Categorizing Fear: The Role of Trauma in a Clinical Formulation

This case report illustrates the difficulties that can accompany the evaluation and treatment of an individual with a history of severe psychological trauma, who meets criteria for several disorders, and presents problems in multiple domains of functioning. Because there were several failed early treatment efforts, the facts of the case are presented as they unfolded.

CASE PRESENTATION

At the time of her presentation, Hope was the 39-year-old mother of one son, whom she had not seen in many years. She was a nonpracticing Protestant of mixed German-American and Anglo-American descent, lived with her third husband, a Catholic Latino, and was supported by his family, public assistance, and odd jobs. She was referred from a gynecology clinic with the chief complaint, “I got depressed when I found out I had to have surgery.”

Although Hope described lifelong feelings of depression and anxiety, she had been relatively well until recently diagnosed with an adnexal cyst. Frightened by the prospect of surgery, Hope experienced agitation, social withdrawal, irritability, poor concentration, insomnia, and crying spells. She also once heard a voice say, “It’s getting bad again,” but realized she was hearing her own thoughts. She had occasionally heard a similar voice in the past that resembled her father’s. Hope also described chronic, although recently worsened, memory problems.

History

Psychiatric. Hope was psychiatrically hospitalized twice in her mid-20s, after jumping from a third-story window and after an overdose of diazepam 1 year later. Treatments for depression and anxiety included amitriptyline, buspirone, and thioridazine, to varying effect (doses unknown).

Hope used alcohol heavily for several years in her late 20s but stopped while in group therapy and never used other substances.

Family. Both of Hope’s parents were alcoholic. Her father suffered from depression and had made at least one suicide attempt. Hope’s maternal grandmother had been diagnosed with early-onset Alzheimer’s disease, and her mother had had memory problems in her early 40s.

Social. Hope was the eldest of six children born to an army officer and his wife stationed overseas. Both parents physically abused all of the children, but especially Hope. In the initial interviews, she reported that her father had sexually abused her throughout her childhood and adolescence, but further details were not elicited. Hope’s entire immediate family died in a car accident when she was 30.

Throughout her 20s and 30s, Hope was employed in a variety of workplaces, from roadside diner to construction site to assembly line in a pickle factory. She attended 2 years of college, then left to marry her first husband. He physically abused her throughout their 3-year marriage and ultimately kidnapped their 2-year-old son and left the country. Hope has not seen her son since. She married her second husband within a year but left after he became physically abusive. Hope married her current husband at the age of 35. He is 12 years her junior and is not abusive. At the time of her presentation, she was working 2 days per week in a retail clothing store.

Mental Status Examination

The evaluating psychiatrist reported that Hope was a large woman, dressed neatly in casual clothes, who seemed somewhat younger than her stated age. She endeavored to answer questions fully and often smiled nervously when speaking. The psychiatrist noted that Hope had a depressed mood, a full affect, and an emotional detachment when describing the death of her family. There were no psychotic symptoms or suicidal or homicidal ideation. Cognition was reported as normal, although a Mini-Mental State evaluation was not performed.

Initial Treatment

In spite of her history of serious psychiatric problems, Hope was given the diagnosis of adjustment disorder, with depressed and anxious mood, and was referred for weekly supportive psychotherapy to manage the stress of the upcoming surgery. After realizing that her current anxiety was related to previously experienced medical complications, Hope underwent surgery without undue emotional strain.

After the surgery, Hope and her therapist focused on managing the stress of living in a chaotic home. Although Hope’s husband did not drink or use drugs, his sister and brother both abused alcohol and cocaine and often fought violently. When her therapist attempted to explore Hope’s own experience with violence, she became more anxious and avoided further discussion but began to complain of more frequent nightmares.

Shortly after starting therapy, Hope requested a psychopharmacology evaluation. The psychiatrist noted mood lability, insomnia, irritability, restlessness, loud, rapid speech, and increasingly poor impulse control, as evidenced by Hope striking her husband on numerous occasions. Results of laboratory tests, including complete blood count, chemistries, and thyroid function, were all within normal limits.

Diagnoses of major depressive disorder and posttraumatic stress disorder (PTSD) were considered. However, because of prominent mood lability, bipolar disorder, mixed episode, was diagnosed, and valproate, 500 mg h.s., was prescribed; the dose was increased to 1500 mg/day in divided doses (53.6 µg/ml [50–100 µg/ml]). Hope’s mood, irritability, and insomnia improved minimally but worsened again in response to stressors including the death of two friends and increasing conflict at home. Trazodone, 50–100 mg h.s. as needed, was added for insomnia. At that point, her therapist and psychiatrist informed Hope of their departure from the clinic at the end of the academic year (since the teaching clinic’s structure is such that patients are usually assigned new clinicians on a yearly basis). In the meantime, Hope was scheduled to undergo a Nissan fundiplication for peptic ulcer disease and had a relapse of mood symptoms.

The new psychiatrist agreed with the diagnosis of bipolar disorder, mixed state, and valproate (2000 mg/day) and trazodone (200 mg h.s.) were increased. Hope continued in once-weekly supportive psychotherapy with a new therapist.

She successfully underwent surgery but continued to complain of irritability, depression, frequent crying spells, nightmares, and poor sleep. Valproate was increased to 4000 mg/day. It was often unclear whether somatic complaints (gastrointestinal upset, tremor, diarrhea) were due to adverse effects or preexisting conditions. Lithium, 300 mg b.i.d., and bupropion, 75 mg t.i.d., were added, resulting in only temporary improvement of sleep and irritability. After several months of such failed psychopharmacologic interventions, Hope herself decreased the doses of her medications because of side effects and expressed anger with both clinicians regarding her lack of improvement.

During this period, Hope’s psychiatrist and therapist had added borderline personality disorder to the diagnostic formulation on the basis of a pattern of affective instability, suicidal behaviors, and uncontrollable anger. Although she never exhibited self-mutilating behaviors, her clinicians felt that her difficulty focusing on symptom assessment with her psychiatrist, and psychosocial issues with her therapist, was due to splitting. Hope missed several appointments, decided to terminate therapy, and missed all appointments with her psychiatrist until the end of the year, when a new clinician would have to be assigned. Hope’s psychiatrist recommended that she be assigned to a single clinician for both psychotherapy and pharmacotherapy.

DIAGNOSIS AND EVALUATION

When a patient presents in crisis, pressure to intervene quickly sometimes abbreviates the evaluation process, and the clinician must keep in mind the necessity of returning to the evaluation as soon as possible. The first diagnostic impression proved incomplete when, after an initial antidemoralization response to therapy, Hope relapsed into a complex syndrome of affective, anxiety, cognitive, and behavioral symptoms. As Hope’s presentation evolved over several months, the clinicians consulted with senior supervisors and conferred regularly to minimize the potential problems intrinsic to split treatments (discussed later). This case, however, illustrates the unfortunate possibility that clinical nonresponse may lead to a progressively more complex medication regimen as the clinician follows an algorithm for treatment-refractory patients, when the problem lies with the diagnostic formulation. When treatment is not succeeding, reassessment of the working diagnosis and clinical formulation is also in order.

Establishing an Alliance

Hope’s third psychiatrist was able to schedule an initial appointment only after 2 months of phone calls and cancellations. Hope was pessimistic about the possibility of being helped but returned out of desperation and was relieved to be offered weekly sessions. She continued to complain of insomnia, irritability, aggressive threats toward her husband, and low energy, as well as worsening memory.

Her psychiatrist decided to investigate each of Hope’s complaints in turn. Her medication regimen at that time was valproate, 1000 mg b.i.d., lithium, 300 mg b.i.d., trazodone, 100 mg h.s. as needed, and bupropion, 75 mg t.i.d. Serum levels of lithium and valproate were checked (0.9 mmol/ml and 87.0 µg/ml, respectively), bupropion was increased (100 mg t.i.d.), and trazodone was continued. Appointments in the medical clinic were arranged because Hope had not been receiving medical care for several months, despite continued complaints of gastrointestinal upset, tremor, and excessive vaginal bleeding. Neuropsycho­logical testing, single photon emission computed tomography, magnetic resonance imaging, and an evaluation in a memory disorders clinic ruled out a dementing process as the cause of Hope’s memory difficulties.

Over the several weeks it took to accomplish this workup, Hope’s mood and irritability improved dramatically. Weekly psychotherapy focused on support and psychoeducation, increasing Hope’s understanding about her psychiatric diagnosis, clarifying target symptoms for pharmacotherapy, and discussing the results of diagnostic testing.

Split Treatment

The problems in Hope’s treatment are worth highlighting. It appears that the first psychotherapists conducted a supportive, trauma-focused psychotherapy, presumably with a diagnosis of PTSD, while the physician was treating her for bipolar disorder. This probably contributed to her confusion. Like many individuals, Hope had difficulty understanding a model of mental life that divides her experience into “symptoms” to be reported to her physician and “problems” to be discussed with her psychotherapist. She became, understandably, increasingly frustrated with her physician and therapist as she remained ill for many months, and this reaction, along with depressive and anxiety symptoms, may have been misconstrued by the clinicians as a manifestation of borderline psychopathology. The complexity of her presentation, the presence of ongoing psychosocial stressors, and her inconsistent compliance with appointments all confounded the clinicians’ efforts. For all of these reasons, the recommendation that one clinician provide integrated care, and essentially become her primary physician, marks a turning point in the treatment course.

Because Hope’s new psychiatrist met with her weekly, he was able to appreciate the relationship between Hope’s affective lability (which had been the focus of previous pharmacotherapy) and her abuse history. Hope’s psychiatrist administered the Dissociation Evaluation Scale, on which Hope scored 53.2%.

Dissociation

Research suggests that prominent dissociative symptoms are associated with both a history of childhood trauma and general psychopathology (1, 2), although dissociative problems can also occur in the absence of significant childhood trauma (3). Initially, Hope’s flashbacks and intermittent loss of reality testing were considered to be consistent with psychotic affective illness and may have further misled her clinicians toward a diagnosis of bipolar disorder. Such misdiagnosis is by no means uncommon among trauma patients. Using the Dissociation Evaluation Scale helped her psychiatrist reformulate her diagnosis in terms of trauma rather than psychosis, since Hope’s score was several standard deviations above the mean (a score above 30 suggests dissociative psychopathology and is typical of patients with PTSD) (4, 5).

This patient dissociates in response to extreme anxiety or fear, including that due to activation of associative memory networks by real and symbolic stimuli that evoke trauma memories. If extreme states are modulated through either psychotherapy or pharmacotherapy, dissociative symptoms may diminish. It is noteworthy that as will be shown, the dissociative symptoms improved as PTSD symptoms improved, without making dissociation a focus of treatment.

Evaluation of Trauma History

Hope’s psychiatrist felt that the importance of trauma in Hope’s psychopathology may have been missed and decided to take a more thorough trauma history over several weeks. Hope saw the reconstruction of her trauma history as a difficult necessity and felt a new sense of purpose in the therapy.

Hope’s father began raping her when she was 6 years old and threatened her if she told anyone. Hope’s mother questioned him when she found Hope bleeding. The abuse stopped for 6 months but then resumed, continuing with a frequency of several times a week until Hope left home at age 18. The memory of bleeding after her father raped her was a particularly intrusive, recurring image. When she finally told her mother, she was called a liar. Hope also told school officials, who did nothing. She was held back in first grade, “coughed a lot” in school, and “froze” during tests.

Both parents frequently beat all the children with a belt to the point of bleeding. Hope would often submit to her father to protect her younger brothers and sisters.

Three months after her first husband kidnapped their 2-year-old son, Hope jumped out of a window, thinking she had heard her father’s voice, and was hospitalized for 8 months. This protracted stay was due to complications from a bleeding ulcer, requiring transfer to a medical facility. Her hospital roommate committed suicide during her stay. A month after her discharge, Hope married a second time. Her husband beat her severely, however, prompting her to leave him and make her second suicide attempt.

During the ensuing several years, Hope lived in four different states. She was hit by a car shortly after being released from the hospital and was raped twice. She has great gaps in her memory of this time, and her recall of events is often inconsistent. She remembers often feeling enraged and would pinch herself, pull her hair, or bang her head “to release the feelings.”

When Hope was in her early 30s, her mother, father, and five brothers and sisters were killed in an automobile accident on their way to a family reunion. She had recurring memories of the mangled bodies, which she had to identify. Hope then cared for her grandmother until her death. Again, Hope was raped and ended up in a home for battered women. She eventually decided to move to a new city after finding the address of a family friend in her parents’ belongings.

On her arrival, Hope found that the friend had moved away. After her belongings were stolen from a bathroom at the bus station, she moved into a homeless shelter and began working at a construction site, where she met her current husband.

Taking a Trauma History in the Diagnostic Evaluation

Taking a trauma history is important for any diagnostic evaluation and can facilitate the alliance by communicating the clinician’s interest in the whole person. Since disclosing traumatic events can provoke intense emotions of fear, anxiety, rage, or shame, the patient may not volunteer a trauma history. Clinicians can therefore specifically inquire about typical traumatic experiences and the patient’s reaction to them (6). In the National Comorbidity Survey, direct questioning about traumatic experiences led to a several-fold increase in the estimated lifetime prevalence of PTSD in the United States (7). If the patient reports a history of severe or recurrent trauma, the clinician can then assess the patient for diagnoses or clinical problems associated with severe trauma such as PTSD, dissociative disorders, major depression, borderline personality disorder, and prominent anniversary reactions.

Most patients feel relieved after disclosing traumatic experiences to a clinician who responds empathetically. However, individuals like Hope, who have never articulated such experiences, often need additional preparation and encouragement. An exploration of the patient’s fears regarding the disclosure can also be helpful. Individuals typically fear being overwhelmed with emotion, having a “breakdown,” or triggering an uncontrollable flood of horrible memories.

Particularly when recurrent trauma and its aftermath are viewed as the central feature of a patient’s problems, a list of diagnoses does not suffice to guide treatment. Significant traumatic experiences, particularly those perpetrated by family members, will almost certainly be relevant to a clinical formulation, in which the phenomenological presentation is integrated into an understanding of the meaning of the patient’s symptoms in the context of his or her life narrative.

When there have been multiple traumas over the life span, the individual’s characteristic adaptation to a predominantly hostile and unpredictable environment, and to attachments that are profoundly unreliable, will likely be observable in his or her approach to help seeking and capacity to make use of the clinician’s expertise. This is illustrated in Hope’s inconsistent and ambivalent engagement in treatment. However, Hope’s story also illustrates the fact that there is a wide range of outcomes to even severe childhood trauma. Her internal representation of dependent relationships certainly included fears of being exploited, betrayed, or treated with indifference. However, this view of others and the world did not entirely govern her behavior, since she was also described as likeable and engaging and showed no major manifestations of borderline or antisocial personality disorder.

New Diagnostic Formulation

Hope’s psychiatrist reformulated her diagnoses as PTSD, chronic; dissociative disorder, not otherwise specified; major depressive disorder, recurrent; and personality disorder, not otherwise specified. Her psychiatrist began to educate Hope about PTSD and the effect of trauma on cognition, emotion, and perception. Although she continued to experience intrusive and anxious symptoms when there was violence in the home, she said that she no longer felt “like a crazy person.” Because the treatment seemed to be just beginning, Hope’s psychiatrist offered to continue seeing her beyond the usual training clinic term of 1 year.

In retrospect, the patient’s prominent irritability, mood instability, agitation, and decreased sleep were mistakenly formulated as symptoms of mixed mania rather than PTSD. Since most patients with PTSD also meet criteria for comorbid disorders (7), the initial presentation may be a complex mix of affective, anxiety, and PTSD symptoms. Similarly, one particular aspect of PTSD (e.g., affective instability, aggressive outbursts, severe insomnia, multiple somatic complaints) may dominate the clinical picture.

DEVELOPING A TREATMENT PLAN

This patient presented the clinical challenge of developing a parsimonious treatment algorithm to address multiple diagnoses and problems. There is a paucity of scientific data to guide the clinical treatment of such patients, who represent a significant proportion of treatment-seeking individuals, have a more chronic course, and may be less treatment responsive (8).

Pharmacotherapy

Both lithium and valproate were slowly tapered off with no adverse consequences in Hope’s mood but with improvement in her tremor, gastro­intestinal upset, and nocturia. In order to give her current antidepressant regimen a full trial, bupropion was increased to 150 mg t.i.d. and trazodone was increased to 150 mg h.s. After 1 month with little perceivable benefit, both trazodone and bupropion were discontinued and fluoxetine (up to 40 mg/day) was prescribed to treat both the depressive and PTSD symptoms. Mood and intrusive symptoms nearly resolved within 2 months. With fluoxetine, Hope has been symptomatically stable for over a year, the longest period of time in her life.

Only insomnia persisted as a major symptom. Benzodiazepines, doxepin, and a trial of sleep deprivation provided little or at best temporary relief. Finally, olanzapine, 5 mg h.s., was prescribed, on the basis of its sedating properties and nonspecific effects on agitation. Olanzapine provided consistent benefit, and Hope eventually reduced her use to one to two times weekly.

Pharmacotherapy for PTSD

The antidepressants prescribed in the first 2 years of treatment (bupropion, trazodone in low doses) do not have established efficacy in PTSD (9). Case reports have suggested that valproate may be helpful for prominent affective instability, hyperarousal, and aggressive outbursts (common in a subset of PTSD patients) in refractory PTSD. Thus, the partial benefit that Hope experienced with valproate, which encouraged the physician to persist with a treatment plan for bipolar disorder, may have been due to mood stabilization (10).

In contrast, Hope experienced a dramatic improvement with fluoxetine, consistent with controlled and open trials of selective serotonin reuptake inhibitors in PTSD for core symptoms and some types of dissociative symptoms (11, 12). The latter is of interest, since dissociation has traditionally been viewed as unresponsive to pharmacotherapy.

There is no evidence to date in support of the use of traditional neuroleptics in PTSD. However, olanzapine is an antagonist at dopamine and serotonin (5-HT_2A) receptors, as well as other receptor sites, which may explain the observed benefits for insomnia and hyperarousal.

Family Therapy

Hope and her husband met in monthly sessions with Hope’s psychiatrist and a social worker who specialized in family therapy. The focus of these sessions was supportive, psychoeducational, and practical. For example, the couple was encouraged to find ways to decrease the amount of violence in the home. The therapists also helped Hope explain to her husband the link between her history of trauma and current symptoms. Finally, both partners were given the opportunity to express their feelings about the trauma and violence. Hope was moved to hear her husband express his own rage at his powerlessness to relieve her suffering and grew to trust him more. Her husband benefited not only by learning to be a better source of support for his wife, but also by having an outlet to express his own strong feelings. As their relationship improved, Hope and her husband demanded greater stability within the home, and the activities of the more aggressive members of the household were relegated to the outside.

Importance of a Systems Approach

Hope’s physician wisely identified her husband as a supportive figure and included him in the treatment plan. Involving a supportive family member in the treatment process can provide further validation of the patient’s reactions to trauma and enhance social support outside of the therapeutic relationship. In family therapy, the clinicians were able to model both and recommend more adaptive and constructive behavior for the patient and spouse. Individuals from abusive, neglectful family environments may have never experienced collaboration, openness, and constructive problem solving in intimate relationships. Similarly, the therapist can demonstrate a model of compassionate and informed understanding, as well as constructive limit setting if necessary, for the family members.

Psychotherapy

Concrete, supportive interventions—setting up initial medical appointments, teaching self-hypnosis, pursuing a memory workup—characterized Hope’s psychotherapy for nearly the entire first year of treatment. As the therapy progressed, however, Hope’s psychiatrist found that an exploratory approach could be extremely useful, without being any less supportive. One recurring theme—Hope’s reclaiming control of her life—became as important in the transference as it was in other aspects of her life. Hope related a dream in which she was riding in a cab to a bar she had frequented as a young woman called Speakeasy. She never gets to the bar because the cab driver drives her all over town. Hope felt frustrated and helpless in the dream, despite her feeling that the cab driver was avoiding the bar so she would not relapse on alcohol. The psychiatrist decided to interpret the dream in the context of trauma-focused therapy. He made a connection between himself and the cab driver and wondered if perhaps she felt she were occasionally being “driven” somewhere she did not want to go. This led to a discussion about ways Hope could feel more in control of her treatment. Several weeks later, Hope went back to the metaphor of the dream and envisioned an alternative ending in which she directs the cab driver to take her to the bar, where she drinks soda water and enjoys the company of old friends.

Hope actualized this newfound sense of agency several months later. Her psychiatrist had recommended she enroll in a support group for trauma survivors. Once Hope learned that detailed exploration of trauma histories was the focus of the group, she missed several sessions with her psychiatrist and, on returning, admitted to feeling anxious and guilty about not wanting to join the group. Hope’s psychiatrist decided that allowing Hope to control her own treatment was paramount. Hope felt relieved and gratified by her psychiatrist’s support of her choice and returned to her weekly psychotherapy sessions.

Integrative Approach to Psychotherapy

The psychosocial aspects of the last clinician’s work with Hope may be divided into several phases: 1) forming an alliance while simultaneously conducting an extended evaluation, 2) educating the patient about the formulation, 3) teaching behavioral interventions to manage anxiety, 4) conducting family therapy and problem-solving to promote a more successful adaptation to the patient’s chaotic environment, 5) reconstructing of a trauma narrative, and 6) exploring of conscious and unconscious meanings of traumatic experiences as they were continuing to influence her life.

The first several months of systematically investigating and addressing the patient’s problems following a medical model helped to solidify the treatment alliance.

The fact that the patient herself did not understand her problems as the consequence of multiple traumatic experiences is not uncommon in survivors of childhood trauma. Trauma-focused treatments with demonstrated efficacy in PTSD include an educational component in which the typical reactions to trauma are shared with the patient and illustrated in the patient’s own history and experience , 13). Education reassures the patient of the therapist’s knowledge and competence; normalizes the patient’s experience and thereby counters views of the self as weak or defective; identifies specific domains of symptoms and problems as a focus of treatment; and provides a credible, intuitively humane framework for both the psychotherapy and the patient’s self-understanding.

As illustrated in Hope’s treatment course, before productive exploration of the trauma, an alliance must be formed by establishing a professional relationship of relative safety and trust. This is important to counterbalance a tendency to misperceive exploration of the trauma as retraumatization. There may be a series of tests of the clinician’s persistence, concern, awareness of appropriate boundaries, ability to tolerate hearing about frightening experiences, and willingness to remain interested in the patient’s many complicated problems. For Hope, it appears that the eventually successful pharmacologic treatment was an additional, powerful proof of her physician’s competence to help her.

The technique of reconstructing a life narrative is particularly useful for patients with multiple, severe trauma. The technique essentially constitutes an extended exposure therapy treatment, which supports its efficacy but also raises important clinical issues, since exposure therapy is generally time limited because of its stressful nature. The clinician’s sensitivity to the patient’s ability to tolerate this endeavor, active support throughout the process, and flexibility can make the difference between a successful or disastrous trauma-focused therapy.

Hope’s decision not to join a therapy group of survivors of childhood sexual trauma is notable in this regard. She may have intuitively appreciated that such a group can be extremely taxing, and her ongoing struggles at home were difficult enough to tolerate. As the clinician recognized, the manifestation of negative transference at this point (i.e., fearing an angry, tyrannical response from the physician and loss of the relationship) in fact presented a valuable therapeutic opportunity. Recognizing her autonomy and responsibility for her own welfare and responding flexibly may have served as another corrective experience to counteract longstanding and understandable expectations of coercion and exploitation in dependent relationships.

MANAGING THE EMOTIONAL ASPECTS OF TREATING TRAUMA PATIENTS

The existential dilemmas intrinsic to clinical practice—particularly the necessity of taking action in the face of uncertainty—are brought into sharp relief in the care of trauma survivors. Resisting the temptation to embrace dogmatic formulations in the service of reducing one’s anxiety is crucial to remaining flexible and responsive. It can be difficult to persist with a full exploration of the patient’s traumatic experiences, but abandoning this pursuit often represents collusion with the patient’s avoidance and may ultimately prolong suffering.

Recognizing that a difficult patient is a survivor of severe trauma can be particularly helpful in managing countertransference reactions to noncompliant, aggressive, tentatively engaged, or otherwise frustrating patients. At the same time, a conscientious therapist may feel guilty for having such understandably human responses. To strike a balance between working actively to help the patient and accepting one’s limitations, seeking support and consultation when necessary, is widely recommended by clinicians of all levels of experience.

CONCLUSIONS

Currently, Hope still faces many problems. Nevertheless, because the role that trauma has played in her life has been acknowledged by Hope, her family, and her doctor, she appears better equipped than ever to face them. A recent excerpt from her writing illustrates her newfound sense of purpose.

Hope is finally altering her lifelong pattern of despair and powerlessness. As Hope has learned to categorize and understand her fears, she has begun to master her symptoms and to make genuine progress toward recovery.

Received Aug. 13, 1998; revision received Dec. 8, 1998; accepted Dec. 22, 1998. From the Department of Psychiatry, Columbia University College of Physicians and Surgeons/New York State Psychiatric Institute. Address reprint requests to Dr. Marshall, Anxiety Disorders Clinic, 1051 Riverside Dr., New York, NY 10032