to make efforts to take better care of herself, he felt guilty saying no to her.
Mr B’s panic attacks resolved within the first 6 sessions of PFPP, as he began to recognize that his panic arose as a result of intense fears about confronting his employees that were related to his early separation experiences.
In phase 2, his difficulty in tolerating his anger and his fears of abandonment were explored in more depth. It emerged that Mr B felt frightened about reprimanding his employees, because even though he was the boss, he feared that he would be abandoned or punished, just as he felt would happen if he confronted his mother about her selfish behavior. Mr B’s panic attacks were also triggered by his rage at his employees, particularly one whose victimized stance reminded him of his mother. Unconsciously, he feared both loss of control of his anger (Mr B’s shaking arms during his anxiety and panic) and being abandoned and punished for its expression, which he linked to his being sent away as a child. He had avoided his anger heretofore with reaction formation, taking greater care of his mother while fending off his rage.
In the final phase of treatment, Mr B discussed his frustration and abandonment fears as they emerged in the transference with the therapist. He expressed a concern that the therapist would not be able to tolerate his feelings about leaving the treatment and would become angry with him. Recognizing this transference fear as a fantasy deepened his understanding of the conflicts that had been identified earlier in the treatment. As his fears subsided, Mr B became more effective in his functioning as a boss, and was able to set better limits with his mother. On completion of his 24-session treatment, he had a score of 1 on the PDSS.
1. Barlow DH, Gorman JM, Shear MK, Woods SW. Cognitive-behavioral therapy, imipramine, or their combination for panic disorder. JAMA. 2000;283:2529-2536.
2. Craske MG, Brown TA, Barlow DH. Behavioral treatment of panic: a two-year follow-up. Behav Ther. 1991;22:289-304.
3. Craske MG, DeCola JP, Sachs AD, Pontillo DC. Panic control treatment for agoraphobia. J Anxiety Disord. 2003;17:321-333.
4. Otto MW, Tuby KS, Gould RA, et al. An effect-size analysis of the relative efficacy and tolerability of serotonin selective reuptake inhibitors for panic disorder. Am J Psychiatry. 2001;158:1989-1992.