«Behaviour Research and Therapy 44 (2006) 807–817 A pilot study of two-day cognitive-behavioral therapy for panic ...»
ARTICLE IN PRESS
Behaviour Research and Therapy 44 (2006) 807–817
A pilot study of two-day cognitive-behavioral therapy for
Brett Deacona,Ã, Jonathan Abramowitzb
Department of Psychology, University of Wyoming, Dept. 3415, 1000 E. University Ave., Laramie, WY 82071, USA
Department of Psychiatry & Psychology, Mayo Clinic, 200 First St. SW, Rochester, MN 55905, USA Received 14 February 2005; received in revised form 13 May 2005; accepted 23 May 2005 Abstract The present study investigated the short-term efﬁcacy of brief, intensive cognitive-behavioral therapy (CBT) for panic disorder (PD). The treatment involved 9 h of therapist contact over two consecutive days and was developed for the purpose of delivering CBT for PD to a largely rural patient population that must travel long distances to ﬁnd a treatment provider. Ten patients who elected to participate in brief, intensive CBT instead of weekly CBT were recruited from routine clinical practice in a hospital-based anxiety disorders clinic. Patients were not excluded based on the presence of agoraphobia, diagnostic comorbidity, concurrent use of PRN benzodiazepine medications, or previous nonresponse to psychotherapy for PD.
Assessments conducted at pre-treatment and 1-month follow-up revealed large, clinically signiﬁcant reductions in PD symptoms, anxiety sensitivity, body vigilance, and anxiety and depressive symptoms.
Most patients (60%) were panic-free after treatment and evidenced normative levels of symptomatology at follow-up. The present study suggests that brief, intensive treatment may be an effective means of delivering CBT for PD.
r 2005 Elsevier Ltd. All rights reserved.
Keywords: Panic disorder; Cognitive-behavioral therapy; Exposure; Brief treatment; Psychotherapy ÃCorresponding author. Tel.: +1 307 766 3317; fax: +1 307 766 2926.
E-mail address: email@example.com (B. Deacon).
0005-7967/$ - see front matter r 2005 Elsevier Ltd. All rights reserved.
ARTICLE IN PRESSB. Deacon, J. Abramowitz / Behaviour Research and Therapy 44 (2006) 807–817 808 Introduction Cognitive-behavioral therapy (CBT) involving interoceptive exposure is the psychological treatment of choice for panic disorder (PD) (Barlow, 2002). This treatment, when delivered in 12–15 weekly sessions, produces substantial and durable reductions in PD symptoms (Addis et al., 2004; Barlow, Gorman, Shear, & Woods, 2000) and, relative to pharmacotherapy, appears more cost-effective (Heuzenroeder et al., 2004), acceptable and preferable to patients (Deacon & Abramowitz, in press), and less likely to result in attrition (Hofmann et al., 1998). Despite the established efﬁcacy and effectiveness of CBT, many patients with PD are unable to beneﬁt from this treatment; for example, individuals living in underserved rural settings who must commute long distances for weekly appointments. This extra travel time can create a strain on time and ﬁnancial resources, leading to treatment refusal. In the present study, we examined the effectiveness of a brief (2-day), intensive variant of CBT for PD that might be well suited for patients and treatment providers in settings where the aforementioned barriers to obtaining effective treatment exist.
A growing body of research has examined the efﬁcacy of various methods for abbreviating standard treatment. Studies examining bibliotherapy (e.g., Gould, Clum, & Shapiro, 1993), computer-guided self-exposure (Marks, Kenwright, McDonough, Whittaker, & Mataix-Cols, 2004), internet-based treatment (e.g., Carlbring, Westling, Ljungstrand, Ekselius, & Andersson, 2001), and teletherapy (e.g., Swinson, Fergus, Cox, & Wickwire, 1995) indicate that reduced therapist contact interventions may be viable options for many individuals with PD.
These studies highlight the possibility raised by stepped care models (Newman, 2000) that brief CBT might serve as a ﬁrst-line treatment for patients who are likely to beneﬁt from minimal interventions.
Although reducing therapist contact makes therapy more affordable and minimizes the inconveniences associated with frequent ofﬁce visits, the duration of these interventions does not differ appreciably from that of standard CBT in most studies (e.g., Cote, Gauthier, Cormier, & Plamondon, 1994). As a result, such interventions might not make treatment more accessible to patients who lack sufﬁcient time or who desire more immediate symptom reduction. Few studies have examined brief CBT approaches that include the essential features of CBT for PD: (a) education, (b) cognitive restructuring, (c) therapistassisted interoceptive exposure, and (d) therapist-supervised in vivo exposure (e.g., Schmidt, 1999). Fewer still have compared reduced therapist contact interventions to this ‘‘gold standard’’ CBT. Perhaps not surprisingly, the most consistently effective brief treatments for PD are those that emphasize these procedures (e.g., Clark et al., 1999). Notably, several studies indicate that very brief, intensive, exposure-based interventions produce outcomes comparable to standard CBT in a matter of weeks (Westling & Ost, 1999) or even days (Evans, Holt, & Oei, 1991).
In the present study, we describe a novel, 2-day, therapist-directed exposure-based CBT approach for PD that was developed to serve a largely rural patient population. Pilot efﬁcacy data are presented from a sample of PD patients treated in routine clinical practice. Although this study was exploratory in nature, based on previous research we hypothesized that brief CBT would produce clinically signiﬁcant reductions in PD symptoms from pre-treatment to 1-month follow-up.
ARTICLE IN PRESS
Method Participants Ten adults (eight women and two men, all of whom were Caucasian; mean age ¼ 38.4 years;
SD ¼ 11.5; range ¼ 26–62) meeting DSM-IV-TR criteria for PD with agoraphobia (n ¼ 5) and PD without agoraphobia (n ¼ 5) were recruited from a multidisciplinary anxiety disorders clinic within a large academic medical center.
The sample was well-educated: four participants had attended some college and ﬁve had earned at least a bachelor’s degree. Median annual family income was between $50,000 and $60,000 per year. Seven participants had full-time jobs, one was a full-time college student, and two were retired. Eight participants were married or living with a partner. In order to be included in the present study, patients had to have a principal diagnosis of current PD and express a preference for brief CBT rather than standard (i.e., weekly) CBT.
Exclusion criteria included having an untreated substance use disorder, a psychotic disorder diagnosis, current suicidality, concurrent involvement in psychotherapy for PD, or seeking treatment for a problem other than PD (e.g., a different anxiety disorder or depression).
Following their initial clinic assessment, patients who wished to participate in CBT were informed about, and asked to select from, two approaches: a standard, once-weekly meeting schedule and a 2-day, intensive approach (described in Section 2.4). Of the 14 PD patients assessed during the 12-month study period, four selected standard CBT while 10 opted for brief CBT. Primary reasons for choosing brief CBT over standard CBT were desire for rapid improvement (n ¼ 5), rural residence and no access to a local CBT provider (n ¼ 4), and returning to college in 1 month (n ¼ 1).
The mean duration of PD in the sample was 53.9 months (SD ¼ 73.8; range ¼ 1–240).
Responses to item 1 of the panic disorder severity scale (PDSS; see below) indicated that in the past month, ﬁve patients experienced an average of less than one panic attack per week, one patient averaged two attacks per week, one patient averaged more than two per week, and three patients experienced at least one panic attack each day. Three patients had additional, current diagnoses, including one with generalized anxiety disorder and hypochondriasis, one with hypochondriasis and major depressive disorder, and one with social phobia. Seven patients were currently taking SSRI medications, and four were also taking a benzodiazepine. Two patients had never taken medication for PD, and one had previously taken numerous antidepressant and benzodiazepine medications but was currently medication-free. Four patients had previously participated in psychotherapy for PD (relaxation training or biofeedback in each case), whereas six had never had psychological treatment.
All patients had an initial assessment including a 1.5-h semi-structured diagnostic interview with a clinical psychologist using the Mini International Neuropsychiatric Interview (MINI) (Sheehan et al., 1998), which is easily integrated into clinical practice, has good reliability and validity, and a high concordance rate with the SCID diagnosis of PD (kappa ¼.76) (Sheehan et al., 1998). The assessor also conducted a functional analysis of the patient’s panic symptoms and administered PDSS (Shear et al., 1997), which assesses the overall severity
ARTICLE IN PRESS
B. Deacon, J. Abramowitz / Behaviour Research and Therapy 44 (2006) 807–817810
of PD and agoraphobic avoidance in the past month. At the end of the assessment, each patient was provided with feedback about their diagnosis and treatment options. This feedback included discussion of the cognitive-behavioral conceptualization of PD and description of CBT.
Several well-studied self-report instruments, with good psychometric properties, that assess the symptoms of PD, agoraphobia, depression, general anxiety, and panic-related cognitive phenomena, were also administered during the pre-treatment assessment. These included the panic and agoraphobia scale (PAS) (Bandelow, 1999), anxiety sensitivity index–revised (ASI-R) (Taylor & Cox, 1998), body vigilance scale (BVS) (Schmidt, Lerew, & Trakowski, 1997), Beck depression inventory (BDI) (Beck, Ward, Mendelsohn, Mock, & Erlbaugh, 1961), and Beck anxiety inventory (BAI) (Beck, Epstein, Brown, & Steer, 1988).
The PDSS, PAS, ASI-R, BVS, BDI, and BAI were also administered at 1-month follow-up.
The follow-up PDSS interview was conducted over the telephone.
All patients provided informed consent to participate in the study. Each received a copy of the Mastery of Your Anxiety and Panic—3rd Edition client workbook (MAP-3) (Barlow & Craske,
2000) following the initial assessment and was instructed to read selected chapters covering education, cognitive restructuring, and exposure therapy. While they were encouraged to learn the material to facilitate treatment, they were not instructed to do anything beyond simply reading the chapters. The date of the ﬁrst treatment session was scheduled approximately 2 weeks after the initial assessment for each patient.
One month after completing treatment, patients were contacted via telephone by the ﬁrst author who again administered the PDSS. Patients also completed and mailed the study questionnaires at this time. All 10 patients who started the treatment completed it and were assessed at 1-month follow-up.
Brief, intensive CBT was modeled after the 12-session protocol developed by Telch and Schmidt (1990), except that it was delivered over two consecutive days in the present study. The ﬁrst treatment session included 6 h of face-to-face therapist time and began with a review of the cognitive-behavioral conceptual model of PD. Patients learned about the role of catastrophic beliefs about the dangerousness of anxiety-related body sensations in causing panic attacks.
Safety-seeking and avoidance behaviors were described as maintenance factors that prevented patients from learning that their catastrophic panic-related beliefs were inaccurate. The primary goal of CBT was described as helping patients acquire more accurate beliefs about the actual dangerousness (or lack thereof) of their panic-related body sensations.
Next, the therapist and patient reviewed the cognitive, behavioral, and physiological features of the ﬁght-or-ﬂight response, with an emphasis on the unpleasant but harmless nature of the patient’s feared panic-related body sensations. To facilitate cognitive restructuring, patients were taught a step-by-step method for identifying ‘‘threat forecasts’’ (i.e., catastrophic appraisals of panic symptoms) and evaluating the evidence for their likelihood and severity. The therapist and
ARTICLE IN PRESS
B. Deacon, J. Abramowitz / Behaviour Research and Therapy 44 (2006) 807–817 811
patient subsequently reviewed in detail each of the patient’s panic-related threat forecasts (e.g., heart attack, passing out, suffocation). For each threat forecast, the actual likelihood and severity was estimated by drawing on evidence from educational material and the patient’s past experiences. Together, education and cognitive restructuring lasted approximately 2 h and was primarily designed to facilitate exposure by generating doubt about the veracity of the patient’s primary threat forecasts.
The remainder of the ﬁrst session was devoted to therapist-assisted exposure. Following a discussion of the rationale and procedures of exposure, patients brieﬂy participated in nine different interoceptive exposure exercises (e.g., hyperventilation, spinning in a swivel chair, running in place) and subsequently provided verbal SUDS ratings. The purpose of this assessment was to create a fear hierarchy consisting of the patient’s most feared interoceptive stimuli. Following an hour-long lunch break, the next several hours were spent conducting exposures to the patient’s two or three most feared exercises. Each interoceptive exposure continued until the patient’s verbal SUDS were reduced by at least 50% and until the patient reported being convinced that the exercise was not dangerous. The therapist framed the exposures as behavioral experiments and helped patients identify their threat forecast(s) prior to each exercise and reﬂect on their accuracy after completing the exposure. Patients were also instructed not to use safety behaviors during the exercises, and the therapist identiﬁed and discouraged such behaviors (e.g., drinking water) when they were apparent.