Cognitive and behavioural treatments of fear of blushing, sweating or trembling

Behau. Rev. Thu. Vol. 31, No. 2,pp. 155-170, 1993 OOOS-7967/93 $6.00 +O.OO Printed in Great Britain. All rights reserved Copyright 0 1993 Pergamon Press Ltd

COGNITIVE AND BEHAVIOURAL TREATMENTS OF FEAR OF BLUSHING, SWEATING OR TREMBLING*

AGNES Scr-roLrNGt and PAUL M. G. EMMELKAMP

Department of Psychology, University of Groningen, Oostersingel 59, 9713 EZ Groningen, The Netherlands

(Received 9 April 1992)

Summary-Social phobic patients (n = 30) with fear of blushing, sweating or trembling as the predomi- nant complaint were randomly assigned to three treatment conditions: (1) exposure in oivo followed by cognitive therapy, (2) cognitive therapy followed by exposure in vivo, or (3) a cognitive-behavioural treatment in which both strategies were integrated from the start. Each treatment condition consisted of 16 sessions, given in two treatment blocks of 4 weeks each, separated by a no-treatment phase of 4 weeks. Self-report assessments were held before and after the treatment blocks and at 3-months follow-up. No significant differences were found between effects of the first treatment block vs those of the 4-weeks waiting-list period. After the second block treatment was significantly more effective than waiting-list. The analyses showed significant time effects after both treatment blocks and at follow-up, indicating improvement for the group as a whole. After two treatment blocks and at follow-up no significant differences among the different treatment packages were found on target problems, avoidance of social situations, cognitions and somatic complaints. After discussion of the results recommendations for further research will be given.

INTRODUCTION

According to DSM-III-R (American Psychiatric Association, p. 241) the central feature of social phobia is “ . . . a persistent fear of one or more situations in which the person is exposed to possible scrutiny by others and fears that he or she may do something or act in a way that will be humiliating or embarrassing. . . “. In contrast with the criteria according to DSM-III (American Psychiatric Association, 1980), in DSM-III-R the diagnosis of social phobia also applies to individuals who fear more than one social situation, in which case the addition ‘generalized type’ is given. Available data suggest that generalized social phobia is prevalent: 67% of the patients described by Mattick, Peters & Clarke (1989) and 90% of the patients of Turner, Beidel, Dancu & Keys (1986) stated that they experienced anxiety in at least two different situations.

These data implicate that a minority of patients suffer from more specific or discrete social phobias. Until now sparse attention has been paid to the question of which fears must be considered as such. Heimberg, Hope, Dodge & Becker (1990) described a type of discrete social phobia, namely public speaking phobia. In contrast, Beck & Emery (1985) stated “ . . . Certainly the fear of speaking in public should not be included (in social phobia) since a very high proportion of the general population has this fear . . . ” (pp. 150-151).

Marks & Gelder (1966) described examples of social phobia that can be conceived to be specific, like fears of eating, drinking, shaking, blushing, writing or vomiting in the presence of other people. A common feature of those complaints seems to be the fear of showing bodily symptoms that are not under voluntary control of the S and may be noticed by others. A central characteristic is a vicious circle, in which maladaptive cognitive processes (like anticipatory fear of the symptoms) produce distress and, as a self-fulfilling prophecy, subsequently lead to exacerbation of the symptoms. A similar process has been hypothesized for panic disorder (Clark, 1986).

In the last decade, several studies have been conducted on treatment of social phobia, with emphasis on effectiveness of exposure in z&o and cognitive strategies (Butler, Cullington, Munby, Amies & Gelder, 1984; Emmelkamp, Mersch, Vissia & Van der Helm, 1985; Mattick 8z Peters, 1988; Mattick, Peters & Clarke, 1989; reviewed by Heimburg, 1989; and Scholing & Emmelkamp, 1991).

*Parts of this paper were presented at the International Congress of the European Association for Behaviour Therapy, Paris, France, September 1990.

tAuthor for correspondence.

155

156 AGNES~CHOLING and PAUL M.G. EMMELKAMP

At the moment it must be concluded that both exposure in vivo and cognitive therapy are useful treatments for social phobia. Although in some studies a combination of exposure and cognitive therapy was found to be slightly superior to the single treatments, differences were only small and not consistent across measures.

Considering the somewhat specific aspects of fear of bodily symptoms that were described above, it is unclear whether the results of these studies on social phobics can be generalized to patients with fear of bodily symptoms. Treatment studies on fear of bodily symptoms have been predominantly pharmacology-oriented and have concentrated on effects of beta-blockers (reviewed by Welkowitz & Liebowitz, 1990). Although positive results were found in volunteers with stagefright, results for social phobics were less encouraging.

Van Son (1978) found social skills therapy (n = 8) to be significantly more effective than systematic desensitization (n = 9) for patients with fear of blushing (erytrophobics), although at post-test patients reported that they were still afraid of the symptoms. Apart from this study only case studies have been published. Boeringa (1983) Lamontagne (1978) and Mersch, Hildebrand, van Hout, Lavy & Wessel (1993b) all found favourable results for paradoxical intention therapy in the treatment of chronic fear of blushing or trembling, but numbers of patients were small and interpretation of the findings is difficult because they also used other techniques, like ‘more traditional psychotherapy’ (Boeringa, 1983) and rational emotive therapy (Mersch et al., 1992). Only one study (Biran, August0 &Wilson, 1981) compared (therapist guided) exposure in vivo with cognitive therapy (a combination of Ellis’ rational emotive therapy and Goldfried’s cognitive restructuring) in the treatment of 3 patients with writing anxiety (often the consequence of fear of trembling). In contrast with the findings that exposure in vivo and cognitive therapy did not differ much in the treatment of (generalized) social phobia, Biran et al. found exposure in vivo to be clearly more effective than cognitive therapy, with results stable at l- and 9-months follow-up.

The present study aimed at evaluating and comparing the effects of (1) self-controlled exposure in vivo, (2) cognitive therapy and (3) a cognitive-behavioural treatment for patients with fear of blushing, sweating or trembling in social situations.

METHOD

Subjects

Patients between the ages of 18 and 65 were included in the study when they met the following criteria:

-a primary diagnosis of social phobia (DSM-III-R) -fear of blushing, trembling or sweating in social situations as the most prominent fear -absence of severe other psychiatric problems, for example psychotic episodes in the past,

suicidal intentions or substance dependence -no (cognitive-) behavioural treatment in the preceding 3 years.

Patients were referred by general practitioners and mental health institutes or responded to advertisements in local newspapers offering treatment. During a period of 2.5 yr, 188 patients were invited for an intake session; 151 had a primary diagnosis of social phobia. Forty-one patients indicated that fear of showing bodily symptoms in social situations was the main problem. Of this group 2 patients were treated outside this study because they were 6 months pregnant at the time of the intake. Four patients did not accept treatment because of practical reasons. The remaining 35 patients were included in the present study. Patient characteristics are summarized in Table 1.

Experimental design

In this study three treatment packages were compared, each consisting of two treatment blocks, separated by 4 weeks without treatment sessions. The treatment packages were: (1) exposure in vivo followed by cognitive therapy, (2) cognitive therapy followed by exposure in vivo, and (3) a cognitive-behavioural treatment (integration of exposure in vivo and cognitive therapy). After the intake session(s), half of the patients started treatment immediately and were randomly allocated to one of the treatment packages. The other half started with a 4-weeks waiting-list control period,

Cognitive-behavioural treatment 157

Table I. Patient characteristics (completers and dropouts)

Completers Dropouts (n = 30) (n = 5)

Sex Men 14 I Women 16

Mean age* (SD) 30.5 (7. I) 3i.4 (7.8) Marital status

Single 10 0 Living together 7 2 Married 13 3

Mean duration of complaints* (SD) 10.4 (7.6) 5.9 (3.3) Referral

Advertisement IO 3 Referred 20 2

‘In years.

after which they were also randomly assigned to one of the treatment packages. Of the first 30 patients who started treatment 5 patients (17%) dropped out. The remaining 5 patients were randomly assigned to replace the dropouts, so that in each condition 10 patients finished treatment. Ss completed self-report questionnaires before and after both treatment blocks, and at 3-months follow-up. Consequently, a total of 5 (no waiting-list control) or 6 (waiting-list control) assessments were held. Figure 1 shows the experimental design.

Procedure

All patients were screened by the first author with an extended Dutch version of the Anxiety Disorders Interview Schedule-revised (ADIS-R, DiNardo, Barlow, Cerny, Vermilyea, Himadi & Waddell, 1985). In addition, all patients completed the Fear Questionnaire during the intake session. Patients with a primary diagnosis of social phobia (n = 151) were asked whether they suffered from bodily symptoms in social situations. Although the majority of the patients answered confirmative, only patients who stated that these bodily symptoms in fact constituted their only

As0 As-l As2 As3 As4 FU

weekno 1 5 9 13 17 29

ew

co0

integr

co0

exp

integ

intake

ew

Cog

integ

Cog

ew

integr

weekno 1 5 9 13 25

As1 As2 As3 As4 Fu

Fig. 1. Experimental design. As0 to As4, WL-assessment, pretest and post-test first block, pretest and post-test second block; FU, 3-months follow-up; cog, cognitive therapy; exp, exposure in ho; integr., integrated treatment; NWL, no waiting-list preceding treatment; WL, waiting-list preceding treatment.

158 AGNFS SCHOLING and PAUL M. G. EMMELKAMP

problem, and that they were convinced that they would not seek treatment if they could be sure never to blush, tremble or sweat again, were offered treatment in the present study.

After the intake the patient was assigned to one of the therapists. The first contact between therapist and patient was planned l-2 weeks after the intake. This session was primarily aimed at building a sound therapeutic relationship and at self-report assessment. Besides collecting more detailed information about the complaints, attention was given to the social environment and family history of the patient, topics that were not discussed during the treatment blocks. Following this session the first assessment took place. From this session on, all patients kept diaries with respect to their complaints until 4 weeks after the second treatment block, and again for 2 weeks preceding the follow-up. After the first session with the therapist patients were randomly assigned to 4-weeks waiting-list control or (direct) treatment. Patients in the treatment condition were randomly assigned to one of the treatment packages and started treatment a few days after this session. Patients in the waiting-list condition were retested after 4 weeks, and were also assigned to one of the treatment conditions.

Treatment

General aspects. Each treatment block consisted of eight 1-hr sessions, spread over 4 weeks. The two treatment blocks were separated by 4 weeks without treatment sessions. Thus, therapist-patient contact was 16 hr, during an interval of 12 weeks. All patients were treated individually. Each treatment was given according to the description of elaborate manuals. During the sessions, much time was devoted to explanation and rehearsal of the treatment rationale. A common element in the first session of each treatment condition was a discussion of the bodily symptoms. The blushing, trembling or sweating was labelled as an idiosyncratic reaction on stress, and more or less as an individual handicap. It was stated that the aim of the treatment would not be never to blush, tremble or sweat again, but instead to diminish the fear of the symptoms. In addition, the vicious circle was discussed with the patient, in which it was shown how the bodily symptoms were exacerbated as a consequence of negative cognitions and avoidance of difficult situations. This introduction was identical in all conditions. However, explanation of the specific treatment strategy was different, depending on the condition (exposure, cognitive therapy or integrated treatment) to which the patient was assigned.

Exposure in uivo. The first session started with a thorough investigation of avoidance behaviour. Patients were asked which situations they feared (and avoided), and which behaviour they commonly used to prevent that others would notice the symptoms. In addition, the principles of exposure in viuo and response prevention were explained. The rationale emphasized the role of avoidance behaviour in the aetiology and maintenance of the fears. It was stressed that only frequent and prolonged confrontation with the feared stimulus without trying to escape or hide the symptoms would diminish the fear. The role of maladaptive and irrealistic thoughts was ignored. During the first session and the first part of the second session patient and therapist constructed a personal hierarchy of gradually more difficult exercises. Examples of frequently used exercises were: talking about difficult topics while keeping eye contact, wearing blouses with open necks or drinking coffee out of small cups in the company of others, and talking with other people about the fear of the symptoms. Response prevention pointed at behaviours like bending down, wearing sunglasses or make-up, escaping to the toilet, looking in the mirror, coughing, sitting in dark places, avoiding artificial light, putting a glass down, signing cheques at home, holding cups or glasses with two hands, etc. From the second session on the patient was requested to spend at least 90 min on homework exercises (exposure tasks) between each two treatment sessions. The first part of the treatment sessions consisted of evaluation of homework already done, and the second part of elaborate discussion of new tasks patients had to accomplish before the next session. During the sessions no exposure assignments were practised. The importance of prolonged confrontation was repeatedly stressed. With regard to situations that did not allow prolonged exposure (e.g. signing cheques, asking questions in a meeting, etc.) the patient was encouraged to practise as often as possible. If patients found it difficult to bridge the gap between two exposure tasks, therapists helped them to find intermediate tasks. No instruction about socially adequate behaviour was given, and no social skills were practised. Although patients were often instructed to engage in social behaviour, for example to start and maintain conversations or to make eye contact, the aim


was reduction of avoidance behaviour rather than improving social skills. On homework diary sheets patients recorded (1) the tasks they had practised, (2) anxiety before and after the task (both on a O-100 scale), (3) duration of the exercise and (4) problems they encountered in the situation.

Cognitive therapy. Rational emotive therapy (Ellis, 1962) was chosen because it had been found effective with social phobics in previous studies, and because behavioural tests or exposure assignments do not form a necessary and essential part of the treatment and can be omitted relatively easily. During the first session patients’ thoughts about the symptoms and about other people’s opinions were extensively inventarized. Afterwards, the principles of rational emotive therapy were explained, illustrated by means of ABCDE-schemata (Ellis, 1962; Walen, DiGiuseppe & Wessler, 1980). It was stressed that the fear and distress were not automatically elicited by the symptoms itself, but by maladaptive thought patterns on behalf of the patients. The importance of learning to recognize and to change the negative thoughts was explained. No instructions were given about entering difficult situations. If occasionally patients asked whether it was sensible to practise in situations, therapists responded they had to judge for themselves, and that it was most important to change the irrational thoughts that provoked anxiety instead of forcing oneself. After this session patients had to read a booklet, explaining the principles of rational emotive therapy (a Dutch translation of Young, 1974). In the treatment sessions non-realistic demands and wishes were traced and discussed, and replaced by more realistic thoughts. At first, attention was given to discussion of easily reported self-statements about the symptoms themselves, like “Other people will definitely notice that I tremble”, and “Blushing means that you hide something”. Gradually, attention shifted to more basic assumptions and rules, like: “I must have everything under control to prevent other people to hurt me” and “I must be perfect in everything to make other people like me”. The patient was asked to spend at least 30 min on homework every day, by making ABCDE-analyses about difficult situations that had occurred. The analyses were further discussed in the next treatment session. As much as possible, therapists used a Socratic-like dialogue to teach patients to analyse their own thoughts.

Integrated treatment. In this package therapists aimed at an optimal integration between exposure in vivo and cognitive therapy. Referring to the earlier discussed vicious circle it was explained that both avoidance behaviour and unrealistic thoughts were maintaining the fear, and that both aspects would receive attention in the sessions. The first session was devoted to explanation of the exposure principles and construction of the individual hierarchy, comparable to the procedure in the exposure treatment. After determining the hierarchy in the first part of the second session, the remainder of this session was dedicated to explanation of the rationale of cognitive therapy. From the third session on both techniques were integrated. The treatment sessions were spent on selection of exposure tasks that had to be practised as homework, on discussion and reformulation of irrealistic thoughts relating to those tasks, and on evaluating the exercises afterwards. Emphasis in the sessions was on tracing and changing maladaptive cognitions. Homework consisted of 90 min practising between each session, as the exposure condition. As a rule, exposure tasks were practised only after related thoughts were analysed and changed in the session. Generally speaking in both treatment blocks the same strategy was followed. Only in the case that exposure assignments passed very well and (almost) all tasks were mastered was homework amplified by analysis of thoughts using ABCDE-schemata.

Measures

Fear Questionnaire (FQ; Marks & Mathews, 1979). The FQ assesses avoidance of 15 common phobic situations, divided into three subscales: (1) Agoraphobia, (2) Social Phobia, and (3) Blood and Injury Fears. Each situation is rated on a 9-point scale, from 0 (not avoided at all) to 8 (always avoided). Only the Social Phobia subscale (range O-40) was used in the statistical analyses.

Social Cognition Inventory (SCZ; Van Kamp and Klip, 1981). This 35-item questionnaire assesses maladaptive cognitions in social situations. The instrument consists of two sections, one dealing with patients’ attitudes towards other people, e. g. ‘One has to be entertaining in the company of other people’ (17 items), the other consisting of items concerning self-image (18 items), e.g. ‘I often make myself ridiculous’. The items in the first section are rated on a S-point scale from 1 (disagree) to 5 (agree), and the items in the second section on a 5-point scale from 1 (false) to 5 (true). The

160 AGNES SCHOLING and PAUL M. G. EMMELKAMP

SC1 yields one (total) score (range 35-175). The questionnaire has high internal consistency (Cronbach’s a = 0.93: Van Meijgaard, Tros, Van der Molen and Wolters, 1987).

Lehrer- Woolfolk Anxiety Symptom Questionnaire (L WAS&, Lehrer and Woolfolk, 1982; &holing and Emmelkamp, 1992). This 36-item inventory covers somatic, behavioural and cognitive aspects of anxiety. The Somatic subscale (16 items) refers to bodily symptoms of anxiety, the Behavioural subscale (9 items) mainly represents avoidance of social situations, while the Cognitive subscale (11 items) measures a tendency to worry and ruminate. Each item is rated on a 5-point scale from 1 (never) to 5 (almost always). In this study the Somatic and Behavioural subscales were used (respective ranges 1680 and 9-45).

Symptom Checklist (SCL90; Derogatis, 1977). The SCL90 is a multidimensional measure of psychopathology. For this study the Dutch version of the scale (Arrindell and Ettema, 1986) was employed, consisting of the subscales: (1) Agoraphobia, (2) Anxiety, (3) Depression, (4) Somatic Complaints, (5) Insufficiency of Thinking and Acting, (6) Interpersonal Sensitivity and Mistrust, (7) Hostility and (8) Sleeping Problems. Each item is rated on a 5-point rating scale, from 1 (not at all) to 5 (very much). In this study only the subscales Depression and Somatic Complaints were used.

Social Anxiety Self-statements Inventory (SASSI, Mersch, Biigels, Hofman, van Hout, Scholing & Arntz, 1993a). The SASS1 is a 66-item self-report inventory, listing negative and positive self-statements in social situations. The instrument was constructed on the basis of thoughts which social phobic patients reported after they had participated in two social interaction tests (Biigels, Mersch, Arntz, Hofman & Van Hout, 1987). Patients have to rate how often they have the thoughts described on a 5-point rating scale, from 0 (never) to 4 (very often). In this study the negative subscale (SASSI-neg, range O-168, Cronbach’s a = 0.93) and the positive subscale (SASSI-pos, range O-96, Cronbach’s a = 0.92) were used.

Target situations. In the first treatment session patients were asked to name five personally relevant situations they feared and avoided because of fear of bodily symptoms. Examples were: ‘Visiting my parents-in-law and drinking two cups of tea’ or ‘Answering the telephone while being watched by a superior’. After formulation of five situations, patients were asked to rate each situation on a scale from 0 to 8, first on fear (0 = completely at ease, 8 = completely in panic), second on avoidance (0 = never avoided, 8 = always avoided). The same five situations were rated at each assessment. Only the avoidance subscale (range O-40) was used in the analyses.

Therapists

Treatments were conducted by advanced clinical psychology students. All therapists (n = 20) had followed introductory and advanced courses in behaviour therapy before they were admitted to the therapist team. An additional 8-weeks training (SO-100 hr) for adequate use of the treatment manuals was given. Specific parts of the treatments, like explanation of the treatment rationales, were repeatedly practised, and much attention was paid to the demarcation between exposure in vivo and cognitive strategies. In order to control for treatment integrity all treatment sessions were recorded on audiotape. After each session therapists overheard the tape in company of another therapist to discuss the therapeutic interventions. Two-hour supervision sessions were held twice a week with groups of 5-7 therapists, in which the progress of all treatments were discussed and evaluated, if desirable by means of the audiotapes.

RESULTS

Statistical analyses

To minimalize the risk of chance findings the dependent variables were divided into two groups. The first group, used for testing the hypotheses, consisted of four variables. The avoidance subscale of the target situations was used as a personally relevant outcome measure. In addition, three new variables were composed, according to the distinction in (avoidance) behaviour, cognitions, and somatic symptoms (Lang, 1971). The first variable, measuring avoidance of social situations, consisted of the mean of the standardized scores of the FQ Social Phobia subscale and the LWASQ Behaviour subscale. The second variable, measuring social phobic cognitions, was the mean of the


standardized scores on the SC1 and the SASSI-neg. The third variable, the mean of the SCL90 and LWASQ Somatic Complaints subscales, represented the somatic channel. Thus, four variables were chosen on an a priori basis for testing the hypotheses. Other variable (e.g. depression and positive self-statements) were only used for more descriptive purposes. Missing data were treated as follows. In a few cases no mean score could be computed for the new behavioural, cognitive or somatic variable because one of both composing variables was missing. In that case only the value of the other variable was used. Fore example if the FQ Social Phobia score was missing, the standardized score on the LWASQ Behaviour subscale was used in the analyses. Descriptive statistics and tests of hypotheses were calculated listwise: if one variable was missing, the patient was excluded from the analyses. For the analyses on waiting-list effects the results of all patients (n = 35) were included, whereas for testing overall and specific treatment effects only the patients who completed treatment (n = 30) were included. Repeated measures MANOVAs were carried out for testing waiting-list and overall treatment effects. Differential treatment effects were tested with univariate ANCOVAs with the pretests as covariate, because of initial differences between the groups.

Waiting-list us no waiting-list

First the effects of waiting-list vs no waiting-list were evaluated. Improvements during the first treatment block (4 weeks) of those patients who started treatment immediately were compared with the results of the no-treatment phase (4 weeks) of the waiting-list patients. Descriptive results are summarized in Table 2. Analysis of difference scores on the four variables together (MANOVA) yielded no significant effect of treatment vs no treatment: F(4,29) = 1.9, P < 0.13. Although on some variables a slight difference in favour of the treatment phase was suggested, it must be concluded that a 4-weeks treatment did not produce significantly better results than a waiting-list period. Comparison of waiting-list results with improvement during both treatment blocks (12 weeks) showed a significant difference: F(4,27) = 7.5, P < 0.000. These findings will be discussed later.

Overall eflect of treatment

Repeated measures MANOVAs were carried out for the whole group to test for an overall time effect after each treatment block and at 3-months follow-up. The results on the different assessment occasions for all patients who finished treatment are graphically presented in Fig. 2. Figure 2 suggests a consistent trend of overall improvement during both treatment blocks, with a relative arrest during the no-treatment phase between the treatments. Clearly the largest improvement in both treatment blocks was gained on avoidance of target situations.

Two MANOVAs were carried out to test for significant overall time effects after the first and the second treatment block. The multivariate and univariate results are given in Table 3. A highly

Table 2. Descriptive results (x and SD) waiting-list (WL) vs no waiting-list (NWL)

Measure As0 AS1 As2 As4

Target situations: Avoidance WL 23.9 (7.4) 22.0 (8.0) 17.0 (5.3) 12.5 (6.5) NWL 24.0 (7.5) 18.0 (4.9) 12.5 (6.9)

FQ Social Phobia WL 19.3 (8.2) 17.5 (6.3) 13.9 (5.9) 14.6 (6.8) NWL 20.3 (9.3) 16.9 (9.9) 13.3 (7.5)

Behaviour WL 0.4 (1.0) 0.2 (0.9) -0.3 (0.7) -0.7 (0.7) NWL l.O(l.3) 0.3 (1.2) -0.2 (1.0)

Cognitions WL 0.5 (1.3) 0.3 (1.2) 0.1 (1.1) -0.7 (0.9) NWL 0.9(1.1) 0.6 (0.9) -0.7 (0.9)

Somatic complaints WL 0.5 (1.5) 0.1 (1.5) -0.2(1.0) -0.5 (0.9) NWL 0.7 (1.4) 0.1 (1.0) -0.3 (0.8)

SCL90-Depression WL 28.2 (I 5.6) 25.9(13.1) 22.5 (7.6) 20.1 (6.6) NWL 30.9(10.1) 28.2 (8.7) 22.3 (6.0)

n = 15 in WL and NWL. AsO, waiting-list assessment; As1 , pretest first block; As2, post-test first block; As4, post-test second block.

162 AGNB SCHOLING and PAUL M. G. EMMELKAMP

1.50

1 .oo

0.50

0.00

-0.50

-1.00

-1.50

- tar

--at-- avoj

As 1 As2 As3 As4 FU

Assessment occasions

Fig. 2. Treatment effects on target situations (TAR), avoidance of social situations (AVOI), cognitions (COG) and somatic complaints (SOM). As1 and As2, pretest and post-test first block; As3 and As4, pretest

and post-test second block; FU, 3-months follow-up.

significant overall improvement was found after both treatment blocks, and this result was stable at follow-up. Univariate comparisons showed that the improvement was gained on each outcome variable, which was found for all assessment moments. Compared with the pretest of the first treatment the largest improvement was again found on the target situations, especially at follow-up.

Dijgerentiai treatment ezects

In Figures 3-6 the results for the treatment conditions on each of the outcome variables are presented grap~cally. The figures show that the treatment groups showed initial differences on each of the variables, although the differences were not very consistent across measures. The effects of the treatment packages were compared by means of univariate ANCOVAs with the pretest of the first treatment block as the covariate to control for the initial between-group differences. The results of the treatment packages were compared after the first treatment phase (exposure vs cognitive treatment vs integrated treatment), after the second treatment phase (cognitive therapy followed by exposure vs exposure followed by cognitive therapy vs integrated treatment) and at follow-up. The results on differential treatment effects are presented in Table 3.

Results after the first treatment block

On the whole, exposure in vivo, cognitive therapy and the integrated treatment yielded about the same progress at the post-test of the first treatment block on all variables. Although the figures

Table 3. Overall and differential treatment effects after each treatment block

After first After second block block Follow-up

Ouerall effects (MAN0 VA) Overall F(4,26) = 8.9*** F(4.24) = 24.0*** F(4.25) = 1 I .9+** Target situations t = s.o*** t = 8.1*** I = 6.31f9 Avoidance t = 4.0*** t = 5.5*** I = 5.9*** Cognitions I = 2.4’. t = 6.6*‘* t = X8*‘+ Somatic compliants I = 3.31.’ t = 4.7*** t = 3.2”

Di@erential effects (ANCOYA) Target situations F(2,26) = 2.0 F(2,24) = 3.0 F(2,25) = 2.2 Avoidance F(2,26) 3 0.2 F(2,26) = 0.2 F(2,25) = 1.2 Cognitions F(2,26) = 0.8 F(2,26) = 0.2 F(2.25) = 1.2 Somatic complaints F(2,26) = 0.0 F(2,26) = 0.7 F(2.26) = 0.2

l P < 0.1; **P < 0.01; l **p < 0.001.


1 .oo

0.50

0.00

-0.50

-1.00

-1.50

-2.00

+

-a-‘

. ..**..*

exp-cog

cog-exp

integ.

AslAs2As3As4 FU


Fig. 3. Diffemntial treatment effects on target situations (TAR). As1 and As2, pretest and post-test first block; As3 and As4 pretest and post-test second block; FU, 3-months foUow-up; cog-exp, cognitive therapy followed by exposure in t&o; exp-cog, exposure followed by cognitive therapy; integr., integrated

treatment.

suggest a slight superiority of the exposure and integrated treatments above the cognitive therapy, the differences were not significant, so that it must be inferred that exposure in uivo, cognitive therapy and a cognitive-behavioural treatment were equally effective.

Results after two treatment blocks

At the post-test of the second treatment block again no significant differences among the packages were present. This result was found for each of the outcome variables. It is concluded

- exp-cog

--di-. cog-exp

As 1 As2 As3 As4 FU

Assessments occasions

Fig. 4. Differential treatment effects on avoidance of social situations (AWN). As1 and As2, pretest and post-test first block; As3 and As4, pretest and post-test second block FU, 3-months follow-up; cog-exp, cognitive therapy followed by exposure in t&o; exp-cog, exposure followed by cognitive therapy; integr.,

integrated Trident.

164 AGNES%HOL~NG~~~ PAUL M.G.EMMELKAMP

- exp-cog

AS 1 As2 As3As4 FU


Fig. 5. D~fferent~ai treatment effects on cognitions (COG). As1 and As2, pretest and post-test first block; As3 and As4, pretest and post-test second block; FU, 3-months follow-up; cog-exp, cognitive therapy followed by exposure in D~UO; exp-cog, exposure followed by cognitive therapy; integr., integrated

treatment.

that it makes no difference in which order exposure in Vito and cognitive therapy are given, and also that an integrated treatment is not superior to a treatment in which the ingredients are applied consecutively.

Results at 3months follow-up

Although no clear differences were discovered, the figures suggest that patients in the cognitive- exposure condition showed some relapse at follow-up, after having gained the smallest improvement during both treatment blocks. These patients had the highest (worst) mean follow-up score

1.50

1.00

0.50

0.00

-0.50

-1.00

-1.50

--+-- exp-cog

-*-. cog-exp

As 1 As2 As3As4 FU


Fig. 6. Differential treatment effects on somatic complaints (SOM). As1 and As2, pretest and post-test first block; As3 and As4, pretest and post-test second block; FU, 3-months follow-up; cog-exp, cognitive therapy followed by exposure in t&o; exp-cog, exposure followed by cognitive therapy; integr., integrated

treatment.


on three of the four outcome variables, whereas they had relatively low scores at the start of treatment. However, univariate ANCOVAs showed no between-group differences on any of the outcome variables (Table 3).

Descriptive statistics

In Table 4 mean scores and standard deviations on the original variables for the whole group and the three treatment groups on all assessments are presented.

Five patients (17%) dropped out during treatment. Patient 1 (cognitive-exposure) stopped in the 10th session because she did not have confidence in the exposure strategy. Patient 2 (exposure-cognitive) was treated outside the study after the 1 lth session because she got increasingly anxious and chaotic, which seemed to be intensified through the focus on cognitions. Patient 3 (integrated treatment) got a cerebral haemorrhage after the seventh session. Although, after a break, he continued and completed treatment with success, his data were excluded from the analyses. Patient 4 (exposure-cognitive) stopped after four sessions because her father got a heart-attack. Patient 5 (exposure-cognitive) dropped out after the pretest of the second treatment block, because her fear had vanished completely. In Table 5 descriptive statistics for dropouts vs completers are presented. The data suggest that at pretest the dropouts did not differ from the completers on most measures. However, during the first treatment block they showed greater improvements than the completers, which they maintained at the pretest of the second treatment block. This result is clearly illustrated in Fig. 7.

Table 4. Descriptive statistics (R and SD) for patients who completed treatment

Measure As1 As2 As3 As4 FU

Target situations exp-cog 21.9 (8.4) cog-exp 22.2 (7.0) integr. 26.1 (7.5)

FQ Social Phobia exp-cog 21.7(9.2)

‘=g-exp 14.8 (5.3) integr. 20.2 (8.0)

LWASQ Behaviour exp-cog 28.0 (8.7) cog-exp 19.7 (5.1) integr. 21.9 (8.6)

SC1 exp-cog 100.0 (19.5) cog-exp 84.5 (15.1) integr. 91.6 (22.8)

SASS1 Neg exp-cog 87.1 (32.5) cog-exp 63.1 (17.2) integr. 67.0 (33.7)

SCL90 Somatic Complaints exp-cog 18.8 (5.7) cog-exp 19.3 (7.2) integr. 16.8 (7.0)

LWASQ Somatic Complaints exp-cog 30.4 (14.0) cog-exp 24.9(11.0) integr. 26.3 (9.6)

SASS1 Pos exp-cog 63.3 (17.3) cog-cxp 58.9 (11.0) integr. 57.6 (14.9)

SCL90 Depression exp-cog 31.2 (9.8) cog-exp 28.6 (15.7)

28.6 (10.9) 23.0 (7.5) 24.4 (6.4)

27.6 (8.6) 24.9 (10.9) 23.0 (6.3)

22.8 (6.5) 21.6 (8.0) 19.1 (3.8)

32.7 (6.7) 26.2 (8.7) 19.9 (6.5) integr. 24.8 (8.9)

AsO, waiting-list assessment; As1 and As2, pretest and post-test first bloclt; As3 and As4, pretest and post-test second block FU, 3-months follow-up; cog-exp, cognitive therapy followed by exposure in viw; exp-cog, exposure followed by cognitive therapy; integr., integrated treatment.

14.8 (5.2) 12.9 (6.0) 18.3 (4.4) 18.6 (5.2) 19.4 (4.8) 15.7 (6.9)

18.2 (10.0) 12.3 (7.3) 15.6 (6.4)

15.5 (9.0) 12.9 (6.7) 16.7 (8.0)

22.2 (7.9) 18.0 (4.5) 17.5 (6.0)

22.6 (8.5) 18.6 (6.4) 16.9 (5.8)

94.4(16.0) 81.7(13.9) 86.9 (24.9)

88.6 (19.7) 79.4 (9.2)

80.8 (25.1)

66.5 (37.3) 67.3 (19.4) 66.4 (30.9)

67.6 (34.0) 63.2 (23.3) 58.7 (27.7)

16.6 (4.5) 16.5 (5.5) 15.6 (4.2)

16.8 (5.6) 17.2 (6.9) 14.7 (2.9)

25.0 (10.3) 23.3 (8.6) 23.2 (5.7)

25.9(11.3) 25.0 (8.5) 22.2 (6.3)

59.8 (19.2) 55.1 (11.8) 56.2 (13.5)

57.2 (13.8) 60.0 (8.5) 57.7 (13.7)

9.6 (6.7) 15.0 (5.6) 13.2 (6.7)

12.9 (9.6) 11.1(5.7) 12.5 (5.0)

20.0 (8.4) 14.8 (3.8) 14.9 (3.5)

75.6 (18.9) 72.3 (10.3) 69.6 (15.6)

47.9 (32.6) 42.8 (20.7) 42.7 (15.5)

14.5 (3.4) 16.5 (5.2) 13.4 (2.5)

21.838.9) 30.3 (7.7) 21.1 (4.1)

56.0 (20.3) 58.3 (14.0) 46.2 (15.6)

9.6 (4.8) 15.5 (6.5) 12.0 (7.5)

12.7 (8.8) 11.6 (7.6) 12.1 (5.6)

18.4 (8.0) 17.1 (7.0) 16.0 (5.5)

74.0(17.5) 79.0 (9.9)

68.1 (18.8)

52.3 (34.5) 46.0 (17.4) 44.9 (26.9)

14.6 (2.5) 16.6 (5.2) 14.4 (2.6)

22.3 (7.8) 20.8 (4.4) 21.7 (5.1)

53.6(21.5) 56.8 (15.9) 53.6 (19.5)


Table 5. Descriptive statistics (2 and SD) completers (n = 30) vs dropouts (n = 5)

Measure ASI As2 As3

Target situations Completers 23.4 (7.6) Dropouts 23.2 (5.2)

FQ Social Phobia Completers 18.9 (8.0) Dropouts 21.0 (7.5)

LWASQ Behaviour Completers 23.2 (8.2) Dropouts 23.0 (8.3)

SC1 CompIeters 92.0 (19.8) Dropouts 82.2(15.7)

SASSI-neg Completers 72.6 (29.6) Dropouts 81.8(36.6)

SCL90 Somatic Complaints Completers 18.3 (6.5) Dropouts 24.3(11.2)

LWASQ Somatic Complaints Completers 27.2(11.5) Dropouts 34.2 (12.9)

SASSI-pos Compieters 59.0 (14.3) Dropouts 56.8 (14.6)

SCL90 Depression Completers 28.3 (11.8) Dropouts 33.2 (12.2)

17.5 (5.1) 12.8 (4.3)

15.4 (8.2) 18.5 (0.7)

19.2 (6.5) 14.3 (5.5)

87.7 (19.0) 70.3 (2.1)

66.7 (29.1) 45.0 (26.2)

16.2 (4.6) 22.7 (7.8)

23.8 (8.2) 32.7 (15.6)

57.0 (14.8) 59.0 (25.5)

25.3 (8.5) 24.0(13.0)

IS.1 (6.3) 10.8 (5.9)

15.0 (7.8) 13.0(8.1)

19.4 (7.2) llS(2.1)

82.9 (19.0) 69.8 (10.8)

63.0 (27.7) 41.3 (33.9)

16.2 (5.3) 22.0 (8.2)

24.4 (8.8) 26.3 (9.2)

58.4 (11.8) 58.5 (27.1)

25.2 (8.7) 29.0 (22.8)

As1 and As2, pretest and post-test first block; As3, pretest second block.

DISCUSSION

In this study several questions concerning treatment of fear of bodily symptoms were addressed. For the sake of clarity, results will be summarized first. No significant overall differences were found between improvements during 4-weeks waiting-list vs 4-weeks treatment, although descriptive statistics suggested that treatment was slightly more effective, especialiy on avoidance of social

3r

tar avoi cog som Variables

_ completers t&3)

completers (FL0

m chwouts c9s3)

Fig. 7, Mean improvements of completers (at As3 and FU] vs dropouts (at As3). As3, pretest second block; FU, 3-months follow-up; TAR, target situations; AVOI, avoidance of social situations; COG, cognitions;

SOM, somatic complaints.


situations. Further, comparison (at intermediate test) of exposure in viuo, cognitive therapy, and the integrated treatment yielded no significant differences. However, the figures suggested that results of cognitive therapy were worse than results of both other treatments. Comparison at post-test showed that the three packages (1) exposure in vivo followed by cognitive therapy, (2) cognitive therapy followed by exposure in viva, and (3) the integrated package were about equally effective. Differential effects, indicating that exposure especially improves avoidance behaviour and somatic symptoms, and does not alter cognitions, and that cognitive therapy has reverse effects, were not found.

For proper inte~retation of these results, several points deserve attention. First, to our knowledge, this is the first controlled study on (cognitive-)behavioural treatment of fear of bodily symptoms in social phobics, which means that comparison with other studies must be limited to studies on generalized social phobia.

A major finding of this study is the absence of significant differences between the treatment conditions. One explanation for this result is that treatment did not have any effects at all, which is also suggested by the fact that treatment did not yield better results than a 4-weeks waiting-list. Several findings do contradict this explanation. First, a highly significant overall time effect was found for all 3 treatment packages together. Second, comparison of waiting-list vs total treatment gains showed highly significant differences in favour of treatment. Third, on nearly all measures the data reflected a similar trend: improvements during the treatment blocks, with a relative arrest or even a slight relapse during no-treatment phases between and after treatment, which points to specific treatment effects. It is also in contrast with patients’ subjective reports of improvement at the end of treatment. Table 6 gives a review of FQ Social Phobia scores of social phobics before and after treatment and of normals. The data show that the mean improvement scores on the FQ Social Phobia subscale in this study were in the range of those reached in other studies. Jacobson & Revenstorf (1988) formulated two criteria for defining clinical significance of improvement. The first is the patients’ post-test scores are more likely to belong in the functional than in the disfunctional range, as is the case for the patients in this study. The second criterion is that the change (expressed in reliable change or RC-index) should be large enough to make it unlikely that it is an artefact of measurement error. RC-index in the present study ranges from 2.5 to 6.0, clearly above the size that was recommended (1.96). Finally, the use of standardized scores allows direct inference about mean effect sizes. Figure 7 illustrates that effect sizes larger than 1 were found on all variables except somatic complaints. These results indicate that improvements were clinically significant.

Table 6. Severity of complaints (x and SD) at pretest and treatment progress across studies

FO Social Phobia

Pretest Post-test Gain scores

Patienls Butler er al. (1984) 5.0 Mattick & Peters (1988)

exP 19.5 (8.3) 14.8 (8.9) 4.7 comb 21.4(11.1) 14.8 (8.9) 6.6

Mattick ef al. (1989) cr 26.9 (8.5) 15.0 (6.0) 11.9 =xP 16.2 (5.5) 12.0 (5.7) 4.2 comb 22.3 (7.8) 11.9 (6.3) 10.4

Heimberg ef al. (1990) Discrete SP 18.2 (5.8) - - Generalized SP 21.6 (7.0) - -

Scholing & Emmelkamp (this study) exp-cog 21.7 (9.2) 12.9 (9.6) 8.8 cog-exp 14.8 (5.3) 11.1(5.7) 3.1 integr. 20.2 (8.0) 12.5 (5.0) 1.7

NOIWUZIS Koopmans (1990) 10.8 (6.6) - -

9.3(6.7) - - Anindell & Buikhuisen (1992) 10.2 (7.4Y - -

cog, cognitive therapy; comb. combined treatment; cr, cognitive restructuring; exp, exposure in ~iuo; integr., integrated treatment; SP, social phobics.

*Pooled from separate data for men and women.


A second explanation for the absence of significant differences is that the first treatment block was too short to produce a significantly better result than the waiting-list period. Two processes may be responsible for this finding. First, it has been reported that short waiting-list periods, and even the mere talking about complaints in an intake session, can have beneficial effects. In the present study all patients had two sessions (one intake session and a session with the therapist) before they were assigned to waiting-list or treatment. Factors such as telling two different people about the complaints, positive expectations about treatment and keeping complaint-oriented diaries during the waiting-list period may already have produced some progress during the waiting-list period. Second, therapists often reported that treatment progress was reached relatively slowly. Although in the first treatment session it was agreed with the patient that treatment would centre on the fear of the symptoms, most patients in fact hoped that they would never blush, sweat or tremble again. Only after several weeks did they gradually start to accept the symptoms and to be less upset about them. The combination of these factors may be responsible for the absence of significant differences after the first treatment phase.

Several explanations are also possible for the lack of differential treatment effects. In the first place it should be remembered that clear and consistent differences among treatments were not found in studies on more generalized social phobia (Butler et al., 1984; Emmelkamp et al., 1985; Mattick & Peters, 1988; Mattick et al., 1989). It is plausible that similarities between treatment conditions (like discussing the vicious circle, explaining a rationale, working complaint-oriented, time-limited and intensively, encouraging patients to tell their ‘secret’) clearly exceed the differences between exposure and cognitive strategies. This would also explain the finding that neither treatment differentially affected avoidance behaviour, cognitions, or somatic complaints. Biran et al. (1981) did find exposure in uivo to be superior to cognitive therapy on decreasing avoidance behaviour. First, this may be caused by the fact that they treated only patients with writing anxiety, which seems to be a more circumscribed problem than fear of blushing or trembling. Second, exposure in uiuo in their study was therapist guided, which may give better results for this kind of complaint.

The absence of clear treatment differences may be caused partly by a very different factor, viz. the lack of adequate assessment instruments. Although patients with fear of bodily symptoms in general may show less anxiety and avoidance behaviour than generalized social phobics, they are very anxious in specific situations, as is suggested by the mean pretest scores on the target situations. Largest improvements during treatment were found especially for this variable, which suggests that the available self-report measures are not sufficiently attuned to the specific fears of this subgroup of social phobics.

Results of patients who completed treatment vs dropouts contradicted the assumption that patients drop out because their complaints are more severe or because they do not improve. In this study pretest differences were found only on depression and somatic complaints, but not on target situations, avoidance of social situations, or irrational cognitions. In addition, dropouts clearly showed greater progress during the first treatment block than completers on almost all measures. In spite of this, they still reported more depression and somatic symptoms at the moment they ended treatment. This suggests that, although treatment was effective in decreasing social phobia, other problems were more important at the time of dropout.

This study has several implications for further research. Effects of cognitive-~havioural treatments for fear of bodily symptoms should be compared with effect of a longer waiting-list period, placebo treatments, or treatments with beta-blockers, to evaluate whether cognitive-behavioural strategies do give additional improvements. Furthermore, during treatments it often emerged that patients were very ashamed of their complaints, never talked about them with other people, and were convinced that their fears were rare and strange. Talking about their ‘secret’, first with the therapist, then with people around them, taught them that other people either had similar symptoms, or had never noticed the patients’ symptoms, which gave much relief. In that respect, group treatments may be especially promising for this sort of complaint. More research is needed to Iind out which factors contribute to dropout, and to differentiate between patients who stop because they are sufIiciently improved vs patients who are dissatisfied with treatment. Finally, evaluation of treatment studies will remain difficult as long as there is a lack of adequate assessment

Cognitive-behavioural treatment I69

instruments. Research should aim at developing instruments that are specifically sensitive to assessing fear of bodily symptoms and related aspects.

Acknowledgemenrs-This report is part of Agnes &holing’s doctoral dissertation on Etiology and Treatment of Social Phobia, which was conducted at the University of Groningen, The Netherlands, and was supervised by Paul M. G. Emmelkamp. The authors gratefully acknowledge the assistance of the therapists for conducting treatments and collecting data, and of Laura van Bergen for preparing the data for statistical analyses

REFERENCES

American Psychiatric Association (1980). Diagnosfic and statistical manual of mental disorders (3rd edn). Washington, DC: Author.

American Psychiatric Association (1987). Diagnostic and statistical manual of mental disorders (3rd revised edn). Washington, DC: Author.

Arrindell, W. A. & Buikhuisen, M. (1982). Dissimulation and the sex difference in self-assessed fears: A brief note. Eehaviour Research and Therapy, 30, 307-3 11.

Arrindell, W. A. & Ettema, J. H. M. (1986). XL-90, Handleiding bij een muhidimensionelepsychoparhologie indicator. Lisse: Swets & Zeitlinger.

Beck, A. T. & Emery, G. (1985). Anxiety disorders and phobias: A cognirive perspective. New York: Basic Books. Biran, M., Augusto, F. & Wilson, G. T. (1981). In vivo exposure versus cognitive restructuring in the treatment of

scriptophobia. Behaviour Research and Therapy, 19, 525-532. Boeringa, J. A. (1983). Blushing: a modified behavioural intervention using paradoxical intention. Psychotherapy: Theory,

Research and Practice, 20, 441444. Bagels, S. M., Mersch, P. P. A., Arntz, A. R., Hofman, A. & Van Hout, W. J. P. J. (1987). Construction and validation

of a self-statement questionnaire for social phobics. Paper presented at the Congress of fhe European Association of Behaviour Therapy, Amsterdam.

Butler, G., Cullington, A. Munby, M., Amies, P. & Gelder, M. (1984). Exposure and anxiety management in the treatment of social phobia. Journal of Consulting and Clinical Psychology, 52, 642650.

Clark, D. M. (1986). A cognitive approach to panic. Behaviour Research and Therapy, 24, 461470. Derogatis, L. R. (1977). XL-90 Administration, scoring and procedures manual I for the R(evised) version and other

instruments of the psychoparhology rating scale series. Baltimore, MD: Clinical Psychometrics Research Unit, Johns Hopkins University School of Medicine.

Di Nardo, P. A., Barlow, D. H., Cerny, J. A., Vermilyea, B. B., Himadi, W. G. & Waddell, M. T. (1985). Anxiety Disorders Interview Schedule Revised (ADIS-R). Albany. NY: Center for Stress and Anxietv Disorders,

Ellis, A. (1962). Reason and emorion in psychotherapy. New York: Lyle-Stuart. . Emmelkamp, P. M. G., Mersch, P. P., Vissia, E. & Van der Helm, M. (1985). Social phobia: a comparative evaluation

of cognitive and behavioural interventions. Behaviour Research & Therapy. 23, 365-369. Heimberg, R. G. (1989). Cognitive and behavioural treatments for social phobia: a critical analysis. Clinical Psychology

Review, 9, 107-128. Heimberg, R. G., Hope, D. A., Dodge, C. S. & Becker, R. E. (1990). DSM-III-R subtypes of social phobia: comparison

of generalized social phobics and public speaking phobics. Journal of Nervous and Mental Disease, 178, 172-179. Jacobson, N. S. & Revenstorf, D. (1988). Statistics for assessing the clinical significance of psychotherapy techniques: issues,

problems and new developments. Behavioral Assessment, 10, 1333145. van Kamp, I. & Klip, E. (1981). Cognitive aspecten van subassertief gedrag. Gedrugstherapeutisch Bulletin, 14, 45-56. Koopmans, P. C. (1990). Indicatoren van agorafobie: Enkele vragenlijstonderzoeken in klinische en ‘normale’ proefgroepen.

University of Groningen: unpublished manuscript. Lamontagne, Y. (1978). Treatment of erytrophobia by paradoxical intention. Journal of Nervous and Menfal Disease, 166,

304306. Lang, P. J. (1971). The application of psychophysiological methods to the study of psychotherapy and behavior

modification. In Bergin, A. E. 8t Garfield, S. L. (Eds.) Handbook of psychotherapy and behavior change (pp. 75-125). New York: Wiley.

Lehrer, P. M. & Woolfolk, R. L. (1982). Self-report assessment of anxiety: somatic, cognitive and behavioural modalities. Behavioral Assessment, 4, 167-177.

Marks, I. M. & Gelder, M. G. (1966). Different ages of onset in varieties of phobias. American Journal of Psychiatry, 123, 218-221.

Marks, I. M. & Mathews, A. M. (1979). Brief standard self-rating for phobic patients. Behnviour Research and Therapy, 17, 5968.

Mattick, R. P. & Peters, L. (1988). Treatment of severe phobia: effects of guided exposure with and without cognitive restructuring. Journal of Consulting and Clinical Psychology, 56, 251-260.

Mattick, R. P., Peters, L. & Clarke, J. D. (1989). Exposure and cognitive restructuring for social phobia: a controlled study. Behavior Therapy, 20, 3-23.

van Meijgaard, I., Tros A., van der Molen, H. & Wolters, F. (1988). De Sociale Cognitielijyst. Een nadere validatie. De Psycholoog, 22, 383-389.

Mersch, P. P. A., Biigels, S. M., Hofman, A., van Hout. W. J. P. J.. Scholina. A. & Arntz. A. R. (1993al. Construction I and validation of-the Social Anxiety Self-Statements Inventory (SASSI). In preparation.

_ I

Mersch, P. P. A., Hildebrand, M., van Hout. W. J. P. J.. Lavv. E. H. & Wessel. I. (1993b). Somatic svmotoms in social phobia: A treatment method based on RET and paradoxical interventions. Submitted’for publica;on’.

Scholing, A. & Emmelkamp, P. M. G. (1991). Social phobia: nature and treatment. In Leitenberg, H. (Ed.), Handbook of social and evaluation anxiety. New York: Plenum Press.

Scholing, A. & Emmelkamp, P. M. G. (1992). Self-report assessment of anxiety: a cross validation of the Lehrer Woolfolk Anxiety Symptom Questionnaire in three populations. Behaviour Research and Therapy, 30, 521-531.

170 AGNES &HOLING and PAUL M. G. E=LKA~~P

van Son, M. J. M. (1978). Sociale v~r~gheidstherapie bij bloosproblemen en slaapproblemen. In van Son, M, J. M., Sock& vaardighei~fherapie, gedragstherapie en sociaat gedrag. Amsterdam: Swets & Zeitlinger.

Turner, S. M., Beidel, D. C., Dancu, C. V. & Keys, D. J. (1986). Psychopathology of social phobia and comparison to avoidant personality disorder. Journal of Abnormal Psychology, 95, 389-394.

Walen, S. R., DiGiuseppe, R. & Wessler, R. L. (1980). A practitioner’s guide to Rational Emotive Therapy. Oxford: University Press.

Welkowitz, L. A. & Liebowitz, M. (1990). Pharmacological treatment of social phobia and performance anxiety. In Noyes, R., Roth, M. & Burrows, G. D. (Bds), Handbook of anxiety, Vol. 4: The treatment of anxiety. Amsterdam: Elsevier.

Young, H. S. (1974). A Rational Emotiue Counseling Primer. New York: Institute for Rational Emotive Therapy.

Documents

Cognitive and behavioural treatments of fear of blushing, sweating or trembling