TreatmentofPanic - British Journal of Psychiatry (1994),...

Info iconThis preview shows pages 1–11. Sign up to view the full content.

View Full Document Right Arrow Icon
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 2
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 4
Background image of page 5

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 6
Background image of page 7

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 8
Background image of page 9

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 10
Background image of page 11
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: British Journal of Psychiatry (1994), 164, 759-769 A Comparison of Cognitive Therapy, Applied Relaxation and lmipramine in the Treatment of Panic Disorder DAVID M. CLARK, PAUL M. SALKOVSKlS, ANN HACKMANN, HUGH MIDDLETON, PAVLOS ANASTASIADES and MICHAEL GELDER Recent studies have shown that cognitive therapy is an effective treatment for panic disorder. However, little is known about how cognitive therapy compares with other psychological and pharmacological treatments. To investigate this question 64 panic disorder patients were initially assigned to cognitive therapy, applied relaxation, imipramine (mean 233 mg/day), or a 3-month wait followed by allocation to treatment. During treatment patients had up to 12 sessions in the first 3 months and up to three booster sessions in the next 3 months. lmipramine was gradually withdraWn after 6 months. Each treatment included self-exposure homework assignments. Cognitive therapy and applied relaxation sessions lasted one hour. lmipramine sessions lasted 25 minutes. Assessments were before treatment/wait and at 3. 6, and 15 months. Comparisons with waiting-list showed all three treatments were effective. Comparisons between treatments showed that at 3 months cognitive therapy was superior to both applied relaxation and imipramine on most measures. At 6 months cognitive therapy did not differ from imipramine and both were superior to applied relaxation on several measures. Between 6 and 15 months a number of imipramine patients relapsed. At 15 months cognitive therapy was again superior to both applied relaxation and imipramine but on fewer measures than at 3 months. Cognitive measures taken at the end of treatment were significant predictors of outcome at follow-up. Cognitive therapy is a recently developed brief psychological treatment for panic. It is based on the cognitive theory of panic disorder (Beck et al, 1985; Clark, 1986; Ehlers er al, 1986; Salk0vskis, 1988) which proposes that individuals who experience re- current panic attacks do so because they have an enduring tendency to misinterpret benign bodily sensations as indications of an immediately impending physical or mental catastrophe. For example, palpitations may be interpreted as evidence of an impending heart attack. This cognitive abnormality is said to lead to a positive feedback loop, in which misinterpretations of bodily sensations produce increases in anxiety which in turn strengthen the sensations, producing a vicious circle which culminates in a panic attack. Recent experimental studies supporting the cog- nitive theory are reviewed by Clark & Ehlers (1993) and McNally (1990). The main clinical implication of the theory is that it suggests naturally occurring panic attacks can be treated by interventions which help patients to identify and change their mis- interpretations of bodily sensations. Several cognitive- behavioural treatments which attempt to achieve these goals have been devised, including the panic control treatment devised by Barlow, Craske and colleagues (Barlow & Cerny, 1988; Barlow & Craske, 1989) and the cognitive therapy devised by Clark, Salk0vskis, Beck and colleagues (Clark, 1989; Salk0vskis & Clark, 1991). An early indication that cognitive-behavioural treatments may be effective in panic was provided by nine uncontrolled case series (Barlow er a1, 1984; Clark at al, 1985; Salk0vskis er al, 1986; Kopp er al, 1986; Shear et a], 1988; Sokol et a], 1989; Michelson e! a], 1990; Shear er a1, 1991; Welkowitz er al, 1991). In each series, consecutively referred panic disorder patients showed substantial reductions in panic frequency and associated symptoms. Two of the studies (Clark et a], 1985; Sokol er a1, 1989) investigating cognitive therapy also reported long- term (1-2 years) follow-up and both found that the gains made in treatment were maintained at followup. The results of these case series have been con- firmed in three controlled trials. Beck et al (1992) allocated panic patients to 12 weeks of cognitive therapy or 8 weeks of supportive therapy. When assessed at comparable time points (4 and 8 weeks), patients given cognitive therapy had improved signifi- cantly more than those given supportive therapy, suggesting that the effectiveness of cognitive therapy is not entirely attributable to non-specific therapy fac- tors. Barlow et al (1989) compared their panic control treatment with another cognitive treatment (cognitive restructuring plus exposure), progressive muscle 759 6914013476 760 relaxation training, and a waiting-list control group. Both cognitive treatments were consistently superior to waiting-list and were significantly more effective than relaxation in reducing panic frequency but not generalised anxiety. Klosko et a! (1990) compared panic control treatment with alprazolam, placebo and a waiting-list control group. Panic control treatment was significantly more effective than placebo and waiting-list, but did not differ signifi- cantly from alprazolam, which in turn did not differ significantly from placebo. The present study provides a further investigation of the effectiveness of cognitive therapy and addresses three specific questions. The first question concerns the role of non-specific factors in determining therapy outcome. Although the finding of Beck et a! (1992), that cognitive therapy was significantly more effective than supportive therapy, suggests that the effects of cognitive therapy are not entirely due to non~specific factors, they did not assess treatment credibility. Cognitive therapy may have been more effective than supportive therapy because it was more credible. To clarify this issue, patients in the present study rated the credibility and expected effectiveness of each treatment. The second question is how cognitive therapy compares with alternative, established psychological and pharmacological treatments for panic disorder. Two alternative treatments were chosen for in- vestigation, each of which had been shown to be more effective than placebo treatment. The psychological treatment was Cst’s (1987) applied relaxation treat- ment, in which training in a specialised and portable relaxation technique is combined with exposure to feared situations. In a recent controlled trial (Ost, 1988) applied relaxation was significantly more effec tive than traditional progressive muscle relaxation training plus exposure. The pharmacological treat- ment, imipramine, was chosen because it has re- peatedly been shown to be more effective than placebo (Fyer & Sandberg, 1988) and is as effective as alprazolam (Cross-National Panic Study, 1992), the main alternative pharmacological treatment. The third question concerns the long-term effective- ness of treatment. The cognitive theory of panic predicts that sustained improvement after the end of any treatment (whether psychological or pharma- cological) will depend on cognitive change having occurred during the course of therapy (Clark, 1986). To test this prediction, misinterpretations of bodily sensations were measured at the end of treatment and related to outcome at follow-up. Previous work from our group (Mathews et a], 1981), and others (Jansson & Ost, 1982; Marks et a1, 1993; Michelson & Marchione, 1991), has shown CLARK ET AL that exposure in vivo is an effective treatment for patients with marked agoraphobia, and much of the initial impetus behind the development of cognitive therapy arose from the need for a psychological treatment for the less phobic panic disorder patients whose attacks were thought unlikely to be completely eliminated by situational exposure alone. The present study includes a substantial number of such patients. Method Design Patients were initially assigned to cognitive therapy, applied relaxation. imipramine, or a waiting-list. Patients receiving cognitive therapy or applied relaxation had up to 12 sessions of treatment in the first 3 months and up to three booster sessions in the next 3 months. Patients receiving imipramine were given a similar number of sessions during the first 3 months and then maintained at maximum dose for a further 3 months, after which medication was withdrawn gradually. Patients in the waiting-list group received no treatment for 3 months, after which they were assigned randomly to one of the three treatments. Assessments, which included ratings completed by an assessor who was blind to treatment allocation, were at pre-treatment/waiting-list, 3, 6, and 15 months. Patients All Oxfordshire general practitioners, psychiatrists, and psychologists were sent a letter requesting referrals for a study of psychological and pharmacological treatments for panic disorder. It was explained that three treatments were being investigated, all had been shown to be effective, and that it was not known which was most effective. There were no competing treatment centres in the area. Referrals were assessed using the Structured Clinical Interview for DSM- 111-R(SClD; Spitzer & Williams, 1986) and accepted if they met the following criteria: (a) DSM-lll-R (American Psychiatric Association, 1987) panic disorder with no, mild, or moderate agoraphobic avoidance. (b) Current episode duration at least 6 months (this criterion was intended to minimise spontaneous re- mission; a previous study (Catalan et al, 1984) of recent-onset anxiety disorders suggested spontaneous remission is common in the first 6 months). (c) At least three panic attacks in the last 3 weeks. (d) Consider panic their main problem. (e) Age 18—60 years. (1') Willing to accept random allocation. (g) No depressive disorder severe enough to require immediate psychiatric treatment. (it) No cognitive therapy, applied relaxation or imi- pramine in the current episode. (i) No evidence of organic mental disorder, schizo- phrenia, alcohol or drug dependence, cardiovascular disease, asthma, epilepsy, pregnancy or intention to become pregnant. COMPARISON OF TREATMENTS OF PANIC DISORDER Concurrent Axis 11 personality disorder was not a reason for excluding patients unless personality disorder was clearly the primary problem. Of 246 panic disorder patients referred for possible in- clusion, 174 did not med study Reasons for exclusion were: too short duration (28 patients); too low panic frequency (76); severe agoraphobic avoidance (18); concurrent pregnancy/physical disorder (17); previous cognitive- behavioural (7) or tricyclic (10) therapy in current episode; and refused randomisation (18). Of 72 patients meeting acceptance criteria, 3 dropped out (1 per treatment). PM who agreed initially to random allocation refused to take imi— pramine when allocated to that condition. Dropouts and refusers after randomisation were replaced and not included in the data analysis. To be classified as a drop—out, patients had to start treatment but attend no more than two sessions. Patients who attended at least three sessions were considered completers and included in all analyses. Treatments Each treatment comprised up to 12 weekly sessions in the first 3 months and up to three booster sessions in the next 3 months. Patients receiving cognitive therapy or applied relaxation received no further sessions after this time, but patients receiving imipramine were seen occasionally to monitor drug withdrawal. Mean numbers of sessions in cognitive theraDY, applied relaxation, and imipramine were 10.5, 10.3, and 10.5 respectively during the first 3 months; 2.2, 2.0, and 1.6 between 3 and 6 months; and 0.4, 1.5, and 2.8 after 6 months. The treatments did not differ in number of sessions during the first 6 months but imipramine— treated patients had significantly (P< 0.01) more sessions than cognitive therapy patients after 6 months. This is partly because the imipramine patients were seen in order to monitor drug withdrawal and partly because a greater number relapsed and required further treatment during the follow-up period (see Results section). To be classified as relapsed, patients had to have deteriorated sufficiently to meet again the panic frequency criteria for inclusion in the trial. (Relapsed patients received a further course of their original treatment.) The average length of sessions was significantly longer (P<0.05) for cognitive therapy (58 min) and applied relaxation (57 min) than for irnipramine (24 min). This difference is in line with usual clinical practice. Several procedures were common to all treatments. (a) Before the first session, patients were given a booklet with reassuring information about panic (it is a common problem, the symptoms are not dangerous and are highly treatable), a rationale for their treatment, and an overview of the treatment. Borkovec & Nau’s (1972) scale was used to check the credibility of treatment. At the end of the second session patients rated the extent to which they thought their treatment was logical, would be successful, and they would recommend it to a friend. (b) Throughout treatment. patients kept a daily diary covering situations in which the attacks occurred and the symptoms experienced in each attack. The diary included the full DSM-lll-R definition of a panic 761 attack and this was clarified with patients during the screening assessment. (c) During the first 4 weeks of treatment, patients were asked not to expose themselves to feared situations more than usual. They were told that this would allow time for them to develop confidence that the treat- ment was helping control their panics. (d) From week 5 onwards, patients who avoided situations through fear of panic attacks were given weekly instructions to enter these situations as pan of their homework practice. Self-exposure instructions were given equal emphasis in each treatment. Cognitive therapy (CT) was based on the cognitive theory of panic. Several cognitive and behavioural techniques (see Clark, 1989; Salkovskis & Clark, 1991) were used to help patients identify and change misinterpretations of bodily sensations. Cognitive procedures included: identifying and challenging patients’ evidence for their misinterpretations; substituting more realistic interpretations; and restructuring images. Behavioural procedures included: inducing feared sensations (by hyperventilation, focusing attention on the body, or reading pairs of words representing feared sensations and catastrophes) in order to demonstrate possible causes of patients' symptoms; and stopping safety behaviours (such as holding onto solid objects when feeling dizzy) in order to help patients disconfirm their negative predictions about the consequences of their symptoms. Homework assignments also included keeping a daily record of negative thoughts and rational responses, and conducting behavioural experiments to test these thoughts. Applied relaxation (AR) was a modified version of the relaxation training devised by 0st and described in detail by OS! (1987) and Clark (1989). Patients were taught to identify the early signs of panic and to relax rapidly. As panic attacks can occur in any situation, the training consists of a series of stages in which patients are taught to relax more and more quickly while performing everyday activities such as walking and shopping. Homework assignments in- cluded twice-daily relaxation practice. There were two modifications to the ori inal treatment. The first was a change in rationale. 1n st's original treatment, patients were told that panic attacks result from a vicious circle in which the physical symptoms of anxiety are augmented by negative thoughts. To reduce overlap with the cognitive therapy rationale, a more behavioural rationale was sub- stituted. Patients were told that panic attacks occur because the body’s ‘alarm system’ has become hypersensitive as a consequence of chronic stress and responds to slight stressors as if a severe danger is present. This hyper- sensitivity could be eventually eliminated by learning how to use relaxation to control the response as soon as it starts and to lower the heightened levels of general tension which are the basis of the hypersensitivity. The second change to the treatment concerned the time at which exposure to feared situations was introduced. in Ost‘s original version exposure is not introduced until all the stages of relaxation training have been completed (after 8 to 10 sessions). To ensure compatibility with the other treatments, we intro- duced exposure after four sessions. lmipramine (IMIP). The rationale for this treatment was that panic attacks result from a central neurochemical 762 abnormality which causes the body’s ‘alarm system’ to be hypersensitive and this abnormality can be corrected by appropriate pharmacological intervention. A problem with the use of imipramine for panic disorder is that some patients report an initial increase in tension and arousal (Fyer & Sandberg, 1988; Nutt & Glue, 1989). in a pilot study, we found that this effect could be avoided by using a small initial dose followed by gradual increments. For this reason, patients started on 10 mg. The dose was increased in 10 mg steps every 3 days up to 60 mg, then to 75 mg and then in 25 mg steps until either panic ceased or 300 mg was reached, as recommended by Fyer & Sandberg (1988). Mean prescribed dose was 233 mg/day (s.d. 49), reached after a mean of 7.5 weeks (s.d. 2.4). To check compliance, an unexpected blood test was carried out shortly after reaching maximum dose. For all patients the test was positive. Mean plasma levels of imipramine, desipramine, and total imipramine plus desipramine were 115 ng/ml (s.d. 99), 124 ng/ml (s.d. 100), and 239 ng/ml (s.d. 149) respectively. These values are within the expected range for an average intake of 233 mg/day. Once at maximum dose, patients were maintained at that dose until the 6-month assessment, after which the medication was withdrawn gradually over two to three months. Treatment adherence. To check therapists’ adherence to the treatment protocol ten audiotapes per treatment (each from a different patient) were randomly selected and rated for the presence/absence of features which should be unique to that treatment (eliciting and challenging panic-related beliefs; behavioural experiments; relaxation training; re- laxation homework; medication administration; monitoring and explanation of side-effects) and for time spent on procedures which should be common to all treatments (reviewing symptoms; providing support; exposure home- work assignments). There were no protocol violations and the treatments did not differ in times spent on the common procedures. Extra-trial medications. In any trial of treatments for anxiety, a proportion of patients will be taking psychotropic medication before entry into the trial. One way of dealing with this problem is to withdraw all such medications and ask patients not to re-introduce them during the trial. Another is to accept only those medicated patients who have been on a stable dose for several months without showing any further impr0vement and to require them not to increase their medication during the trial. We chose the second strategy and required a minimum steady state period of three months. Our reasons for preferring this strategy were twofold. First, abrupt withdrawal of psychotropic medication in panic patients can induce a transient increase in anxiety and panic. Such an effect would tend to overestimate the size of pre- to post-treatment changes, while under- estimating differences between treatments. Second. not all panic patients are able to withdraw from psychotropic medication, and those who cannot tend to be the more severe cases. lnsisting on withdrawal prior to inclusion may therefore produce a biased sample in which severe cases are under-represented. Percentages of patients taking medications prior to the start of treatment or wait were: 35% (7/20) for C1‘; 40% (8/20) for AR; 30% (6/20) for lMlP; and 43% (7/16) for WL (waiting-list). None of these differences CLARK ET AL are significant (P>0.5). The vast majority of medications were fi-blockers and low-potency benzodiazepines. No patient was taking imipramine or alprazolam, though three were taking another tricyclic at low dose (imipramine equivalents <50 rug/day). Patients were questioned in detail about medication usage during assessment interviews but urine samples were not taken. Only two patients (both AR) reported increasing medication during treatment or wait. Several patients, including all three on low-dose tricyclics, stopped their extra-trial medication during treatment or wait (4/7 CT; 1/9 AR; 3/6 IMIP; 1/7 WL). Therapists The psychological treatments were administered by two clinical psychologists (PS and AH). Each treated an equal number of patients with cognitive therapy and applied relaxation. Both had extensive prior experience in the use of cognitive and behavioural treatments for anxiety. lmipramine was administered by a psychiatrist (HM) who had extensive experience in the pharmacological treatment of anxiety and had also received training in various forms of psychotherapy. Before starting the trial, all three therapists were given further specific training in the treatments being studied. This training involved regular supervision and listening to tape recordings of treatment sessions while treating at least ten pilot cases. Regular supervision continued throughout the trial with supervision for the psychological treatment being given by another experienced clinical psychologist (DMC) and supervision for pharma- cological treatment being given by a senior psychiatrist (MG). Measures A comprehensive battery of panic and anxiety measures was used. Panic attacks. Frequency of panic attacks and panic- related distress/disability were assessed on separate rating scales covering the last two weeks. Each scale was completed by both patient and assessor. Panic frequency was rated on a five-point scale where 0 represented ‘no panic attacks’, 1 ‘one panic attack per fortnight’, 2 ‘one or two panic attacks per week’, 3 ‘at least three panic attacks per week but less than one per day’ and 4 ‘one or more panic attacks per day’. The DSM-lll-R definition of a panic attack was printed above the scale. Frequency of limited symptom attacks was not assessed. Panic-related distress/disability was rated on a 9-point scale where 0 represented ‘not at all disturbing and/or disabling’, 2 ‘slightly', 4 ‘definitely‘, 6 ‘markedly’ and 8 ‘very disturbing/disabling’. A further panic measure was derived from patients’ diaries by counting the number of patients who recorded zero panic attacks in the two weeks before each assessment. General anxiety was assessed by the Beck Anxiety inventory (BAl; Beck et a1, 1988) and by the Hamilton Anxiety Rating Scale (HARS; Hamilton, 1969). The HARS was intended as a measure of general anxiety and tension. For this reason the assessor excluded panic episodes from the rating. Patient and assessor also completed a 9-point rating of general tension and anxiety. COMPARISON OF TREATMENTS 0F PANIC DISORDER Avoidance was assessed by a modified version of the Fear Questionnaire (Marks & Mathews, 1979) which has three avoidance subscales (agoraphobia, social phobia, and blood-injury phobia). Each subscale item was rated on a 4-point scale where 0 represented ‘never’, 1 ‘sometimes’, 2 ‘often‘ and 3 ‘always’. Situational fear and fear ofbodily sensations. Situational fear was assessed by the modified Marks & Mathews Questionnaire with the instructions changed to cover fear instead of avoidance. Fear of bodily sensations was assessed by the Bodily Sensations Questionnaire (BSQ; Chambless el al, 1984). The anchors on the situational fear scales were the same as those on the BSQ. Panic-related cognition. Two measures of panic-related cognition were used. The Body Sensations Interpretation Questionnaire (BSIQ; Clark et al, 1988) panic scale assesses misinterpretation of a range of bodily sensations and has been shown to discriminate panic patients from other anxious patients. The Agoraphobic Cognidons Questionnaire (ACQ; Chambless er al, 1984) is a checklist of thoughts which also discriminates panic patients from other anxious patients. Depression was also assessed, by using the Beck Depression Inventory (BDI; Beck er al, 1961). Statistical analysis To identify any differences between groups before treatment or wait, initial scores for the three treatment groups and waiting-list were compared with the liberal procedure of separate one-way analyses of variance (ANOVAs) for each measure. To identify any differences between the groups after treatment or wait, one-way analyses of covariance (ANCOVAs) were performed on the data collected at 3, 6. and 15 months using initial scores as c0variates. Dunam’s multiple comparison procedure was used to identify the source of any significant between-group differences. Because of the large number of panic/anxiety outcome measures a two-stage procedure was used to control experiment-wise error. First, a single unweighted composite panic/anxiety measure was created and analysed by ANCOVA. Only if the composite revealed significant between-group differences were further ANCOVAs performed on individual panic/anxiety measures. The composite was generated by the procedure recommended by Rosenthal & Rosnow (1991) and Hollon et al (1992). Patients’ scores on each (of 17) panic/anxiety measures were standardised (mean = 0, s.d. 1) across all assessment points by converting to 2 scores, and then averaged across measures. Dichotomous variables were analysed using 763 Fisher’s exact test. In comparisons between treatments and waiting-list, sample size is 16 per group. In comparisons between the three treatments, sample size is 20 per group as these comparisons also include all patients (12) who continued to meet trial inclusion criteria at the end of the waiting period and were then allocated to treatment. Reauhs The mean age of the 64 patients included in the trial was 34.6 years (s.d. 9.2) and the mean duration of the current episode of panic disorder was 3.3 years (range 0.5-15). Seventy-eight per cent were female; 19% had no agora- phobic avoidance, 48% had mild agoraphobic avoidance, and 33% had moderate agoraphobic avoidance. Among patients with no agoraphobic avoidance. 59% of panic attacks occurred at home and attacks outside the home were not confined to agoraphobic situations. There were no significant differences between groups on any of these characteristics. Solubility of treatment and expectation of improvement Patients rated the three treatments as equally logical, indicated that they would be equally likely to recommend them to a friend, and had equivalent expectations of improvement (means (s.d.): CT=8.6 (1.5), 6.8 (2.5), 8.3 (1.9); AR = 8.8 (1.3), 7.1 (1.5), 8.4 (1.6); lMlP = 7.8 (1.7), 6.7 (2.4), 7.8 0.9)). Comparisons between each treatment and waiting-list Table 1 shews patients’ panic/anxiety composite and BDl scores before treatment or wait and 3 months later. ANOVAs were used to compare pre-treatment/waiting scores for the composite and all individual measures. None were significant, indicating that the groups did not differ before the start of treatment or wait. To determine whether more improvement in panic/anxiety occurred in the treatments than in the waiting-list group, a one-way ANCOVA was performed on the 3-month panic/anxiety composite scores. This was highly significant (F= 15.9, d.f. = 3,59, P< 0.001). Duncan multiple comparisons indicated that cognitive therapy. applied relaxation and imipramine all differed significantly from waiting-list (P< 0.01, P< 0.05 and P< 0.05 respectively). Analysis of individual measures (means and s.d.s available from authors) showed cognitive therapy was significantly superior to waiting-list on all panic/anxiety measures. Applied relaxation was superior to waiting-list on 12 panic/anxiety measures: panic frequency (patient and assessor); panic-related distress (patient); general tension and anxiety (patient); BAl; HARS; Table 1 Pre-treatment/wait and 3-month scores for each treatment and waiting~|ist Measure Occasion Cognitive Applied lmipramine Waiting-list therapy relaxation Panic/anxiety composite score Pre-treetment 0.98 (0.4) 1.12 (0.7) 1.05 (0.5) 1.22 (0.6) 3 months -0.78 (0.4) -0.01 (0.7) -0.06 (0.8) 0.94 (0.8) Beck Depression Inventory Pre-treatment 18.1 (8.0) 18.6 (10.5) 18.0 (9.5) 21.3 (9.2) (0-63) 3 months 7.3 (6.6) 9.8 (7.4) 8.2 (8.2) 20.9 (11.7) n- 16 for each group; s.d.s are given in parentheses. 764 CLARK ET AL Table 2 Comparisons between cognitive therapy, applied relaxation, and imipramine Measure Occasion Cognitive Applied lmipramina Summary of significant therapy relaxation between-treatment comparisons‘ Panic/anxiety composite score Pro-treatment 0.83 (0.6) 1.06 (0.7) 1.15 (0.7) - 3 months —0.82 (0.5) 0.02 (0.9) 0.01 (1.0) CT <AR, MP 6 months —0.75 (0.5) 0.05 (0.9) -0.33 (0.9) CT. lMlP <AR 15 months -0.88 (0.8) -0.08 (0.9) -0.05 (1.1) CT <AR, IMlP Panic frequency (0—4) (assessor) Pro-treatment 2.5 (1.0) 2.8 (1.0) 2.8 (0.9) - 3 months 0.3 (0.6) 1.3 (1.5) 1.0 (1.2) CT <AR 6 months 0.3 (0.4) 1.4 (1.5) 0.7 (0.9) CT. (MlP <AR 15 months 0.4 (0.8) 1.1 (1.3) 0.9 (1.2) - Panic frequency (0—4) (patient) Pie-treatment 2.6 (1.0) 2.8 (1.0) 3.0 (0.9) - 3 months 0.3 (0.7) 1.4 (1.4) 1.1 (1.3) CT <AR 6 months 0.4 (0.7) 1.6 (1.5) 0.6 (0.9) CT. IMlP<AR 15 months 0.5 (1.0) 1.3 (1.6) 0.9 (1.4) - Panic-related distress/disability Pie-treatment 5.1 (1.6) 5.8 (1.3) 5.6 (1.4) - (0-8) (assessor) 3 months 1.3 (1.2) 3.2 (2.2) 3.1 (2.4) CT (AR, (MIP 6months 1.4 (1.1) 2.9 (2.0) 2.2 (2.1) - 15 months 1.3 (1.8) 3.0 (2.1) 2.8 (2.2) CT <AR, IMIP Panic-related distress/disability Pie-treatment 5.2 (2.0) 6.1 (1.7) 6.0 (2.2) - (0-8) (patient) 3 months 1.2 (1.1) 3.0 (1.8) 3.0 (2.5) CT <AR, (M)? 6 months 1.4 (1.2) 2.9 (2.0) 1.8 (2.0) CT. IMIP <AR 15 months 1.1 (1.6) 2.8 (2.0) 2.6 (2.3) CT <AR. IMIP Beck Anxiety Inventory (0—63) Pre-treatment 25.4 (8.4) 29.2 (9.1) 29.9 (8.2) - (patient) 3 months 8.2 (4.6) 13.8 (9.9) 17. (12.5) CT <IMIP 6 months 10.0 (5.7) 15.7 (11.3) 15.0 (12.1) - 15 momhs 8.0 (7.4) 14.0 (9.7) 16.8 (11.6) - Hamilton Anxiety Rating Scale Pre-treatment 19.7 (3.9) 18.6 (4.6) 21.6 (6.5) - (0-56) (assessor) 3 months 10.5 (6.2) 13.5 (6.5) 15.9 (8.6) - 6 months 9.3 (6.1) 13.3 (8.0) 11.5 (7.0) - 15 months 7.9 (6.6) 12. (6.8) 13.4 (8.6) - General tension and anxiety Pretreatment 4.5 (1.4) 4.1 (1.5) 4.9 (1.7) - (0-8) (assessor) 3 months 2.4 (1.5) 3.3 (1.8) 3.2 (2.2) - 6m0nths 2.2 (1.3) 3.2 (1.4) 2.6 (1.7) - 15 months 1.7 (1.6) 3.3 (1.7) 2.9 (1.9) CT <AR, IMIP General tension and anxiety Pie-treatment 5.2 (1.6) 5.2 (1.4) 5.5 (2.2) — (0-8) (patient) 3 months 2.3 (1.6) 2.9 (1.7) 3.0 (2.0) - 6months 2.7 (1.2) 3.3 (2.0) 2.8 (2.3) - 15 months 2.3 (2.0) 3.5 (2.0) 3.2 (2.5) - Agoraphobic avoidance (0-15) Pro-treatment 5.9 (3.6) 6.8 (4.2) 6.5 (4.5) - 3 months 2.9 (2.5) 5.3 (3.8) 4.9 (2.8) CT <AR, lMlP 6months 3.2 (2.9) 4.7 (4.4) 4.3 (3.3) - 15 months 2.6 (2.9) 5.1 (4.4) 4.7 (3.9) - Social avoidance (0—15) Pre-treatment 5.2 (3.6) 6.8 (3.8) 6.2 (3.5) — 3months 3.4 (3.2) 5.5 (4.2) 5.5 (3.8) - 6months 3.2 (2.4) 5.0 (4.1) 4.6 (3.3) - 15 months 3.1 (3.3) 5.1 (4.2) 5.0 (4.5) - Bloody/injury avoidance (0-15) Pie-treatment 5.1 (3.7) 3.9 (3.0) 5.1 (4.2) - 3 months 2.4 (2.3) 3.4 (3.8) 3.9 (3.5) CT <AR, (MIP 6months 3.0 (3.0) 3.3 (3.5) 4.8 (3.8) - 15 months 2.3 (3.0) 3.7 (2.9) 4.6 (3.8) CT <AR, IMIP Agoraphobic (ear (5-20) Pre-treatment 13.4 (5.7) 13.1 (5.5) 13.6 (4.7) - 3 months 8.3 (4.0) 10.8 (5.2) 10. (3.7) CT <AR, lMlP 6 months 9.0 (4.2) 10.7 (5.9) 8.6 (3.7) - 15 months 8.0 (3.7) 10.7 (5.3) 11.3 (5.0) CT (AR, (MIP Social (ear (5—20) Pro-treatment 11.1 (5.1) 12.9 (4.6) 12.5 (5.7) - 3months 8.2 (3.6) 10.9 (4.6) 11.0 (5.6) - 6 months 8.1 (3.2) 10.9 (5.1) 10.0 (5.1) s 15 months 8.0 (3.8) 10.2 (4.4) 10.2 (5.1) - (continued) COMPARISON OF TREATMENTS OF PANIC DISORDER 765 Table 2 (continued) Comparisons between cognitive therapy. applied relaxation, and imipramine Measure Occasion Cognitive therapy Bloodfinjury fear (5-20) Pre-treatment 12.6 (4.5) 3 months 10.0 (3.5) 6 months 10.1 (3.1) 15 months 9.5 (3.7) Body Sensations Questionnaire Pretreatment 36.6 (9.5) (17-68) 3 months 22.5 (5.2) 6 months 24.8 (7.2) 15 months 22.7 (6.0) Agoraphobic Cognitions Pretreatment 27.9 (6.5) Questionnaire (15-75) 3 months 19.4 (3.3) 6 months 20.5 (5.3) 15 months 19.6 (5.3) Bodin Sensations Interpretations Pro-treatment 13.4 (4.4) Questionnaire-panic items 3 months 7.3 (0.6) (7-21) 6 months - 7.7 (1.7) 15 months 8.2 (2.1) Beck Depression Inventory Pro-treatment 17.0 (7.8) (0—63) 3 months 6.3 (6.4) 6 months 6.0 (6.1) 15 months 5.3 (5.2) Applied lmipramine Summary ol significant relaxation between-treatment comparisons‘ 11.9 (3.7) 12.5 (4.9) - 10.4 (5.0) 10.9 (5.4) - 10.4 (4.1) 10.8 (5.4) - 10.4 (4.4) 11.3 (5.3) - 45.4 (9.9) 42.2(13.0) - 35.9 (13.2) 33.2 (9.3) CT <AR,IM(P 36.4(11.5) 31.1 (8.5) CT <AR 34.0 (10.9) 37.3 (11.2) C7 <AR, (MI? 32.9 (8.2) 32.5 (7.5) - 27.0 (7.4) 26.2 (8.3) CT <AR,(MIP 28.2 (8.7) 24.3 (6.8) CT,lMIP <AR 25.2 (6.1) 25.9 (8.7) -' 13.6 (4.1) 13.7 (4.5) - 9.4 (3.3) 10.6 (3.3) CT <AR,lMIP 9.5 (2.9) 10.1 (3.8) CT <AR,lMlP 9.1 (2.5) 10.2 (4.3) -’ 19.0 (9.6) 20.6 (12.0) - 10.2 (7.1) 9.8 (10.4) - 11.4 (8.3) 7.9 (9.4) CT,lMIP<AR 10.5 (6.7) 8.3(10.8) — n- 20 (or each group except for applied relaxation at 15 months when n: 19 as one patient in this treatment was unavailable (or the 15-month assessment. a.d.s are given in parentheses. (n the last column. '<' means 'superior to'. 1. P< 0.05 or less, Duncan’s multiple comparison procedure. 2. Although the overall ANCOVA just failed to reach significance for these measures at 15 months. pairwise ANCOVA comparisons indicated that at 15 months CT was superior to AB on BSlO-penic (P< 0.05) and was superior to (M)? on the Agoraphobic Cognition; Questionnaire (P<0.05). Agoraphobic, Social, and Blood-injury Fears; Body Sensations Questionnaire; BSlQ-panic; and ACQ. lmipramine was superior to waiting-list on 8 panic/ anxiety measures: panic frequency (patient and assessor); panic-related distress/disability (patient); general tension and anxiety; BA); Body Sensations Questionnaire; BSlQ— panic; and ACQ. A one-way ANCOVA on the three-month BDI scores was also significant (F: 8.7, d.f. = 3.59. P< 0.001). Pairwise comparisons indicated all three treat- ments were significantly superior to waiting-list. Comparison between the three treatments Table 2 shows mean composite and individual measure scores before treatment and 3. 6 and 15 months later. Twelve patients still met trial inclusion criteria when allocated to treatment after the wait period. In order to increase statistical power, it was decided (a priori) to include these patients in all analyses comparing the effectiveness of the three treatments. ANOVAs indicated that there were no significant pre-treatment differences between these 12 patients and patients immediately allocated to treatment and their inclusion did not influence effect sizes. To compare the effectiveness of the treatments, separate analyses were performed on the data at 3, 6 and )5 months. A13 months ANCOVA revealed a significant treatment effect on the panic/anxiety composite (F= 5.6, d.f. = 2.56, P< 0.01) but not on the ED]. Multiple comparisons on the panic/anxiety composite indicated that cognitive therapy was superior to both applied relaxation (P< 0.01) and imipramine (P< 0.0l). Applied relaxation and imipramine were not significantly different from each other. Analyses of individual panic/anxiety measures are summarised in Table 2. Cognitive therapy was superior to applied relaxation on 10 panic/anxiety measures and was superior to imipramine on 9 panic/anxiety measures. Table 3 shows the percentage of patients who were panic free and the percentage who were high end-state functioning at each assessment. To be classified as panic free, patients had to have experienced no panic attacks during the preceding two weeks. Following Craske er al (l991), high end-state function was defined as panic free and an assessor panic-related distress/disability rating <2 (‘slight’). Fisher’s exact tests indicated that at three months both the proportion of patients who were panic free and the proportion who were high end-state functioning was significantly (P<0.05) greater in cognitive therapy than in either applied relaxation or imipramine. Ar 6 months there was a significant ANCOVA treatment effect on the panic/anxiety composite (F: 4.6, d.f. = 2,56, P<0.05) and on the BDl (F: 3.2. d.f.=2,56. P<0.05). Figure I shows that on the panic/anxiety composite imipramine patients continued to improve between 3 and 6 months while cognitive therapy and applied relaxation afforded little change. This is reflected in the multiple comparisons which indicated that at 6 months there was 766 Table 3 Percentages of patients panic free and high end-state functioning at different time points and percentage who relapsed Assessment Cognitive Applied lmipramine point therapy relaxation Percentage of patients panic free Pre-traatmant 0 0 0 3 months‘ 90 50 55 6 months 75 40 70 15 months 85 47 60 Percentage of patients reaching high end-state function Pie-treatment 0 0 0 3 months' 80 25 40 6 months 65 35 55 15 months 70 32 45 Percentage of patients who relapsed and required ' further treatment '- 6-15 months 5 26 40 n-20pargroupaxcaptforappliadrelaxationat 15monthawhenn- 19. 1. For comparison purposes at three months 796 of waiting-list patients were panic free and 7% of waiting-list patients were high end-state functioning. no longer a significant difference between cognitive therapy and irnipramine and both were superior to applied relaxation (P< 0.01 for cognitive therapy; P< 0.05 for imiprarnine). Table 2 summarises the individual panic/anxiety measure analyses. Cognitive therapy was superior to applied re- laxation on 6 panic/anxiety measures and irnipramine was superior to applied relaxation on 4 panic/anxiety measures. Fisher’s exact tests indicated that cognitive therapy and irnipramine did not differ in the proportion of patients who were panic free or high end-state functioning at 6 months but patients treated with cognitive therapy tended (P= 0.054) to be more likely to be panic free than those treated with applied relaxation. Multiple comparisons on the 3D] in- dicated that cognitive therapy and imipramine were both superior to applied relaxation (P<0.05). 1.5 r‘ o $ Panidanxiety composite score .0 .0 O U! 15months Pre- 3 rnomhas months mumnt Fig. l Panic/anxiety composite scores at pre—treatment, 3, 6. and 15 months (cognitive therapy. -I-; applied relaxation, ---El---; irnipramine, "4»). CLARK ET AL A! 15 months there was again a significant ANCOVA treatment effect on the panic/anxiety composite (F: 3.5, d.f. = 2,55, P< 0.05) but not on the 8D]. Imipramine was gradually withdrawn after the 6-month assessment and Figure l shows that on the panic/anxiety composite irnipramine patients tended to deteriorate between 6 and 15 months. This is reflected in the multiple comparisons which indicated that at is months cognitive therapy was again superior to irnipramine (P<0.05) and applied re- laxation (P< 0.05), which did not differ from each other. Analysis of individual panic/anxiety measures (Table 2) indicated that cognitive therapy was superior to applied relaxation and irnipramine on 6 measures. Fisher’s exact tests on the proportion of patients who were panic free and the proportion who achieved high end-state function showed cognitive therapy was superior to applied relaxation on both measures (P< 0.05) but did not differ from irnipramine. However. the comparison with irnipramine is complicated by additional treatment. As Table 3 shows, patients treated with irnipramine were significantly (P- 0.02) more likely to relapse and receive additional treatment between 6 and 15 months than patients treated with cognitive therapy. Additional analyses. Thirty-five per cent (2] out of 60) of patients were taking psychotropic medication at the start of the trial. As only 2 of these patients, who were both allocated to applied relaxation. reported increasing extra- trial medication during the study, it seemed unlikely that extra-trial medication had influenced the results reported above. However, as an additional check, the panic/anxiety composite analyses were repeated using only those patients who were not taking any extra-trial medication. The results of these additional analyses were identical to those based on the full sample. At 3 months cognitive therapy was superior to both applied relaxation (P<0.01) and irnipramine (P<0.01). At 6 months coytitive therapy and irnipramine did not differ from each other but both were superior to applied relaxation (P<0.0l for cognitive therapy; P<0.05 for irnipramine). At 15 months, cognitive therapy was again superior to both applied relaxation (P<0.01) and irnipramine (P< 0.05). Prediction of long-term outcome The cognitive theory of panic predicts that long-term outcome will depend on the extent to which interpretations of bodily sensations have changed during treatment (Clark, 1986). In particular, patients who continue to misinterpret bodily sensations at the end of treatment should have a worse outcome during the follow-up than patients who have ceased to do so. To test this prediction, two analyses were performed. First, for the total sample, we examined the relationship between the cognitive measures (Bodily Sensations Interpretation Questionnaire-panic items, and Agoraphobic Cognitions Questionnaire) at the end of the active treatment (6-month assessment) and the panic/ anxiety composite at the longest follow-up point (15-month assessment), while controlling for panic/anxiety composite scores at the end of treatment. To avoid criterion contamination the panic/anxiety composite used in this analysis excluded the cognitive measures. Second, for those patients who were panic free at 6 months, we examined the COMPARISON OF TREATMENTS OF PANIC DISORDER relationship between cognitive measures at the d of treatment and whether patients subsequently relapsed (indexed by additional treatment for panic during follow- up and/or experiencing panic attacks at the 15-month assessment). Both analyses supported the prediction. For the total sample, there was a significant correlation between BSlQ—panic at 6 months and the panic/anxiety composite at 15 months (r(,,,=0.65, P<0.001) and this remained significant when composite scores at 6 months were partialled out (r(,.,,:=0.32, P< 0.05). Within patients who were panic free at 6 months, there was a significant correlation between BSlQ-panic at 6 months and sub- sequent relapse (rm, 2 0.49, P< 0.01). Correlations between the Agoraphobic Cognitions Questionnaire and subsequent symptomatology were in the same direction as those obtained with the BSlQ—panic scale but failed to reach significance. This difference in results is in line with the nature of the two measures. The BSlQ-panic scale is a schema measure which was designed to assess the tendency to misinterpret bodily sensations, even if a patient is currently asymptomatic. The ACQ was designed as a measure of the automatic thoughts that occur during anxiety episodes. Discussion A major aim of the present study was to compare the effectiveness of cognitive therapy with that of applied relaxation and imipramine. Previous studies had suggested that all three treatments are effective, and this was confirmed in the present study as all three treatments were associated with significantly greater impr0vement than the waiting-list condition. Comparisons between treatments indicated that the relative efficacy of the treatments varied with assess- ment occasion and when there were differences between treatments these were not evident on all measures. Nevertheless, cognitive therapy was the most consistently superior treatment, followed by imipramine. On no occasion was applied relaxation more effective than either cognitive therapy or imipramine. The second aim of the study was to investigate the role of non-specific factors in cognitive therapy. For this question the most informative comparison is that between cognitive therapy and applied relaxation. Both of these treatments were equated in terms of treatment credibility, session frequency, session duration, therapists, and the use of situational exposure homework assignments but cognitive therapy was consistently superior. It is therefore clear that these non-specific therapy factors cannot entirely account for the effectiveness of cognitive therapy. Three main procedures differentiated cognitive therapy from applied relaxation (and imipramine): (a) detailed discussion of the evidence for and against patients‘ negative beliefs, (b) behavioural 767 experiments explicitly designed to test these beliefs, and (c) interoceptive exposure. We do not know whether all three procedures contributed to the enhanced effectiveness of cognitive therapy and to decide this will require studies evaluating each component separately. Salkovskis et a1 (1991) pro- vided preliminary evidence that the cognitive procedures alone are effective: in a series of single case experiments good results were obtained with a modified version of cognitive therapy from which interoceptive exposure, situational exposure and breathing retraining had been excluded. Margraf & Schneider (described in Margraf et al, 1993) also found that cognitive procedures alone (without interoceptive or situational exposure) were effective. In addition, Margraf & Schneider found that situational and interoceptive exposure without explicit cognitive restructuring was equally effective. However, change in panic-related cognitions was a significant predictor of improvement in both the pure cognitive and the pure exposure treatments, suggesting that the two different psychological procedures may have their effects through a common psychological mechanism, change in misinterpretations of bodily sensations. The final aim of the study was to investigate long term outcome. Cognitive theory predicts that sustained improvement after the end of treatment is dependent on cognitive change having occurred during the course of treatment. Analysis of the long- terrn (15-month) outcome data provides preliminary support for this hypothesis. Misinterpretations of bodily sensations at the end of treatment were significant predictors of subsequent symptoms in the total sample and of relapse in patients who were panic free at the end of treatment. If confirmed in other follow-up studies, this finding has important implications for the conduct of both psychological and pharmacological treatments for panic. In particular, it suggests that clinicians should aim not only for a panic-free state but also for marked cognitive change. The present data suggest that without such change patients are likely to have a relatively poor long-term outcome. The present study has several limitations which should be borne in mind when interpreting the results. First, we modified Ost’s (1987) applied relaxation treatment by reducing overlap with cognitive therapy’s rationale and introducing exposure earlier. These changes may have reduced the effectiveness of the treatment and a comparison between cognitive therapy and unmodified applied relaxation is required to clarify this issue. Second, we adopted a low starting dose, gradual build-up regime with imipramine in order to reduce dropouts. The regime appears to have been effective, as we 768 had a lower dropout rate (5% v. 25—30%) and a higher end dose (233 mg/day v. 155 mg/day) than the recent multicentre study (Andersch et al, I991; Cross-National Panic Study, I992) comparing alpra- zolam and imipramine. However, the slowness of the build-up (mean = 7.5 weeks) is likely to have contri- buted to imipramine’s relatively poor outcome at the 13-week assessment. Further studies are needed to determine whether others can achieve a similar low drop-out rate with a faster dose build-up. Third, imipramine was withdrawn after 6 months. While this was common practice at the time the study was planned, more recent studies (Frank er a], 1990) indicate that in depression further relapse prevention can be achieved by patients on meditation for more than a year after recovery. It is, therefore, possible that less relapse would have been observed with imipramine if we had a longer maintenance phase. Fourth, as with most trials of treatments for anxiety, some of our patients were taking psycho- tropic medication before entry into the trial. Prompted by a desire not to exclude patients who were unable to withdraw, we chose to deal with this problem by requiring that medicated patients had been on a stable dose for several months without improvement and to prohibit increases in dose or the introduction of new medications. Very few patients incrwsed their extra-trial medication and comparisons between treatments remained significant when re- stricted to patients taking no extra medication. It therefore seems unlikely that extra-trial medi- cation influenced our results. However, complete pre-trial drug withdrawal with exclusion of patients who are unable to withdraw is a more common re- search design and confidence in the present findings would be enhanced by a replication using this design. Margraf & Schneider (see Margraf et 01, I993) used the withdrawal design and obtained improvements with cognitive therapy and waiting-list comparable to those in the present study but to date no studies have used the withdrawal design and compared cognitive therapy with irnipramine. Finally, because our interest was in the relatively less phobic panic disorder patients, those with severe agoraphobic avoidance were excluded. Further comparative treatment trials with this population would be of interest, especially as all three of the treatments used here and situational exposure appear to be effective with this group of patients (Michelson & Marchione, I991). Acknowledgements This research was funded by grants from the Medical Research Coundl of the United Kingdom, the North Atlantic Treaty CLARK ET AL Organization and the Wellcome Trust. The authors would like to thank Sarah Durbin, Carolyn Fordham-Walker, Johanna Flack. Ann Jeavons, Katherine Koehler, John Ludgate, Anthony Morrison, David Nutt. Lars—Goran 0st and Adrian Wells for their assistance. In addition, Aaron Beck, David Barlow. Diane Chambless. and Steven Hollon provided helpful criticism of an earlier manuscript. References Asa-11cm PsvcmAmc Asocwnow (I987) Diagan and Statistical Manual of Mental Disorders (3rd edn, revised) (DSM-lll-R). Washington, DC: APA. Anosascn, S.. Rosanna, N. K.. KULLINGSLO. H., et al (I991) Efficacy and safety of alprazolarn. imipramine and placebo in treating panic disorder. A Scandinavian multicenter study. Acta Psychiatrica Scandinavian (suppl. 365), 18-27. BAaLow, D. H. a Canny. J. A. (I988) Psychological Treatment of Panic. New York: Guilford. , COHEN, A. S.. WADDELI... M. T., et al (I984) Panic and generalised anxiety disorders: nature and treatment. Behavior Therapy, 15, 431—449. 8: Cum, M. G. (I989) Mastery of your Anxiety and Panic. Albany: Oraywind Publications. , , Cam, I. A., et aI (I989) Behavioural treatment of panic disorder. Behavior Therapy. 20. 26l-282. BECK, A. T., Essen. G. a GREENIEIG, R. 1.. (I985) Anxiety Disorders and Phobias. New York: Basic Books. , EPSTEIN. N., BROWN. 0.. et a! (I988) An inventory for measuring clinical anxiety: psychometric properties. Journal of Consulting and Clinical Psychology, 56, 893-897. , SOKOL, I... CLAax, D. A.. et aI (I992) Focused cognitive therapy of panic disorder: a crossover design and one year follow-up. American Journal of Psychiatry. I47, 778-783. , WAan. C. H.. MW. M., etal(I9bl) An inventory for measuring depression. Archie's ofGeneral Psychiatry. 4. 561-571. Boakovec, T. D. & Nut, 5. D. (I972) Credibility of analogue therapy rationales. Journal of Behaviour Therapy and Experi- mental Psychiatry. 3. 257—260. CATALAN, 1., GATH, D., Eamonns, (3., et al (1984) The effects of non-prescribing in general practice: I. Controlled evaluation of psychiatric and social outcome. British Journal of Psychiatry, 144. 593—603. CHAMBLBS. D. I... CAPUTO, G. C., Britain. P., et al (I984) Assessment of fear of fear in agoraphobics: the Body Sensations Questionnaire and the Agoraphobic Cognitions Questionnaire. Journal of Consulting and Clinical Psychology, 52, 1090-1097. amt. D. M. (I986) A cognitive approach to panic. Behaviour Research and Therapy. 14, 461-470. (1989) Anxiety states: panic and generalized anxiety. 1n Cognitive Behaviour Therapy for Psychiatric Problem: A Practical Guide (eds K. Hawton. P. Salkovskis, 1. Kirk, et al), pp. 52-96. Oxford: Oxford University Press. a EHLeas, A. (I993) An overview of the cognitive theory and treatment Of panic disorder. Applied and Preventive Psychology, 2, I3I-139. , SAonvsxts. P. M. a CHALKLEY. A. J. (I985) Respiratory control as a treatment for panic attacks. Journal of Behaviour Therapy and Experimental Psychiatry, I6, 23—30. CIASKE, M. 0.. BROWN, T. A. a BAntow, D. H. (1991) Behavioural treatment of panic disorder: a two year follow-up study. Behaviour Therapy, 22. 289-304. Caoss-NA'rtONAL COLLABORATIVE PANIC STUDY, SECOND PHASE INVBTIGATORS (I992) Drug treatment of panic disorder: com- parative efficacy of alprazolam. imipramine and placebo. British Journal of Psychiatry, 160. 191-202. EHLEas, A., MAaoaAr, J. & ROTH, W. T. (I986) Seiective information processing, interooeption and panic attacks. In Panic and Phobias ll (eds 1. Hand 3: H. U. Wittchen). New York: Springer-Verlag. COMPARISON OF TREATMENTS 0F PANIC DISORDER FRANK. E., Kuvrsa, D. 1., Penn, J. M., et al (1990) Three-year outcomes for maintenance therapies in recurrent depression. Archim of General Psychiatry, 47, 1093-1099. FYER, A. J. a SANDBERG. D. P. (1988) Phannacolop'cal treatment of panic disorder. In Review of Psychiator (eds A. J. Frances a R. E. Hales). New York: American Psychiatric Press. HAMILTON. M. (1969) Diagnosis and rating of anxiety. British Journal of Psychiatry (special publication no. 3), 76-79. HOLLON, .1. D., De RUIEIS, R. J., Evans, M. D., et al (1992) Cognitive therapy and pharmacotherapy for depression: sineg and in combination. Archives of General Psychiatry, 49. 774-781. Jansson. L. 4t Os'r. L. G. (1982) Behavioural treatments for agoraphobia: an evaluative review. Clinical Psychology Review. 2, 311-336. KLOSKO, .1. S., BaaLow, D. H.. Tasman, R., et al (1990) A comparison of alprazolam, and behaviour therapy in the treatment of panic disorder. Journal of Consulting and Clinical Psychology, 58, 77-84. Kory, M., MILHALY. K. a Vaoasz. P. (1986) Agoraphobics es panikneurotikus betegek legzesi kontroll keyelese. ldeggyogyaszati Szemle, 39. 185—196. MAIGRAF, .1., BAILOW, D. 1-1., CLARK. D. M., et al (1993) Psychologiml treatment of panic: work in progress on outcome. active ingredients, and follow-up. Behaviour Research and Therapy. 3]. 1-8. Mnxs, l. a MATHEWS, A. M. (1979) Brief standard self-rating for phobic patients. Behaviour Raearch and Therapy. 17, 263—267. . Swmson, R. P., BASOGLU, M.. etal (1993) Alprazolam and exposure alone and combined in panic disorder with agora- phobia: a controlled study in London and Toronto. British Journal of Psychiatry, 162. 776-787. MAYHEWS, A. M., GELDER, M. G. & JOHNSTON, D. W. (1981) Agoraphobia: Nature and Treatment. New York: Guilford Press. MCNALLY. R. (1990) Psychological approaches to panic disorder: a review. Psychological Bulletin. 108, 403-419. MrcnstsoN. L. K. & MAICHION’E, K. (1991) Behavioral, cognitive, and pharmacological treatments for panic disorder with agora- phobia: critique and synthesis. Journal of Consulting and Clinical Psychology. 59, 100—114. 769 —, —. GREENWALD, M., et al (1990) Panic disorder: cognitive-behavioural mnent. Behaviour Research and Therapy, 28, 141-153. Nu'n, D. .l. & GLUE, P. (1989) Clinical pharmacology of anxio- lyu'cs and antidepressants; a psychophannacological perspective. Pharmacology and Therapeutics. 44, 309-334. 051'. L. G. (1987) Applied relaxation: description of a coping technique and review of controlled studies. Behaviour Ruearch and Therapy. 25, 397—410. (1988) Applied relaxation vs. progressive relaxation in the treatment of panic disorder. Behaviour Rsearch and Therapy, 26. 13-22. ROSENTHAL, R. & Rosnow, R. L. (1991) EssentiaLs of Behavioral Research: Methods and Data Analysis (2nd edn). New York: McGraw-Hill. SALKOVSKIS. P. M. (1988) Phenomenology, assessment and the cognitive model of panic. In Panic: Psychological Perspectiva (eds S. Rachman a J. Maser). New Jersey: Erlbaum. & CLARK, D. M. (1991) Cognitive therapy for panic disorder. Journal of Cognitive Psychotherapy, 5, 215-226. , t HACKMAN'N, A. (1991) Treatment of panic attacks using cognitive restructuring without exposure or breathing retraining. Behaviour Raearch and Therapy. 29, 161—166. . JONES. D. R. 0. a Cunt, D. M. (1986) Respiratory control in the treatment of panic attacks: replication and extension with concurrent measurement of behaviour and pCO,. British Journal of Psychiatry, 148, 526-532. SHEAR. M. K., BALL, 6.. FITZPATRICK, M.. et al (1991) Cognitive- behavioural therapy for panic: an open study. Journal of Nervous and Mental Disease, 8, 463-474. SOKOL. L., BECK. A. T.. GREENIERG, R. L., et al (1989) Cognitive therapy of panic disorder: a non-phannacological alternative. Journal of Nervous and Mental Disease. 177, 711-716. Srr'rzsa, R. L. a WILLIAMS. J. B. (1986) Structured Clinical Interview for DSM—lll-R, Patient Version (SCID—P, 1/1/86). New York State Psychiatric institute, 722 West 168th Street, New York. NY 10032: Biometrics Research Depanmem. WELKOWITZ, L. A.. Parr, L. A., erras, M., et al (1991) Cognitive-behaviour therapy for panic disorder delivered by psychopharmacolop'cally oriented clinicians. Journal of Nervous and Mental Disease, 179. 473-477. ‘David M. Clark, MA, MPhil, DPhil, Wellcome Principal Research Fellow, Department of Psychiatry, University of Oxford; Paul M. Salkovskis, BSc, MPhil, PhD, Wellcome Senior Research Fellow in Biomedical Science, Department of Psychiatry, University of Oxford; Ann Hackmann, BA, MSc, Research Clinical Psychologist, Department of Psychiatry, University of Oxford; Hugh Middleton, MA, MD, Research Psychiatrist, Department of Psychiatry, University of Oxford and now Clinical Lecturer, Department of Psychiatry, University of Cambridge; Pavlos Anastasiades, BA, Research Psychologist, Department of Psychiatry, University of oxford; Michael Gelder, MA, MD, Handley Profesor of Psychiatry, University of Oxford 'Correspondence: Department of Psychiatry, University of Oxford, Warneford Hospital, oxford 0X3 71X (First received June 1993, final revision October 1993, accepted November 1993) ...
View Full Document

This note was uploaded on 08/12/2008 for the course PSY 364 taught by Professor Telch during the Fall '06 term at University of Texas at Austin.

Page1 / 11

TreatmentofPanic - British Journal of Psychiatry (1994),...

This preview shows document pages 1 - 11. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online