Copyright by Jasper Antonius Smits 2004
The Dissertation Committee for Jasper Antonius Smits Certifies that this is the approved version of the following dissertation:
Facilitating Public Speaking Fear Reduction by Increasing the Salience of Disconfirmatory Evidence
Committee:
Michael J. Telch, Supervisor
Eric Stice
James Pennebaker
Patrick Randall
Stephanie Rude
Facilitating Public Speaking Fear Reduction by Increasing the Salience of Disconfirmatory Evidence
by
Jasper Antonius Smits, M.A.
Dissertation Presented to the Faculty of the Graduate School of The University of Texas at Austin in Partial Fulfillment of the Requirements for the Degree of
Doctor of Philosophy
The University of Texas at Austin May, 2004
Facilitating Public Speaking Fear Reduction by Increasing the Salience of Disconfirmatory Evidence
Publication No.______
Jasper Antonius Smits, Ph.D. The University of Texas at Austin, 2004
Supervisor: Michael J. Telch
Abstract: The primary objective of the current study was to investigate whether adding videotape feedback procedures to exposure treatment would facilitate public speaking fear reduction. Participants meeting DSM-IV criteria for social phobia were randomized to receive non-pill placebo, exposure without videotape feedback, exposure with audience videotape feedback, or exposure with performance videotape feedback. Assessments were conducted at pretreatment, posttreatment, and 1-month follow-up. Having participants contrast actual audience responses to responses they imagined did not facilitate changes in participants’ predictions about the negative consequences of appearing anxious in front of others, nor did it result in greater anxiety reduction compared to exposure alone. On the other hand, having participants contrast their imagined performance to their actual performance accelerated the speed of improvement in public
iv speaking anxiety. Further, this differential effect appeared to be cognitively mediated. Specifically, the speedier improvement in the performance videotape feedback condition was partially accounted for by changes in participants’ estimates of the likelihood of an anxious appearance when performing in front of others. Follow-up data revealed no differences among the exposure treatment conditions, suggesting that the enhanced efficacy associated with performance feedback was not durable. In general, the pattern of the findings underscored that social phobia is a severe form of anxiety pathology, that warrants a high dose of treatment. Theoretical and clinical implications are discussed.
v Table of Contents
Chapter 1: Introduction ...... 1 1.1 Defining Features of Social Phobia...... 1 1.1.1 Nature and Epidemiology...... 2 1.1.2 Indirect and Direct Costs of Illness ...... 5 1.2 Etiology of Social Phobia...... 8 1.2.1 Neurobiology...... 8 1.2.2 Learning Theory...... 12 1.2.3 Cognitive Theory...... 15 1.3 Treatment Approaches for Social Phobia...... 19 1.3.1 Pharmacological Approaches...... 19 1.3.2 Psychosocial Approaches...... 20 1.3.2.1 Behavioral Treatments ...... 20 1.3.2.2 Cognitive-Behavioral Treatments ...... 21 1.3.2.3 Other Psychosocial Treatments...... 22 1.4 Treatment Outcome...... 23 1.4.1 Short-Term Efficacy...... 23 1.4.2 Long-Term Efficacy...... 26 1.4.3 Quality of Life...... 27 1.4.4 Cost Effectiveness...... 28 1.4.5 Predictors of Clinical Response ...... 28 1.5 Mechanisms of Fear Reduction...... 30 1.5.1 Habituation and Extinction...... 30 1.5.2 Cognitive Reappraisal ...... 32 1.5.3 Emotional Processing...... 33 1.5.4 Self-Efficacy...... 35
vi Chapter 2: The Present Study...... 37 2.1 Facilitating Fear Reduction ...... 37 2.2 Hypotheses ...... 40 2.2.1 Outcome Hypotheses...... 40 2.2.2 Process Hypotheses ...... 41 3.1 Participants...... 42 3.2 Experimental Design...... 42 3.3 Assessment...... 43 3.3.1 Diagnostic Assessment...... 43 3.3.2 Liebowitz Social Anxiety Scale ...... 43 3.3.3 Speech Task...... 44 3.3.3.1 Subjective Peak Fear ...... 45 3.3.3.2 Threat Appraisals ...... 45 3.3.4 High End-State Functioning...... 46 3.3.5 Relapse ...... 46 3.4 Procedure...... 47 3.4.1 Screening...... 47 3.4.2 Treatment Conditions...... 48 3.4.2.1 Exposure Conditions ...... 48 3.4.2.2 Placebo Condition ...... 50 3.5 Steps for Enhancing Treatment Integrity ...... 52
Chapter 4: Statistical Analyses...... 53 4.1 Baseline Differences ...... 53 4.2 Treatment Outcome...... 53 4.2.1 Effects at Posttreatment...... 54 4.2.2 Effects at Follow-up ...... 54 4.3 Mediator Analyses...... 55 4.4 Moderator Analyses ...... 55
vii Chapter 5: Results ...... 56 5.1 Characteristics of Participants...... 56 5.2 Drop Outs and Number of Sessions Attended ...... 57 5.3 Baseline Differences ...... 57 5.4 Effects at Posttreatment...... 58 5.4.1 Within-Group Effects...... 58 5.4.2 Between-Group Effects ...... 61 5.4.3 High End-State Functioning...... 61 5.5 Effects at Follow-up...... 62 5.5.1 Within-Group Effects...... 62 5.5.2 Between-Group Effects ...... 64 5.5.3 Relapse and High End-State Functioning ...... 64 5.6 Mediator Analyses...... 65 5.6.1 Effect of Treatment on Mediator (Step 1)...... 66 5.6.2 Effect of Treatment on Outcome (Step 2)...... 67 5.6.3 Relationship between Mediator and Outcome (Step 3) ...... 67 5.6.4 Effect of Treatment after Controlling for Mediator (Step 4)...... 68 5.7 Moderator Analyses ...... 69 5.8 Clinically Meaningful Differences and Statistical Power...... 70
Chapter 6: Discussion...... 72 6.1 Efficacy ...... 72 6.2 Mechanisms of Change ...... 76 6.3 Clinical Implications ...... 79 6.4 Limitations ...... 81
References ...... 85
Vita ...... 99
viii CHAPTER 1: INTRODUCTION
1.1 Defining Features of Social Phobia
It is almost her turn. Her heart is pounding. She notices hot flashes, and is breathing faster. With her shaking hands she takes another sip of water, and again rehearses sentences mentally in order to manage her anxiety. Last night, she did not sleep well because she was worried that this was going to happen; the course instructor had asked the students to briefly introduce themselves. As each student before her speaks, it feels as if the grim reaper is closing in to take her life. In a state of panic and dissociation she introduces herself to the class. A big wave of relief spills over her as the girl next to her starts her introduction. It feels like the end of a horrible nightmare. This moment of relief is quickly followed by a series of thoughts about her performance. She is convinced that she was blushing and that other people picked up on it. Knowing that her classmates saw how afraid she was when she spoke, she is sure that she made a fool of herself. She leaves class and heads to the student advisor’s office. It is the twelfth day of class; she can still drop her classes without a financial penalty. Although the term social phobia was first used by Janet in 1903 (as cited by Heckelman & Schneier, 1995), descriptions of social fears appear throughout history. Perhaps most well known is the report of Greek philosopher, Hippocrates (as quoted in Marks, 1969): "He dare not come in company for fear he should be misused, disgraced, overshoot himself in gesture or speech or be sick; he thinks every man observes him."
1 1.1.1 Nature and Epidemiology
According to the DSM-IV (American Psychiatric Association, 1994), the central feature of social phobia, also referred to as social anxiety disorder (Liebowitz, Heimberg, Fresco, Travers, & Stein, 2000), is a marked and persistent fear of social or performance situations driven by the unrealistic expectation of negative evaluation by others. Although recognizing that the fear is unrealistic and excessive, social phobic sufferers avoid social situations or endure them with significant anxiety and distress. Moreover, a DSM-IV diagnosis of social anxiety disorder warrants that the fear of social situations and related avoidance cause significant interference with the normal routine, occupational functioning, or social activities and relationships (American Psychiatric Association, 1994) The DSM-IV (American Psychiatric Association, 1994) makes a distinction between generalized and non-generalized social phobia. Individuals suffering from generalized social phobia report fear in most public performance and social interaction situations. Non-generalized social phobia sufferers typically report fears related to one single performance situation or several social situations. There is now ample evidence suggesting that generalized social anxiety represents a more severe form of psychopathology. Compared to people with the non- generalized subtype, people with generalized social phobia report higher symptom persistence, higher rates of Axis-I and Axis-II comorbidity, and more significant functional impairment (Heimberg, Hope, Dodge, & Becker, 1990; Kessler, Stein, & Berglund, 1998; Stein & Chavira, 1998; Turner, Beidel, & Townsley, 1992; Wittchen, Stein, & Kessler, 1999).
2 Estimates of the lifetime prevalence of social phobia in the general population vary from study to study. Studies conducted in the 1980s yielded a lifetime prevalence rate of approximately 2% (Pollard & Henderson, 1988; Schneier, Johnson, Hornig, Liebowitz, & Weissman, 1992). Diagnostic interviews in these studies followed DSM-III criteria (American Psychiatric Association, 1980), in which social phobia was more strictly defined than in the later editions of the DSM. The most significant change from the DSM-III to the DSM-III-R ( American Psychiatric Association, 1987) was that persons who met criteria for avoidant personality disorders were no longer excluded from the diagnosis of social phobia. Utilizing the DSM-III-R criteria, the National Comorbidity Survey (NCS; Kessler et al., 1994) collected diagnostic interview data from a sample of over 8,000 USA citizens ranging from age 15 to age 64. Social anxiety yielded a prevalence rate of 13.3%, thereby ranking among the most prevalent psychiatric disorders, following only alcohol dependence and major depressive disorder (Kessler et al., 1994). Follow-up analyses have revealed higher life-time prevalence among younger individuals (between age 15 and 34) than among older individuals (between age 36 and 64), suggesting an increase in the prevalence of social phobia (Magee, Eaton, Wittchen, McGonagle, & Kessler, 1996). Moreover, of the individuals suffering from social phobia, one-third report only fear connected to public speaking. Two-thirds reported multiple performance and interactional fears, thereby resembling the generalized social phobia subtype (Kessler et al., 1998). Recently, Heimberg and colleagues (Heimberg, Stein, Hiripi, & Kessler,
3 2000) demonstrated that the observed increase in prevalence of social phobia was only evident for the generalized subtype. A related question is the incidence of new cases per year. Neufeld, Swartz, Bienvenu, Eaton, and Cai (1999) estimated the incidence to be approximately 5 per 1,000 per year. According to NCS data, social phobia is more common in women (15.5%) than in men (11.1%) (Kessler et al., 1994). This 3:2 female-to male ratio has been observed in most prevalence studies (Pollard et al., 1988; Schneier et al., 1992; Turk et al., 1998). Although present in children, the onset of social phobia typically occurs in early to late adolescence. The Epidemiological Catchment Area (ECA) Study, which surveyed over 13,000 adults, yielded a mean onset age of 15.5 years (Schneier et al., 1992). Comparisons of the two subtypes have revealed that generalized social phobia patients have an earlier onset age than non-generalized social phobia patients (Brown, Heimberg, & Juster, 1995; Mannuzza et al., 1995). Social phobia is a chronic unremitting illness when left untreated. Retrospective and prospective studies have reported a mean average duration of illness ranging from 18 years (Reich, Goldenberg, Goisman, Vasile, & Keller, 1994) to 29 years (Chartier et al., 1998). Psychiatric comorbidity is often observed in those with social phobia (Amies, Gelder, & Shaw, 1983; Schneier et al., 1992; Sanderson, DiNardo, Rapee, & Barlow, 1990; Turner, Beidel, Borden, Stanley, & Jacob, 1991), with approximately 50% of patients meeting criteria for one or more comorbid Axis-I or Axis-II disorders (Sanderson et al., 1990). In most cases, social phobia
4 precedes the onset of these comorbid disorders (Schneier et al., 1992). With a rates ranging between 20% and 33%, generalized anxiety disorder is the most common comorbid diagnosis among social phobic sufferers (Barlow, DiNardo, Vermilyea, Vermilyea, & Blanchard, 1986; de Ruiter, Rijken, Garssen, Van Schaik, & Kraaimaat, 1989; Sanderson et al., 1990; Turner et al., 1991). Panic disorder, obsessive-compulsive disorder and specific phobias also frequently co- occur with social phobia. Other commonly observed additional Axis-I diagnoses include substance abuse and mood disorders. Schneier et al. (1992) found a formal diagnosis of alcohol abuse in 68 of 361 (19%) social phobic sufferers. Similarly, drug abuse was observed in 13% of the sample. While only a small subset met diagnostic criteria for bipolar disorder (5%), high rates were found for dysthymia (13%) as well as for major depression (17%), Among Axis-II diagnoses, avoidant personality disorder is the most common comorbid disorder, followed by obsessive-compulsive personality disorder. Turner et al., (1991) found that these two personality disorders accounted for over 70% of the Axis-II diagnoses observed among social phobic sufferers. It is not surprising that avoidant personality disorder often co-occurs with social phobia, as the DSM lists similar criteria for the two conditions. In fact, Herbert, Hope, and Bellack (1992) found that the vast majority of persons suffering from avoidant personality disorder also met criteria for social phobia.
1.1.2 Indirect and Direct Costs of Illness
Social phobia is associated with significant impairment. This perception was first reinforced by results from the ECA study. Even after controlling for
5 comorbidity, people suffering from social phobia showed an elevated rate of suicidal ideation compared to people without social phobia (Schneier et al., 1992). Similarly, compared to non-anxious controls, social phobic sufferers reported lower levels of self-confidence (Davidson, Hughes, George, & Blazer, 1994), and greater impairment in areas such as general health (Davidson et al., 1994; Katzelnick et al., 2001; Safren, Heimberg, Brown, & Holle, 1997; Schneier et al., 1994), family relationships, marriage or romantic relationships, and social network (Katzelnick et al., 2001; Schneier et al., 1994). Kessler et al. (1998) examined whether the rate of impairment varied as a function of social phobia subtype. Although non-generalized social phobia (i.e., fear associated with public speaking) was associated with significant disability, significantly more pervasive impairment was seen in generalized phobic sufferers. In addition to emotional and health impairment, social phobia often contributes to significant impairment in occupational functioning. The ECA study found that the prevalence rate for social phobia was inversely related to social economic status (Schneier et al., 1992). More than 85% of a sample of social phobic persons indicated interference with occupational functioning (i.e, work or academic functioning) as result of their anxiety (Turner, Beidel, Dancu, & Keys, 1986). Compared to non-anxious controls, social phobic sufferers reported more work attendance problems or poorer school grades (Davidson et al., 1994), lower education, and lower wages (Katzelnick et al., 2001) and financial dependency (Schneier et al., 1992).
6 Given the adverse effects of social phobia on health and quality of life, it is not surprising that the disorder is associated with increased use of medical services. Data from a recent survey conducted in two outpatient clinics of a large mental health organization indicated significantly more outpatient visits among people with social phobia than those with no diagnosis (Katzelnick et al., 2001). The results showed approximately 9 annual visits among social phobia patients and 11 annual visits among social phobia patients with comorbid psychopathology. These rates were comparable to those observed in patients with major depression (10 annual visits). The ECA study also reported increased health care seeking (Schneier et al., 1992; Davidson et al., 1994). Moreover, Davidson et al. (1994) found a fourfold rate of tranquilizers utilization (relative to the normal controls) among people with social phobia compared to healthy controls. The staggering costs of anxiety disorders have been well documented. Using the data from the NCS, Greenberg et al. (1999) estimated the total costs of anxiety disorders at 63.1 billion in 1998 dollars. This amount comprised medical treatment costs, psychiatric treatment costs, indirect workplace cost, and mortality costs. Greenberg and colleagues did not report the specific amount accounted for by social phobia, but showed that most of the cost of social phobia was comprised of workplace costs. This suggests that treatment expenses among social phobia sufferers are not significantly different from those reported by non-sufferers. A recent study from the UK corroborated these findings (Patel, Knapp, Henderson, & Baldwin, 2002). They found that social phobic suffers reported higher levels of drug dependency and use of prescribed medication. However, they did not report
7 higher total health care costs, except for expenses resulting from general practitioner visits. In summary, social phobia is the most prevalent of all anxiety disorders, and it is associated with impaired quality of life. Moreover, social phobia sufferers commonly develop additional psychological disorders such as substance abuse and depression. The financial costs of social anxiety are considerable, mostly due to lost productivity.
1.2 Etiology of Social Phobia
1.2.1 Neurobiology
Several approaches have helped advance the understanding of the neurobiology of social anxiety. Twin studies, where a comparison is made between monozygotic and dizygotic twins have enabled us to make inferences regarding the genetic contribution to the development social anxiety. A series of studies by Kendler and colleagues (Kendler, Myers, Prescott, & Neale, 2001; Kendler, Neale, Kessler, Heath, & Eaves, 1992) suggest only moderate genetic influences in social phobia. Their first study was an investigation of over 2,000 female twin pairs. The heritability index was estimated at 30%, as concordance rates for social phobia were 24.4% for monozygotic twins and 15.3% for dizygotic twins (Kendler et al., 1992). Similar results were obtained in their latest study, which examined approximately 1,200 male twins. The authors concluded that both genetic and environmental components account for significant variance in the prevalence of social phobia (Kendler et al., 2001).
8 Consistent with these conclusions are the findings from family studies. Family studies typically assess the first-degree relatives of those with a present diagnosis of the disorders of interest (i.e., probands), and compare those to first- degree relatives of control participants. Fyer, Mannuzza, Chapman, Liebowitz, and Klein (1993) examined a sample of 30 probands and 77 controls, and found a significantly increased risk among relatives of social phobia probands compared to relatives of healthy controls (16% vs. 5%). A follow-up study revealed that this increased risk was limited to the generalized social phobia subtype (Mannuzza et al., 1995). These findings have been replicated by Stein et al. (1998). Recently, Stein (Stein, 1998; Stein, Chartier, Lizak, & Jang, 2001) argued that it is unlikely that the disorder has a familial basis. Instead, he suggested that it may be risk factors such as behavioral inhibition that are transmitted. Preliminary findings are consistent with this hypothesis (Mick & Telch, 1998; Stein et al., 2001; Stein, Jang, & Livesley, 2002). Prospective studies are needed to address this hypothesis more appropriately. Laboratory challenge studies such as inhalation of carbon-dioxide enriched air have greatly advanced the understanding of the etiology of panic disorder (Zvolensky & Eifert, 2001; Sinha, Papp, & Gorman, 2000). Although few in number, findings of laboratory challenge studies are indicative of an increased sensitivity to respiratory arousal among social phobic patients compared to healthy control participants (Caldirola, Perna, Arancio, Bertani, & Bellodi, 1997; Gorman et al., 1990; Papp et al., 1993; Pine et al., 2000).
9 Another challenge paradigm applied to social phobia is the impromptu speech task, where the objective is to measure autonomic nervous system functioning. Research findings suggest that autonomic dysfunction may be evident in the generalized subtype, as they tend to show little physiological arousal as measured by heart rate reactivity, despite their heightened self-reported anxiety. This incongruity is not evident in the non-generalized subtype (Boone et al., 1999; Heimberg et al., 1990; Hofmann, Newman, Ehlers, & Roth, 1995). Neuroimaging studies have provided some support for the hypothesis that altered dopaminergic functioning is implicated in the etiology of social phobia. Using single photon emission computerized tomography (SPECT), Tiihonen et al. (1997) examined striatal dopamine reuptake site densities in social phobic persons and healthy controls. They found lower dopamine reuptake site density among social phobic sufferers, which led them to conclude that it is probable that social phobic patients have a smaller number of dopaminergic synapses and neurons in the basal ganglia. Schneier et al. (2000) applied a similar methodology to assess dopamine (D2) receptor binding in 10 social phobia sufferers and 10 healthy control subjects. The mean dopamine receptor binding potential was significantly lower for social phobia persons compared to the healthy controls. However, the binding potential was not significantly correlated with self-report measures of social anxiety among the social phobia participants. Few other neuroimaging methods have been used to examine the central nervous system of social phobic patients. Birbaumer et al. (1998) used functional magnetic resonance imaging (fMRI) to determine the activation of the amygdala
10 during exposure to slides of neutral faces as well as aversive odor stimuli. The sample consisted of seven social phobic persons and five healthy controls. Consistent with the hypothesis, the amygdala was selectively activated in the social phobic participants during the exposure to faces. Tillfors et al. (2001) applied positron emission tomography (PET) to measure regional cerebral blood flow (rCBF). A comparison of a social phobia group to a nonphobic comparison group during a public speaking task revealed not only increased self-reported anxiety, but also and heightened rCBF in the amygdala among the social phobic participants. In a follow-up study, Furmark et al. (2002) demonstrated that the improvement in symptomatology resulting from treatment was associated with attenuation of activity in the amygdala. The authors concluded that their results were consistent with theoretical outlines implicating increased activity of the amygdala during fearful responding. The conclusion that amygdala dysfunction is implicated in the etiology of social phobia is premature. The greater activation may simply be a consequence of anxiety. A comparison between a social phobia group and a control group that displays equivalent levels of anxiety but not social phobia is needed to further test this hypothesis. In summary, social phobia has a modest genetic component and tends to run in families. Research studies have illuminated several potential biological risk factors for the disorder. However, the cross-sectional nature of these studies precludes causal inferences regarding the role these neurobiological factors play in the development of social phobia. Longitudinal studies are needed to address the etiologic significance of these research findings.
11 1.2.2 Learning Theory
The learning theory as a model for the etiology of fear and avoidance has its origins in Pavlov’s empirical work (Pavlov, 1927). Pavlov started his famous work on classical conditioning studying the indigestion systems of dogs. He found that a novel stimulus (the sound of a bell) could elicit a reflex (saliva) after repeated pairing with a natural eliciting stimulus (food). He used the term unconditioned response (UR) to refer to the reflex response. Similarly, the stimulus naturally eliciting the UR was referred to as the unconditioned stimulus (US). The novel stimulus was termed the conditioned stimulus (CS) as it came to yield the reflex response after repeated pairing with the US. Likewise, the response that was elicited by the CS was referred to as the conditioned response (CR). Pavlov continued his work on classical conditioning examining what then was referred to as neurosis (Pavlov, 1927). He demonstrated that dogs could discriminate between an ellipse and a circle, as they received food along with the presentation of a circle, whereas a shock was administered with the presentation of an ellipse. However, when the administration of stimuli was changed to where it became more difficult to distinguish the ellipse from the circle, the dogs began to show agitation. These findings demonstrated that confusion or unpredictability was conditioned to elicit anxiety. The first study that demonstrated that fear could be acquired through classical conditioning in humans was conducted by Watson and Rayner (1920). In this study, also referred to as the “Little Albert” experiment, Watson and Rayner
12 conditioned an 11-month-old boy named Albert to fear a white rat through repeated pairing with a loud noise. In addition, Albert’s conditioned fear generalized, as he started displaying fearful responses when presented with other white furred objects and small animals. This outcome of this study suggested that anxiety can be learned through pairing of a noxious stimulus with an innocuous stimulus. In his two-factor theory, Mowrer (1939) outlined the two processes involved in the acquisition of avoidance behavior. First, fear is elicited through classical conditioning. Thus, a previously innocuous stimulus is conditioned (CS) to elicit a fear response (CR). Second, this link between the CS and the CR implies that termination of the CS is associated with fear reduction, a negatively reinforcing response. This negative reinforcing quality of fear reduction drives the termination of the CS. Thus, Mower states that avoidance is a learned response, acquired to terminate the CS and the associated fear response. Based on the works by Pavlov (1927) and Mowrer (1939) and the findings of subsequent examinations of their theories, Rachman (1977) put forth a modified learning theory of fear acquisition. There were three major additions. First, Rachman proposed that the strength and the likelihood of the fear would be directly proportional to a) the number of repetitions of the association between the stimuli and the fear response, and b) the intensity of the fear response in the presence of the stimuli. Second, he suggested that fear reactions are more likely to be elicited in situations of excessive confinement. Third, in addition to direct conditioning, Rachman proposed that fear could also be acquired vicariously (i.e.,
13 modeling), or though transmission of information. Evidence consistent with this theory has come from a number of studies (King, Clowes-Hollins, & Ollendick, 1997; Merckelbach, de Ruiter, van den Hout, & Hoekstra, 1989). Questionnaire studies provide some evidence suggesting that traumatic conditioning may be implicated in social phobia. Ost and Hugdahl (1981) found that 58% of social phobic sufferers attributed the onset of social phobia to a traumatic conditioning experience. Only 13% of the persons with social phobia indicated vicarious learning experiences as a cause of their phobia. Hofmann, Ehlers, and Roth (1995) asked diagnosed speech phobic persons about past public speaking experiences. They found that 89% of the sample reported traumatic experiences in the past. However, none of these speech phobics reported that this traumatic event preceded the onset of their phobia. In addition, 57% reported vicarious conditioning events, but none indicated these experiences to have a role in the acquisition of their fear. Thus, while direct conditioning may be implicated in social phobia, the data suggests that it is unusual for social phobic fears to be acquired vicariously. Work by Ohman and colleagues has added on to the existing conditioning theories for social anxiety. Building on the concept of “preparedness”, which was introduced by Seligman (1971), Ohman and colleagues conducted a series of experiments examining whether social fears may be driven by a predisposition to acquire fears of angry, critical or rejecting faces. In their first experiment, they found that conditioning with angry faces as CSs was superior to conditioning with happy faces as CSs (Ohman & Dimberg, 1978). The authors followed up this
14 study with an examination of the hypothesis that prepared conditioning would only occur when the gaze would be directed toward the subject. Consistent with their hypothesis, results of two experiments revealed superior conditioning, as indexed by enhanced resistance to extinction, when the CSs was a angry face directed toward the subject relative to when the angry face was directed away from the subject (Dimberg & Ohman, 1983). This led the authors to conclude that people have an evolutionary predisposition to fear someone with an angry face, but only when the gaze is directed at them. As Barlow (1988) pointed out, these findings seem to be particularly relevant to the avoidance of direct eye contact, behavior frequently observed in social phobia. Taken together, empirical data is consistent with the learning model of fear acquisition. Laboratory experiments illuminate the potential of an evolutionary based predisposition to drive the fear and avoidance of rejection. Moreover, questionnaire data suggests the importance of direct learning experiences in the origins of social phobia. It should be noted however that the method of recall is problematic when making inferences regarding cause and effect.
1.2.3 Cognitive Theory
The cognitive approach to anxiety disorders assumes that anxiety is a response to the perception of threat (i.e., Beck, Emery & Greenberg., 1985). Based on this general assumption and empirical evidence, Rapee and Heimberg (1997) recently proposed a cognitive model for the maintenance of social phobia. It should be noted that this model is an extension of earlier models and has
15 considerable overlap with the cognitive model developed by Clark and Wells (1995). Rapee and Heimberg’s model is presented in Figure 1. The authors propose that immediately following the entrance or anticipation of a social situation, a social phobic individual forms a mental representation of his/her appearance from an audience perspective. In other words, the mental representation is based on how the individual thinks the audience views his/her performance. Recently, Wells, Clark, and Ahmad (1998) confirmed that social phobics show this tendency to recall social situations from an observers-perspective. When asked to recall images of social situations, social phobics, relative to controls, were significantly more likely to view themselves from an observer’s perspective. Both groups generally took a field perspective for non-social situations. Empirical evidence further suggests that the observer’s image taken by social phobics is distorted. First, comparisons between observer ratings and self- ratings have demonstrated that socially anxious people significantly underrate their performance (Rapee & Lim, 1992). Second, as Rapee and Heimberg point out, socially anxious people have shown to overpredict the degree to which people notice their symptoms of anxiety (Alden & Wallace, 1995). Based on findings of a large body of research examining attention and memory processes in anxiety disorders, Rapee and Heimberg (1997) propose that the socially anxious individual will direct his/her attention to the threat, in this case, the mental representation of their external appearance. This assumption as it relates to social phobia has received support from several lines of research. Rapee
16 and Heimberg refer to numerous Stroop-paradigm studies (e.g., McNeil et al., 1995) that have demonstrated that socially phobic individuals take longer to recall words related to negative evaluation compared to neutral words, which suggests a tendency to allocate attentional resources to threatening material. In addition, several studies have found high correlations between social anxiety and the time spent focusing on the self (Melchior & Cheek, 1990). It has also been demonstrated that social phobic individuals focus their attention towards themselves when confronted with a fearful social situation (Beidel, Turner, & Dancu, 1985; Stopa & Clark, 1993). Rapee and Heimberg (1997) further propose that the person’s perceptions of the likelihood and the perceived negative consequences of negative evaluation are a direct function of the discrepancy between the individual’s mental representation and the individual’s prediction of the standards audience hold for him/her. Thus, a formal situation where an audience holds higher standards for the performer is more likely to increase the performer’s perceived probability and cost of negative evaluation than an informal situation. Support for this assumption comes from studies that have demonstrated the link between the degree of self- reported anxiety and perceived importance of the audience, as measured by size and status (e.g., Turner, Beidel, & Larkin, 1986). The degree to which negative evaluation is perceived as likely along with the extent to which negative evaluation is considered to be harmful is assumed to determine the level of anxiety. Anxiety is expressed physiologically (i.e., increased heart rate, sweating, blushing), cognitively (i.e., more negative
17 thoughts), and behaviorally (i.e., avoidance). Rapee and Heimberg further point out that behavioral avoidance can be subtle. Examples of these subtle avoidance strategies, also referred to as safety-seeking behaviors, include avoiding eye contact and reducing voice volume. The model that Heimberg and Rapee (1997) present is cyclical, namely the experience of anxiety ultimately leads the socially anxious person to strengthen his/her distorted mental representation of his/her performance. More specifically, when a social phobic person experiences increased somatic arousal (e.g. blushing), he/she will predict that this is noticed by the audience, and therefore will expect that the audience will view his/her performance as inferior. Empirical evidence is consistent with this cognitive approach to social phobia. However, it should be noted that this model presents only limited insight as to the causes of social phobia. More specifically, Rapee and Heimberg (1997) base their model on the notion that social phobic persons a) assume that other people are inherently critical and b) attach importance to being positively appraised by others (Rapee & Heimberg, 1997, p.742). Therefore, this model provides a perspective on how social phobia can be produced while acknowledging the interaction of various other factors including those biological in nature.
18
Figure 1. A model of the generation and maintenance of anxiety in social/evaluative situations. Source: Rapee & Heimberg (1997)
1.3 Treatment Approaches for Social Phobia
1.3.1 Pharmacological Approaches
As in the case for other anxiety disorders, several pharmacological agents have established clinical efficacy in the treatment of social phobia. Phenelzine, an irreversible Monoamine Oxidase Inhibitor (MAOI) was the first medication to demonstrate clinical efficacy for social phobia. More recently, double-blind 19 placebo controlled clinical trials have demonstrated the efficacy of several other agents such as selective serotonin inhibitors (SSRIs) including paroxetine, fluvoxamine and setraline, and high potency benzodiazepines (BZDs) such as clonazepam, bromazepam, and alprazolam.
1.3.2 Psychosocial Approaches
1.3.2.1 Behavioral Treatments
Exposure, applied relaxation, and social skills training are the most researched behavioral interventions. Exposure aims to “unlearn” anxiety using the principles of conditioning. The exposure technique consists of having the patient repeatedly confront the feared situation. Typically, exposure is delivered in a graduated fashion, where the patient and therapist first develop a hierarchy. Patients start with a situation that produces mild fear, and are instructed to repeatedly enter this situation until the fear has diminished. The patient will then approach the next situation on the hierarchy. Aside from using real life feared situations, exposure can also be delivered through imagery (in vitro exposure), virtual reality or via role play. Where patients are instructed to remain focused on the feared situation during exposure treatment, applied relaxation (AR) training (Ost, 1987) teaches the patient to decrease the level of physiological arousal while being exposed to the feared situation. Thus, AR is a coping technique designed to facilitate relaxation while engaging in anxiety-provoking situations. AR consists of several steps. Patients first learn the difference between the sensation associated with anxiety and the sensations associated with relaxation. They are then taught a
20 relaxation technique through muscle tension exercises. These exercises involve a series of trials where patients first increase and then quickly release the tension of particular muscle groups. Patients are instructed to focus on the sensations associated with relaxation. Finally, using exposure exercises, patients are taught to relax while approaching a feared situation. A third behaviorally-oriented intervention is social skills training (SST). Similar to AR, SST aims to provide the patient with a skill that can be applied when confronted with the feared situation. SST is designed to teach patients skills required for effective social interactions (Turner, Beidel, Cooley, Woody, & Messer, 1994). Using techniques such as modeling, rehearsal, corrective feedback, and homework assignments, SST focuses on enhancing patients’ interpersonal skills such as conversation initiation as well as performance related skills such as formal presentations. SST is often offered along with exposure (Turner et al., 1994).
1.3.2.2 Cognitive-Behavioral Treatments
The underlying assumption of cognitive-behavioral treatment is that anxiety disorders are maintained by faulty threat appraisals. Cognitive theories of social phobia implicate the tendency to overpredict both the likelihood and the negative valence of negative evaluation. In addition to behavioral techniques such as exposure, contemporary cognitive behavioral treatments for social phobia include specific cognitive techniques to alter these maladaptive beliefs (Heimberg, Juster, Hope, & Mattia, 1995).
21 In cognitive restructuring, patients are first taught to identify their specific threat appraisals, including both estimates of the likelihood and the negative valence associated with the threat. Next, patients are taught to evaluate the accuracy of these beliefs using evidence provided by Socratic questioning, past experiences, and exposure exercises. Lastly, patients are taught to replace their beliefs based on the evidence illuminated in the previous step. As Heimberg (2001) pointed out, the focus of exposure exercises during cognitive restructuring is different from traditional exposure techniques in that they are designed to provide evidence that is inconsistent with threat forecasts. In order to facilitate this process, patients are encouraged to fade maladaptive defensive behaviors such as rehearsing sentences in mind before giving a speech, and walking close to walls to manage the fear of tripping. It is thought that fading these safety behaviors enhance their threat disconfirmation process as the individual discontinues to erroneously attribute safety (failure to be harmed) to their safety behavior (Salkovskis, 1991).
1.3.2.3 Other Psychosocial Treatments
Interpersonal psychotherapy (IPT) for social phobia has received some attention in the literature, albeit less than behavioral and cognitive approaches. IPT is based on the assumption that interpersonal problems play a role in the etiology and maintenance of psychiatric disorders. IPT has demonstrated efficacy in the treatment of depression (Elkin et al., 1989) and bulimia nervosa (Fairburn et al., 1991). Recently, Lipsitz, Markowitz, Cherry, and Fyer (1999) developed an 16-session IPT manual for social phobia. As role transition appears as the most
22 salient interpersonal problem in social phobic patients, the treatment aims to address this interpersonal problem area using techniques such as emotional expression and role play. Preliminary findings are encouraging (Lipsitz et al., 1999). The psychodynamic perspective on social phobia stresses the importance of shame experiences, which result from the underlying unconscious wish of to be the center of attention and to receive supportive responses from others. Along with the sense of shame, social phobic persons are thought to have guilt feelings and separation anxiety. The treatment focus is to reveal these unconscious factors (Gabbard, 1992). The psychodynamic approach to the treatment of social phobia has not received any empirical support.
1.4 Treatment Outcome
Since 1981, approximately 20 randomized controlled clinical trials have been published investigating the efficacy of pharmacotherapy in the treatment of social phobia, and an additional 17 controlled studies have examined the efficacy of CBT for social phobia. Moreover, a review of the literature revealed 16 controlled trials that have investigated the efficacy of behavioral treatment and 7 controlled studies that examined cognitive restructuring. Only 3 controlled studies have investigated combined medication plus psychological treatment.
1.4.1 Short-Term Efficacy
In the last decade, the efficacy of CBT and pharmacological therapy for social phobia has been systematically reviewed in several meta-analytic studies (Fedoroff & Taylor, 2001; Gould, 1997; Taylor, 1996). A meta-analysis utilizes
23 either within-subject or controlled effect-sizes to evaluate the efficacy of various interventions. A within-subject effect size represents the degree of pre-to posttreatment change in standard deviations. It is derived by subtracting the mean of a social phobia composite (i.e., the average across all dependent measures) at posttreatment from the mean at pretreatment, and then dividing by the pooled standard deviation (pre- and posttreatment). Using within-subject effect sizes, Federoff and Taylor (2001) evaluated the efficacy of pharmacological and cognitive-behavioral treatments in a meta-analysis of 108 studies that included 11 different treatment conditions. As can be seen in Table 1, both modalities were effective in reducing the prominent features of the disorder as evidenced by the large mean pre- to posttreatment effect sizes. Controlled effect sizes provide a more stringent test of the relative efficacy of various interventions. Thus, the controlled effect size represents the degree to which a treatment produces more pre-to posttreatment change than a control condition. It is derived by subtracting the control group posttreatment mean of a social phobia composite (i.e., the average across all dependent measures) from the treatment group posttreatment mean of a social phobia composite, and then divided by the control group posttreatment standard deviation. Gould et al. (1997) conducted a meta-analysis of 24 controlled studies that reported on 40 separate treatment by control comparisons. The results are presented in Table 1. A comparison of the specific treatment modalities revealed
24 no significant differences between medication, CBT or their combination (effects sizes (ES), 0.74, 0.62, and 0.49 respectively). In addition, mean attrition rates did not differ among patients receiving medication (13.7%) from those receiving CBT (10.7%) or a combination treatment (6.7%). Among the cognitive-behavioral treatments, exposure-based interventions showed an advantage over those interventions using only cognitive-restructuring techniques. Contrary to expectations, there was no significant advantage of group- delivered CBT over individual CBT (ES, 0.88 and .44 respectively). Comparisons among pharmacological agents suggest superiority of SSRIs, as they yield strong outcomes (ES = 1.89) and, moreover, low dropout rates (1.5 %). These data suggest that in the short-term, cognitive-behavioral and pharmacological treatments have comparable clinical efficacy. However, since some of the studies in these meta-analyses suffer from significant methodological problems, the validity of meta-analytic results may be questionable. One specific methodological problem in clinical trials is the effect of allegiance. To address this problem, Heimberg et al. (1998) compared the efficacy of group-administered CBT (CBGT), and Phenelzine in a large sample (N=133) of carefully diagnosed social phobic patients. The study was conducted at two sites with different therapeutic expertise, and thus controlled for allegiance effects. Patients were randomly assigned to one of four treatment conditions: (a) CBGT, (b) Phenelzine; (c) Educational-supportive group therapy (ESGT); or (d) pill placebo. Treatment efficacy was evaluated at three different phases (12-weeks of acute treatment, and 6-month maintenance, and 6-month treatment-free follow-up). After acute
25 treatment, both active treatments showed marked improvements that were significantly greater than those observed for the pill placebo and ESGT condition. Controlled effect sizes for CBGT and Phenelzine over placebo were in the medium to high range (ES .44 and .71 respectively). Moreover, 77% of patients receiving Phenelzine, and 75% of patients who had undergone CBGT were classified as responders, compared to 35 % for ESGT and 41% for pill placebo.
Table 1. Short-term efficacy for treatments of social phobia. Source: *Gould et al., (1997) and ¥Fedoroff & Taylor (2001)
Condition Uncontrolled ES¥ Controlled ES* % Drop out* M N M N M N Medication BZDs 2.10 4 .72 2 12 2 SSRIs 1.70 12 1.89 2 2 2 MAOIs 1.08 15 .64 5 14 5 Psychosocial EXP 1.08 7 .89 9 13 9 AR .51 4 - - - - SST .64 7 .60 3 4 3 CR .72 7 .60 4 15 4 CR+EXP .84 21 .80 8 10 8 Combination Medication plus CBT - - .49 2 7 2 Control Wait-list .03 9 - - - - Attention Placebo .45 4 - - - - Pill Placebo .66 17 - - - -
1.4.2 Long-Term Efficacy
Since control groups are usually not assessed at follow-up, durability of treatment is typically determined using within-subject effect sizes. In order to evaluate the long-term treatment efficacy, Gould et al. (1997) calculated mean posttreatment to follow-up effects sizes of studies using a minimum follow-up period of 3 months. All except one of the 8 studies included in this analysis
26 employed CBT interventions. The mean effect size was 0.23, indicating further improvement after discontinuation of treatment. The one study examining pharmacotherapy showed an effect size of 0.07, suggesting maintenance of gains over the follow-up period. No data regarding the long-term efficacy of combined treatments is yet available. Findings of the multi-site study (Liebowitz et al., 1999) underscored the long-term advantage of CBT over medication. Seventeen percent of CBT completers relapsed, whereas 50% of Phenelzine completers relapsed over the course 6 months following discontinuation of treatment.
1.4.3 Quality of Life
The question whether treatment effects go beyond improvements in indices of social anxiety has been addressed in only a few studies. Findings reported by Gelernter et al. (1991) indicated that the relative clinical improvement observed among patients treated with Phenelzine (compared to a placebo condition) was associated with improved clinician ratings on a measure of work and social disability. In contrast, Stein, Fyer, Davidson, Pollack, and Wiita (1999) reported clinical superiority of Fluvoxamine over placebo, but failed to find differential improvement in social life functioning, as measured by self-report. Safren and colleagues (Eng, Coles, Heimberg, & Safren, 2001; Safren et al., 1996-1997) found that CBGT for social anxiety leads to significant and durable improvements in quality of life, as measured by self-report ratings. However, global quality of life ratings following treatment were not comparable to those of healthy control persons. Based on these findings, the authors suggested
27 that additional interventions aiming to improve life satisfaction should be incorporated in current treatment approaches for social phobia.
1.4.4 Cost Effectiveness
In addition to treatment efficacy and tolerability, treatment costs should be considered in the overall evaluation of the utility of a treatment. In their meta- analyses of interventions for social phobia, Gould et al. (1997) compared the expenses for CBT to those for pharmacological treatment. Their findings indicated that group-administered CBT was the most cost-effective intervention. The total cost over a 12-month period was approximately $600 for group CBT, compared to a total yearly cost of approximately $1000 for Clonazepam, $1200 for individual CBT and Phenelzine, and $2900 for Fluvoxamine.
1.4.5 Predictors of Clinical Response
Incidence of treatment non-response and relapse suggests that there are factors that moderate the efficacy of existing interventions. Safren, Heimberg, and Juster, (1997) examined the effect of treatment expectancies on treatment outcome for CBGT. Participants completed a reaction to treatment questionnaire tapping treatment credibility and confidence in improvement during the first and fourth treatment session. The results revealed that expectancies significantly predicted improvement, even after controlling for pretreatment social phobia severity. Since homework serves an integral part of cognitive-behavioral treatment, it has been hypothesized that compliance to homework would affect the beneficial effects of treatment. Consistent with predictions, homework compliance has
28 shown to be associated with greater improvement (Edelman & Chambless, 1995; Leung & Heimberg, 1996). Several studies have demonstrated an association between social phobia subtype and treatment outcome (Brown et al., 1995; Hope, Heimberg, & Bruch, 1995; van Velzen, Emmelkamp, & Scholing, 1997). While the generalized subtype displays greater impairment at both pre- and posttreatment, the degree of improvement is similar to the rate of improvement among the nongeneralized subtype. A similar pattern of results has been observed with regards to the effects of Axis-I comorbidity such as depression (Turner, Beidel, Wolff, Spaulding, & Jacob, 1996; van Velzen et al., 1997), generalized anxiety disorder (Mennin, Heimberg, & Jack, 2000), and specific phobias (Turner et al., 1996) and Axis-II comorbidity such as avoidant personality disorder (Brown et al., 1995; Hope et al., 1995; van Velzen et al., 1997). Thus while associated with greater severity, the co-occurrence of other psychiatric disorders does not affect the beneficial effects of treatment. In summary, two decades of research have demonstrated that both pharmacotherapy and CBT are markedly effective in the short-term, and cognitive-behavioral treatments seem to have an advantage with respect to durability of treatment gains. Traditional moderators such as treatment alliance also operate in the treatment for social phobia, but contrary to predictions, social phobia subtype and comorbidy do not negatively influence the benefits of treatment. There is an absence of data on the effects of combining psychosocial
29 and medication treatments. Finally, the question whether these efficacy data generalize to the real world of clinical practice remains to be addressed.
1.5 Mechanisms of Fear Reduction
1.5.1 Habituation and Extinction
One behavioral explanation for fear reduction is habituation. Habituation refers to a decrease in a response to a stimulus. Habituation occurs when the stimulus is presented in a repeated fashion (Pavlov, 1927). For example, loud noises initially elicit startle responses. However, the startle response diminishes after extended duration of a loud noise. It has been demonstrated that particularly physiological responses are susceptible to habituation. In the context of fear reduction, habituation is thought to explain the decline in fear responses, specifically physiological indices of fear, following repetitive exposure to fear- provoking situations. Although seemingly suitable, habituation falls short as a theory of fear reduction. First, habituation is a short-term effect, as the sensitivity to a stimulus is recovered only after a short rest period (Thompson & Spencer, 1966). Fear reduction, however, has demonstrated to be durable. Second, the presentation of stimuli that provoke a high intensity physiological response are resistant to habituation. In fact, learning theories predict sensitization in these situations. Inconsistent with this condition is evidence from treatment outcome studies suggesting that prolonged exposure to intense fear-provoking situations (i.e. flooding) can be effective in reducing anxiety (e.g., Turner, Beidel, & Jacob, 1994).
30 Another behavioral explanation for fear reduction is extinction. Similar to habituation, extinction refers to the decrement in responding to a stimulus. There are several differences between extinction and habituation (Rachman, 1990). Extinction pertains to a decrease in learned responses, whereas habituation refers to a decrease in unlearned responses. Moreover, while habituation occurs following repeated presentation of the relevant stimulus, extinction occurs through repetitive presentation of the relevant stimulus in the absence of reinforcement (Pavlov, 1927). Thus, in the context of fear reduction, extinction results from repeated exposure to feared situations without experiencing the feared consequences (i.e., negative evaluation in social phobia). Extinction is thereby a better explanation for the effects of flooding, in which the fearful stimulus is presented without the aversive consequences. Extinction has mostly been applied to behavioral indices of anxiety (i.e., avoidance) (Barlow, 1988). As Barlow (1988) points out, extinction implies a learning experience. Patients undergoing repeated exposure are provided with evidence that the feared consequences do not occur, and therefore reduce avoidance behavior. The experience of learning might explain why extinction is more permanent than habituation, which only requires repeated exposure to the feared stimulus (with or without aversive consequences) (Barlow, 1988). Both habituation and extinction play part in contemporary theories of fear reduction (Barlow, 1988; Rachman, 1990). Changes in physiology during exposure are consistent with habituation, whereas changes in behavioral indices of anxiety during exposure to disconfirmatory evidence match extinction
31 processes. The fact that repeated exposure does not always produces fear reduction undermines the applicability of extinction and habituation as the mechanism underlying fear reduction. Similarly, learning theories fall short in explaining fear reduction resulting from techniques other than exposure (i.e., cognitive restructuring).
1.5.2 Cognitive Reappraisal
According to cognitive theory, pathological anxiety stems from irrational beliefs regarding danger (Beck et al., 1985). More specifically, cognitive models assume that pathological anxiety is driven by overestimations of the degree or likelihood of threat (Beck & Clark, 1997). Accordingly, therapeutic effects are assumed to be mediated by the modification of threat expectancies. Following this cognitive approach, the improvement of social phobic symptomatology is expected to be mediated by a change in the mental representation of the self and a change in the person’s beliefs that performing poorly in a social situation will have catastrophic consequences (Rapee & Heimberg, 1997). Evidence from treatment outcome studies provides some support for this mediation hypothesis. For example, Woody, Chambless, and Glass (1997) found that social phobic patients decreased their self-focused attention over the course of cognitive-behavioral treatment. Similarly, Hofmann (2000) observed a positive correlation between the decrease in negative self- focused thoughts and pre-post difference scores in a self-report measure tapping social phobia symptoms. In another study, Wells and Papageorgiou (1998) demonstrated that exposure therapy where patients focus on the external
32 environment is more effective than exposure therapy where patients are just to focus on the feared situation. They found that these increased effects were associated with change from a self-focused perspective to a field perspective. Cognitive reappraisal is a mediation hypothesis highly regarded in accounts for both depression and anxiety reduction (Beck et al., 1985; DeRubeis et al., 1990). Pre-posttreatment studies show that a change in threat appraisals is associated with improvement. However, this particular design precludes causal inferences. Therefore it can be argued that instead of a mediator, threat reappraisal may simply be a consequence of anxiety reduction.
1.5.3 Emotional Processing
In line with the assumption that fear reduction occurs through the modification of erroneous ideas of threat is the emotional processing theory (Foa & Kozak, 1986). Foa and Kozak refined the emotional processing perspective, which was first introduced by Rachman (1980). The theory is based on Lang’s “bio-informational conceptualization of fear” (Lang, 1979), which proposes that fear is represented in a network that includes information about (a) the feared stimulus (b) verbal, physiological and overt behavioral responses, and (c) interpretation about the meaning of the stimulus and response elements of the structure. Consistent with Rapee and Heimberg’s (1997) model, Foa and Kozak (1986) suggest that fear stimuli or responses are associated with unrealistically high probability of harm, and that this harm has a high negative valence. Exposure is assumed to produce fear reduction through a modification of the existing fear structure. There are two processes that occur. First, exposure
33 activates the fear structure. Second, exposure provides the information that is incompatible with the elements that exist in the fear structure, so that a new memory can be formed. Foa and Kozak (1986) further propose that the long-term fear reduction necessitates short-term (within-session) habituation and long-term (across-session) habituation. Short-term habituation results in a decoupling of the stimulus and response propositions associated with threat valence (i.e., if the unpleasantness of being criticized passes quickly, then it is no so bad to be criticized). Thus, during repeated exposures, mild criticism ceases to induce physiological arousal, which guides the patient to decrease the negative consequences of being criticized. Exposure to incompatible information is thought to enhance this change in threat valence estimations. On the other hand, repeated exposures across sessions are assumed to weaken the inaccurate probabilities of harm (i.e., the fact that negative evaluation has failed to occur every time the feared situations was entered means that it is unlikely that it will occur in the future). Several studies have investigated the validity the emotional processing theory. In general, changes derived from exposure-based treatment can be well explained in terms of emotional processing. For example, Telch, Valentiner, Ilai, Petruzzi, and Hehmsoth, (2000) demonstrated that increasing the salience of disconfirmatory evidence was associated with greater beneficial effects of exposure. Similarly, safety-seeking behaviors which are thought to interfere with disconfirmatory processes have shown to impede the effects of treatment (Sloan & Telch, 2002). Data pertinent for social phobia comes from a study by Foa,
34 Franklin, Perry, and Herbert (1996). They found that 58% of the improvement following CBT for social phobia was accounted for by a change in cost estimates. The authors concluded that successful treatment for social phobia warrants targeting the overestimation of the disastrous consequences of negative evaluation.
1.5.4 Self-Efficacy
Bandura (Bandura, 1977; Bandura, Adams, & Beyer, 1977) proposes that the perception of increased mastery over a feared situation (i.e, self-efficacy) mediates fear reduction. According to Bandura, a situation is fear-provoking when people believe that they cannot control threats in a situation. The self-efficacy model proposes that fear reduction is the consequence of an increase in the perceived ability to control potential threats. Four sources of information are assumed to influence the degree of self-efficacy: a) enactive experiences; b) vicarious experiences; c) verbal persuasion; and d) physiological information. Several studies have provided evidence for the potential mediating effects of self-efficacy. Guided mastery treatment, which combines in vivo exposure with techniques designed to enhance mastery (e.g., gradual reduction in therapist involvement) has demonstrated to produce greater fear reduction in the treatment of specific phobias and agoraphobia compared to exposure alone and no treatment control conditions (Williams, Dooseman, & Kleinfield, 1984; Williams, Kinney, & Falbo, 1989; Williams, Turner, & Peer, 1985). Moreover, perceived self- efficacy was the most accurate predictor of pre- to posttreatment behavioral changes.
35 Although no studies have yet examined self-efficacy as a potential mediator for treatment change in social phobia, Hofmann (2000) provides some indirect evidence that is consistent with this hypothesis. He refers to a clinical trial for public speaking anxiety (Newman, Hofmann, Trabert, Roth, & et al., 1994), which showed that exposure therapy led to an increase in performance self- ratings, but observer-ratings remained unchanged. The current literature emphasizes the importance of both behavioral and cognitive variables in fear reduction. Studies that carefully examine processes during treatment are needed to address the significance of these models in explaining the mechanism of action of treatments for social phobia.
36 CHAPTER 2: THE PRESENT STUDY
2.1 Facilitating Fear Reduction
Social phobia is the most prevalent of all anxiety disorders (Kessler et al., 1994) and is associated with significant impairment in work and social functioning. Moreover, social phobia sufferers often develop additional psychological sequalae such as substance abuse and depression. Numerous clinical trials have demonstrated the efficacy of Cognitive Behavioral Treatment (CBT) in treating social phobia (cf. Gould et al, 1997). However, approximately 25% of social anxiety patients do not benefit (Heimberg et al., 1998) and of those who do benefit, most continue to display residual symptoms and impaired work/academic impairment. Consequently, there is a considerable need for developing more potent psychological treatments for social phobia. The emotional processing theory (Foa et al., 1986) is well accepted and often considered as a model of reference in the investigation of mechanisms governing the reduction of pathological fear. Foa and Kozak (1986) suggest that two conditions are necessary for emotional processing to occur. First, fear relevant information must be made available in a manner that will activate the fear structure. Second, the information made available must contain elements that are incompatible with the elements that exist in the fear structure, so that a new memory can be formed. Accordingly, fear activation in addition to fear reduction during exposure sessions (i.e., within-session habituation) and fear reduction across exposure sessions (i.e., between-session habituation) are indicators of
37 emotional processing of fear. Foa and Kozak (1986) propose that the durability of emotional processing requires alteration of threat perceptions connected to the fear stimulus. Preliminary data suggest that the provision of videotape feedback following a speech may enhance speech anxiety reduction. Rapee & Hayman, (1996) compared performance ratings of participants who received videotape feedback following a speech to performance ratings of participants who did not receive videotape feedback. The results showed a greater discrepancy between self-ratings and ratings made by independent observers among those who did not receive feedback, compared to participants who watched their speech on tape. The findings are consistent with the idea that feedback provides disconfirmatory evidence, and thereby corrects the distorted perceptions. In a follow-up study, Harvey, Clark, Ehlers, and Rapee (2000) examined whether cognitive preparation would increase the beneficial effects of videotape feedback. High and low socially anxious individuals performed a speech and received videotape feedback of their performance. Half of the participants were instructed to contrast their imagined performance to their actual performance, which was provided by the videotape feedback. The other half of the sample did not receive any specific instructions. Participants who received cognitive preparation rated their performance higher compared to those who were not given cognitive preparation. These results suggest that the beneficial effects of video feedback may be increased when the client carefully examines the discrepancy between imagined performance and actual performance.
38 The current study examined whether exposure-based treatment for public speaking anxiety, the most common feared situation reported among social phobic sufferers (Schneier et al., 1992), can be enhanced through the addition of videotape feedback procedures. In addition to investigating the effects of videotape feedback where the participant receives information regarding his/her performance, the current study examined the effects of feedback of external stimuli through videotape feedback where the camera was focused on the audience. This form of feedback was hypothesized to address the overestimation of the cost associated with appearing anxious in front of others, as the participant is provided with information suggesting that exhibiting anxiety is not associated with negative consequences. In addition, instructing the social phobic patients to focus on external stimuli rather than on the self has been shown to increase the effects of exposure (Wells & Papergeorgiou, 1998). In contrast, it was hypothesized that the information made available through performance videotape feedback would address the overestimation of the likelihood that the person will exhibit anxiety symptoms. This change in speech-related appraisals is thought to enhance participants’ performance ratings, and thereby decrease public speaking anxiety. The study design was a 4-arm randomized clinical trial. Participants who meet DSM-IV criteria for Social Phobia were randomly assigned to one of four treatment conditions: (a) Exposure to public speaking plus Videotape Feedback focusing on the participants’ performance (EVS); (b) Exposure to public speaking plus Videotape Feedback focusing on the audience reaction (EVA); (c) Exposure
39 to public speaking without feedback (EO); or (d) Credible Placebo Treatment consisting of providing pulsed audio and photic stimulations using a Digital Audio Visual Integration Device (DAVID). This device flashes lights through goggles and clicks like a metronome through headphones at 12Hz. Participants assigned to each of the three exposure conditions received three 15-min. exposure sessions. Participants randomized to the placebo condition received three 15-min. treatment sessions with the D.A.V.I.D. device but did not receive any direct exposure to public speaking. Outcome assessments were conducted prior to treatment, one week posttreatment, and at 1-month follow-up.
2.2 Hypotheses
2.2.1 Outcome Hypotheses
It was expected that providing videotape feedback would be associated with greater fear reduction. More specifically: 1. Participants in the exposure conditions (i.e., EVS, EVA, and EO) were expected to display greater decline in fear than participants in the DAVID condition; 2. Participants in the videotape feedback conditions (i.e., EVS and EVA) were expected to show greater response to treatment compared to those who received exposure without videotape feedback (i.e., EO). 3. Participants who received videotape feedback, where the camera was focused on the audience (EVA) were expected to show greater fear
40 reduction at posttreatment compared to participants who received videotape feedback, where the camera was focused on the speaker (EVS).
2.2.2 Process Hypotheses
It was expected that fear reduction was achieved through a change in threat appraisals, resulting from disconfirmation. More specifically: 1. It was expected that reductions in threat appraisals specific to public speaking would mediate the increased improvement in public speaking anxiety observed in the exposure conditions relative to the placebo condition. 2. It was expected that changes in estimates of the negative consequences of an anxious appearance would mediate the increased improvement in public speaking anxiety observed in the EVA condition. 3. It was expected that the increased improvement in self-ratings of performance would mediate reductions in estimates of the likelihood to appear anxious observed in the EVS condition 4. It was hypothesized that reductions in estimates of the likelihood to appear anxious would mediate the increased improvement in public speaking anxiety observed in the EVS condition.
41 CHAPTER 3: METHOD
3.1 Participants
Study participants were recruited from the pool of introductory psychology students at the University of Texas at Austin as well as from the Austin community. All patients met the following entry criteria: (a) principal Axis-I diagnosis of social phobia as determined by Composite International Diagnostic Interview (CIDI-Auto; World Health Organization, 1997); (b) significant fear of public speaking as determined by peak fear ratings during an impromptu speech task; (c) fluent in English (both written and verbal); and (d) negative for current psychosis, bipolar disorder, or actively suicidal or past history of seizures (this exclusion criterion was added because there is a slight increase in seizure risk upon exposure to the D.A.V.I.D. device among individuals with a history of seizures); and (e) no recent change in psychotropic medications.
3.2 Experimental Design
Participants were randomly assigned to one of four treatment conditions:, (a) Exposure to public speaking plus Videotape Feedback focusing on the participants’ performance (EVS); (b) Exposure to public speaking plus Videotape Feedback focusing on the audience reaction (EVA); (c) Exposure to public speaking without feedback (EO); or (d) Credible Placebo Treatment consisting of providing pulsed audio and photic stimulations using a Digital Audio Visual Integration Device (DAVID). Participants assigned to each of the three exposure conditions received three 15-min. exposure sessions. Participants randomized to the placebo condition
42 received three 15-min. treatment sessions with the D.A.V.I.D. device but did not receive any direct exposure to public speaking. Outcome assessments were conducted prior to treatment, one week posttreatment, and at 1-month follow-up.
3.3 Assessment
3.3.1 Diagnostic Assessment
Assessment of DSM-IV diagnosis of social phobia was conducted using the anxiety module of the computerized version of the Composite International Diagnostic Interview (CIDI-auto; World Health Organization, 1997). The CIDI- auto has been widely used for the assessment of DSM-IV diagnoses. It has demonstrated good psychometric properties including good sensitivity (.91) and acceptable specificity (.62). Moreover, the agreement between the clinical standard diagnosis (i.e., LEAD standard diagnosis) and CIDI-Auto diagnosis for social phobia was acceptable (71%), and similar to the clinician administered version of the CIDI (Peters & Andrews, 1995).
3.3.2 Liebowitz Social Anxiety Scale
The Liebowitz Social Anxiety Scale (LSAS; Liebowitz, 1987) was originally developed as a clinician-administered interview to assess social phobia through the evaluation of fear and avoidance of social situations. The scale consists of 24 items, 13 items describing performance situations (i.e., writing while being observed - for example, signing a check in a bank), and 11 items describing social interaction situations (i.e., talking with people you don't know very well). Items are rated for fear (0 = no fear; 3 = severe fear) and avoidance (0 = never avoid (0%); 3 = usually avoid (66-100%)). In addition to a total score, the 43 LSAS yields eight subscales: 1) Social Interaction-Fear; 2) Social Interaction- Avoidance; 3) Public Speaking-Fear; 4) Public Speaking-Avoidance; 5) Observation by Others-Fear; 6) Observation by Others-Avoidance; 7) Eating and Drinking in Public-Fear; and 8) Eating and Drinking in Public-Avoidance (Safren et al., 1999). The LSAS has shown good internal consistency, test-retest reliability, and has demonstrated to be sensitive to treatment change (Heimberg et al., 1999). Examination of the self-report version of the LSAS (LSAS-SR) has yielded similar results. Baker, Heinrichs, Kim, & Hofmann (2002) found that the LSAS- SR correlated highly with the clinician-rated version (r=.85), showed adequate test-retest reliability (r=.83), and demonstrated adequate sensitivity to treatment change. Moreover, recent data showed that the LSAS-SR subscales have high internal consistency (αs ranged from .75 to .95; Oakman, Van Ameringen; Mancini; and Farvolden, 2003).
3.3.3 Speech Task
The speech task was administered at pretreatment, posttreatment, and at 1- month follow up. The objective was to measure threat expectancies and subjective fear while giving a speech. Participants were instructed to speak about a general topic for 3 minutes. The speech was completed in a conference room in front of a video camera and in the presence of four audience members (2 males and 2 females). The mean audience member age was 22.87 (SD=3.71), and the ethnic breakdown was similar to that of the study sample.
44 Participants were first asked to select a topic. Possible topics included nuclear power, the American health system, seatbelt laws, and corporal punishment in schools. These topics have been used in previous studies (e.g., Hofmann et al., 2000). Following topic selection, participants were given 5 minutes to prepare a general outline. Although they were allowed to take notes during preparation, notes were not allowed during the actual speech. Participants were then instructed to start their speech. The following measures were taken during the speech task.
3.3.3.1 Subjective Peak Fear
Immediately following the speech task, participants rated the following question on a 0 (no fear) to 100 (extreme fear) scale: “What was the highest level of fear you experienced during your three-minute speech.”
3.3.3.2 Threat Appraisals
Likelihood of Anxious Appearance. Prior to the speech task, participants rated the following question on a 0% (Not likely at all) to 100% (Extremely likely) scale: “Estimate the likelihood that the audience will notice your anxiety during the speech.” Cost of Anxious Appearance. Prior to the speech task, participants rated the following question on a 0% (Not bad at all) to 100% (Extremely bad) scale: “Estimate how bad it would be when the audience notices your anxiety during the speech.”
45 Self-Rated Performance. Immediately following the speech task, participants rated the following question on a 0% (Not good at all) to 100% (Extremely Well) scale: “Estimate your overall performance.”
3.3.4 High End-State Functioning
Following recommendations by Jacobson and Truax (1991), participants were classified as responders if they achieved a level of improvement that was statistically reliable. Commensurate with previous trials (Sloan & Telch, 2001;
Rentz, Powers, Smits, Cougle & Telch, 2003), the reliable change index (RCI; x2- x1/Sdiff) was computed for the subjective fear rating obtained during the speech task (Subjective Peak Fear). Those participants with RCIs greater than 1.96 on this outcome measure were defined as responders. Participants were classified as showing high end-state functioning (HEF) if: (a) they showed responder status; and (b) their level of functioning at posttreatment as measured by Subjective Peak Fear fell outside the range of the social phobic population, as defined by two standard deviations from the mean of that population. This latter criterion was selected due to the unavailability of normative data for a non-social phobic population (see Jacobson and Truax,
1991).
3.3.5 Relapse
Relapse was defined as a statistically reliable increase in fear from posttreatment to follow-up as defined by the RC index. Consistent with recommendations put forth by Jacobson and Truax (1986), responders with RCIs greater than 1.96 were defined as showing relapse.
46 3.4 Procedure
3.4.1 Screening
The screening consisted of two stages. During the Stage 1, potential student participants (N=6014) completed an online version of the Liebowitz Social Anxiety Scale (LSAS-SR; Liebowitz, 1987). Those scoring in the social phobic range and who reported significant fear and avoidance associated with public speaking (N=829) were invited via email for an individual assessment at the Laboratory for the Study of Anxiety Disorders facility (Stage 2). Of these 829 potential participants, 75 participants agreed to take part in Stage 2 of screening. At the start of Stage 2 of the screening, student and community participants (N=75) completed the CIDI-auto (World Health Organization, 1997) along with an assessment battery consisting of self-report questionnaires. Those who met DSM-IV criteria for social phobia (N=65) were administered the speech task. Individuals who reported only mild fear (i.e., less than 50) during the pretreatment speech were deemed insufficiently phobic and excluded from the study (N=2). Of 63 participants who were found eligible, 4 decided against participation, leaving 59 participants who were randomized.
47 Screening Phase I (N=6014)
Potentially Eligible Ineligible (N=829) (N=5185)
Screening Phase II Not Interested (N=75) (N=754)
Ineligible Randomized Not Interested* (N=12) (N=59) (N=4)
Post Drop-Out Assessed at Post (N=4) (N=55)
FU Drop-Out Assessed at FU (N=10) (N=45)
Figure 2. Flow chart of screening.
3.4.2 Treatment Conditions
3.4.2.1 Exposure Conditions
Eligible participants started treatment immediately following eligibility assessment. They received an exposure rationale emphasizing that fear of public speaking is often fueled by exaggerated beliefs about being negatively evaluated by others. The role of avoidance in the maintenance of pathological fear was explained along with instructions emphasizing the importance of repeated confrontations as a method for overcoming fears. To enhance credibility of the treatment, participants received videotaped instructions delivered by Michael J. Telch, Ph.D, clinical psychologist and director of the Laboratory for the Study of Anxiety Disorders. In order to assess perceived credibility of the treatment, all participants completed a reaction to treatment questionnaire (RTQ; Borkovec & Nau, 1972 ) following instructions. The RTQ is a four-item scale assessing (a) the
48 participants’ perceived credibility of the exposure treatment, and (b) the expected level of improvement from the treatment. Items are rated on a 0 (not at all) to 10 (extremely) Likert-type scale. Participants assigned to each of the three exposure conditions received a total of 45 min. of in vivo exposure to public speaking. The exposure was divided into five, 3-min. speech trials, across three treatment sessions. Treatment was completed in a one-week period. Instructions for the treatment speech trials were similar to those given during the baseline speech task. Participants were instructed to select a new topic from the list during sessions 1 and 2 , while during session 3 they were instructed to give a 3-minute speech on themselves. Participants were then allowed 10 minutes to prepare a general outline for their speech. Exposure-Videotape Feedback of Self (EVS). Along with the standardized exposure rationale, participants assigned to the EVS condition received specific instructions outlining the utility of videotape feedback, where the camera is focused on the speaker. More specifically, participants were told that this feedback was designed to help examine their performance more objectively and thus help lessen their concerns about the appearing weird or appearing overly uncomfortable. Following each speech trial, participants watched a tape demonstrating their performance during their speech. Commensurate with suggestions outlined by Harvey et al. (2000), participants were specifically instructed to pay attention to how they appeared rather than to how they felt during the speech.
49 Exposure-Videotape Feedback of Audience (EVA). Along with the standardized exposure rationale, participants assigned to the EVA condition received specific instructions outlining the utility of videotape feedback, where the camera is focused on the audience. Following each speech trial, participants watched a tape showing the reactions of the audience during their speech. Participants were specifically instructed to pay attention to how the audience was reacting to their speech rather than to how they felt during the speech. Exposure Only (EO). Participants in the EO condition did not receive videotape feedback. Following each speech trial, these participants watched a tape on a neutral topic.
3.4.2.2 Placebo Condition
Participants assigned to the placebo condition (DAVID) received a similar rationale emphasizing that fear of public speaking is often fueled by exaggerated beliefs about being negatively evaluated by others, along with instructions emphasizing the beneficial effects of relaxation. To enhance credibility of the treatment, participants received the following videotaped instructions delivered by Michael J. Telch, Ph.D, clinical psychologist and director of the Laboratory for the Study of Anxiety Disorders:
“An effective strategy for reducing fear is to induce heightened beta wave brain activity with a device called the Digital Audio Visual Integration Device or DAVID. Beta waves are high frequency, low amplitude brain waves seen while people are awake and relaxed immediately prior to the alpha wave activity of Stage 1
50 of sleep. The DAVID induces these brain waves by delivering pulsed audio and visual stimuli. These goggles will deliver flashing lights at 12Hz (cycles per second) and these headphones will deliver audible ticks (like a metranome) also at 12Hz (cycles per second) to induce the beta wave relaxation. Prior research has shown that the delivery of pulsed audio and visual stimuli is an effective strategy for enhancing beta wave activity associated with relaxation. The enhanced relaxation brought on by the enhanced beta wave activity will allow you to feel less anxious.” The Digital Audio Visual Integration Device (DAVID) developed by
Comptronic Devices (9876-A 33rd Ave., Edmonton, AB) is used by health care professionals as a relaxation device. It is a small soundboard about the size of a stereo receiver, which includes a headset and plastic mask. The headset emits controllable ticking sounds, similar to those made by a metronome. The plastic mask resembles ski goggles, and delivers pulsed orange lights at controllable rates. In this study, the audio and video stimulus frequency will be set at 12 Hz (cycles per second), which is the rate at which the device is suggested to maximally produce relaxation and meditative states. Participants assigned to the DAVID conditions received a total of 45 min. of exposure to the DAVID. The exposure was divided into five, 3-min. trials, across three treatment sessions. Treatment was completed in a one-week period, with spacing of sessions identical to the exposure conditions. Participants
51 completed ratings of self-efficacy and threat expectancy as if they were to give a speech prior to each trial.
3.5 Steps for Enhancing Treatment Integrity
In order to assure the greatest possible treatment integrity, assessments and treatments were manualized and administered by trained staff. Manualized Protocol. The protocol is a 60-page manual divided into separate sections for each session (pretreatment, treatment, posttreatment, and follow-up). The treatment section is further divided into separate sub-sections for each treatment condition (EVS, EVA, EO, and DAVID). Scripts are provided throughout the manual to be read aloud verbatim by the staff member. Staff Training. The training of staff involved a) didactic orientation to the project provided by the PI; b) observation of assessment and treatment procedures; and c) role-plays of procedures with trained experimenters. Experimenters were observed and monitored, and provided with feedback regarding adherence to the experiment protocol by the first author (JS). Experimenters were not allowed to administer assessments or treatments until they had demonstrated proficiency with the protocol.
52 CHAPTER 4: STATISTICAL ANALYSES
4.1 Baseline Differences
To examine whether the randomization procedure had resulted in equivalent groups, baseline scores on the outcome measures were subjected to a univariate analysis of variance (ANOVA). Differences in treatment credibility and expectancy, as measured by the RTQ, were also examined with an ANOVA.
4.2 Treatment Outcome
All outcome analyses were intent-to-treat. The intent-to-treat analyses were performed using the Last Observation Carried Forward1 method (Mazumdar, Liu, Houck, & Reynolds, 1999). To get an impression of the magnitude of improvement associated with each of the conditions, Cohen’s D uncontrolled effect sizes were calculated using the formula: uncontrolled effect size = (mean pretreatment score - mean posttreatment/follow-up score) ÷ pooled SD. Similarly, to get an idea about the magnitude of the differential treatment effects, I computed controlled effect sizes using the formula: Cohen’s D controlled effect size = (posttreatment/follow up covariance adjusted mean of condition X – posttreatment/follow-up covariance adjusted mean of condition Y) ÷ pooled SD. Effect sizes for differences between
1 I acknowledge that when applied to follow-up data, this method may seem optimistic as it assumes maintenance of treatment gains. However, results of analogue studies with both claustrophobic and spider phobic samples conducted in our laboratory show low rates of relapse, but mostly continual improvement during this brief follow-up period (e.g., Sloan & Telch, 2001; Smits, Randall, & Telch, 2002), suggesting that carrying baseline scores forward is overly conservative.
53 proportions were computed using probit transformations (see Table 5.5 in Glass, McGaw & Smith, 1981). Cohen (1988) proposes the following classification of effect sizes: small (0.20 – 0.49), medium (0.50- 0.79), and large (0.80 and above).
4.2.1 Effects at Posttreatment
Within-group effects were examined with repeated measures MANOVAs. For each group separately, pre- and posttreatment measures were entered as the DVs and time as the within-group factor. Separate analyses were performed for Public Speaking Anxiety (as measured by Subjective Peak Fear (i.e., situation- specific public speaking anxiety), LSAS-Public Speaking-Fear (i.e., generalized public speaking anxiety), and LSAS-Public Speaking-Avoidance (i.e., generalized public speaking avoidance), and General Social Anxiety (as measured by LSAS- Social Interaction-Fear, LSAS-Social Interaction-Avoidance, LSAS-Observation by Others-Fear, LSAS-Observation by Others-Avoidance, LSAS-Eating and Drinking in Public-Fear, LSAS-Eating and Drinking in Public-Avoidance). The differential treatment effects at posttreatment were tested using MANOVAs with residualized change scores of the outcome measures as the DVs and the group variable as the IV. Separate analyses were performed for Public Speaking Anxiety and General Social Anxiety. Differences in HEF rates were examined using chi-square analyses.
4.2.2 Effects at Follow-up
The durability of treatment effects was examined in similar fashion. Pretreatment and follow-up scores were entered as DVs in the within-group analyses, and residualized change scores (pretreatment to follow-up) served as the
54 DVs for the between-group analyses. Differences in relapse rates were examined using chi-square analyses.
4.3 Mediator Analyses
The hypothesis that the effects of treatment would be mediated by changes in threat expectancies were tested in accordance with the analytic steps outlined by Baron and Kenny (1986). Conditions to be met for mediation include (1) a significant effect of treatment on the mediator; (2) a significant effect of treatment on outcome; and (3) a significant correlation between the mediator and outcome. According to Baron and Kenny (1986), evidence for full mediation exists when the relationship between treatment and outcome is no longer significant after controlling for the effects of the mediator; whereas evidence for partial mediation exists when the relationship between treatment and outcome is significantly attenuated (but still significant) after controlling for mediator effects.
4.4 Moderator Analyses
To examine whether the differential treatment effect was moderated by pre-randomized individual factors such as gender, community status (Student vs. Community), or diagnostic status (GSP vs. NGSP), moderator analyses were conducted in accordance with the analytic steps outlined by Baron and Kenny (1986). In these analyses, sex, ethnicity and diagnostic status were entered as categorical independent variables along with the treatment variable. Moderator status was assigned to those factors that yielded significant interactions with the treatment variable.
55 CHAPTER 5: RESULTS
5.1 Characteristics of Participants
As can be seen in Table 2, most participants met criteria for generalized social phobia2 (GSP; 71%). The sample consisted primarily of students (73%). Fifty-four percent of the sample was female, and ages ranged from 18 to 51 (M = 21.73; SD = 6.23). The ethnic breakdown of the sample was 76% Caucasians, 10% Asian-American, 9% Mexican-American, 4% African-American, and 1% Indian-American.
Table 2. Demographic characteristics
Demographic DAVID EO EVA EVS Characteristic (N=14) (N=16) (N=14) (N=15)
Age (yrs) M 18.93 21.50 22.93 23.47 SD 1.07 6.04 6.01 8.48 Sex (%) Female 80 50 64 27 Ethnicity (%) African-American 7 0 7 0 Asian-American 7 13 7 20 Caucasian 65 81 86 73 Indian-American 0 6 0 0 Mexican American 21 0 0 7 Community Status (%) Student 92 63 71 67 Social Phobia Status (%) GSP 86 63 64 67
2 Clark et al. (in press) observed a pretreatment LSAS-SR mean of 75.01 (SD=23.29) in a recent clinical trial with GSP patients. Based on a traditional standard (M-1SD), one could argue that participants with a score below 51 (75-24) can be classified as Non-GSP. Mennin et al. (2002) recommended using a cut-off score of 60 for the LSAS clinician administered version. In order to minimize the number of false positives, I decided to use 60 as the cut off score. 56 5.2 Drop Outs and Number of Sessions Attended
Four participants withdrew before the end of treatment (1 in EVA, and 3 in EO). For these participants, assessments at pretreatment were carried forward for the acute and follow-up analyses. Ten additional participants failed to return for follow-up assessment3 (2 (12%) in DAVID, 2 (12%) in EO, 3 (21%) in EVA, and 3 (20%) in EVS). Their posttreatment scores were carried forward for the follow-up analyses. During informed consent, all participants had agreed on a 3- treatment session protocol. Among completers, 13 participants attended only 2 sessions (3 (21%) in DAVID, 3 (23%) in EO, 4 (30%) in EVA, and 3 (21%) in EVS). Among the 4 participants who dropped out during the acute phase, 1 attended only two sessions (1 in EVA). The conditions did not differ significantly in attrition rates or the number of sessions attended. Moreover, those who dropped out did not differ significantly on pretreatment measures from completers.
5.3 Baseline Differences
The groups did not differ on any of the outcome measures at pretreatment (see Table 3 and Table 4). However, gender appeared unequally distributed across groups, χ2=8.62, p<0.05, with relatively more females in the DAVID condition, and relatively fewer females in the EVS condition (see Table 2). An ANOVA confirmed that the mean credibility as well as expected level of improvement as measured by the RTQ (Borkovec & Nau, 1972) was comparable across the three exposure conditions. However, participants in the
3 The high degree of attrition that occurred from posttreatment to follow-up may be best explained by the fact that we could not require students to return for assessment following the completion of the semester. 57 DAVID condition provided lower credibility ratings compared to participants in the exposure conditions, F(1, 53)=8.73, p<0.05. The mean RTQ-expected level of improvement score in the DAVID condition did not differ from the mean observed in the exposure conditions.
5.4 Effects at Posttreatment
Table 3 and Table 4 present changes in public speaking outcome indices and general social anxiety indices, respectively.
5.4.1 Within-Group Effects
Results of repeated measures MANOVAs revealed significant pretreatment to posttreatment improvement in public speaking anxiety for all conditions (ps<.0.05). The improvement reported by the EVS condition was most robust, as indicated by statistically significant pre- to posttreatment effect sizes for each of the three public speaking anxiety indices (ps<.0.001; See Figure 3). The EO and EVA conditions showed statistically significant improvements on Subjective Peak Fear (ps<.0.001), whereas the DAVID condition showed statistically significant reductions in both Subjective Peak Fear, F(1, 14)=15.05, p<0.01 and on LSAS-Public Speaking Fear, F(1, 14)=7.05, p<0.05.
58 PL 2.50 2.28 EO EVA 2.00 EVS 1.51 1.50 1.34 1.18 1.19 1.21
1.00 Pre-Post ES Pre-Post 0.62 0.54 0.460.42 0.50 0.33 0.14
0.00 Speech Fear LSAS-PS Fear LSAS-PS Avoidance
Figure 3. Pre-to posttreatment effect sizes for public speaking anxiety indices.
Results of repeated measures MANOVAs for the general social anxiety measures were only significant for the EVS and DAVID conditions (ps<0.05). Examination of the individual indices revealed that the significant effect observed in the DAVID condition was qualified by a significant improvement on the LSAS-Social Interaction Avoidance (ES: 0.73, F(1, 13)=10.85, p<0.05), whereas changes on the other scales did not reach statistical significance (ES ranging from
-0.07 to 0.45). Participants in the EVS condition showed statistically significant improvement on both the LSAS-Social Interaction Fear and Avoidance (ES: 0.53, F(1, 14)=10.33, p<0.01, and 0.92, F(1, 14)=23.76, p<0.001 respectively). Other scales did not change significantly (ES ranging from 0.05 to 0.31).
59 Table 3. Means and SDs for Public Speaking Anxiety Indices DAVID EO EVA EVS N M SD N M SD N M SD N M SD Peak Subjective Fear (0-100) Pre 14 82.14 14.77 16 85.00 11.55 14 80.71 13.85 15 78.00 82.14 Post 14 58.57 19.94 13 49.23 27.53 13 53.08 23.23 15 41.33 58.57 FU 12 49.17 20.65 11 48.18 26.39 10 42.00 25.30 12 41.67 49.17 LSAS-Public Speaking Fear (0-15) Pre 14 11.79 2.01 16 10.63 2.50 14 11.36 1.86 15 11.40 2.29 Post 14 10.64 2.24 13 9.23 3.52 13 10.08 3.15 15 8.73 2.19 FU 12 10.08 3.06 11 9.00 2.83 10 9.50 3.72 12 8.42 2.39 LSAS-Public Speaking Avoidance (0-15) Pre 14 10.93 3.22 16 10.06 3.47 14 9.71 3.47 15 10.73 2.52 Post 14 10.00 2.29 13 8.31 3.79 13 9.54 3.71 15 7.53 2.75 FU 12 9.67 3.60 11 7.36 3.59 10 8.20 4.47 12 6.92 2.11 Table 4. Means and SDs for General Social Anxiety Indices DAVID EO EVA EVS N M SD N M SD N M SD N M SD LSAS-Social Interaction Fear (0-24) Pre 14 16.64 3.52 16 14.06 3.92 14 15.43 5.40 15 14.60 5.12 Post 14 15.14 3.11 13 11.38 5.53 13 13.69 7.15 15 12.13 4.21 FU 12 14.00 3.16 11 10.36 4.82 10 15.80 5.75 12 11.50 3.53 LSAS-Social Interaction Avoidance (0-21) Pre 14 16.00 4.40 16 12.63 4.13 14 14.21 6.19 15 14.13 5.58 Post 14 12.86 4.15 13 9.08 5.51 13 10.31 5.81 15 9.07 5.38 FU 12 11.67 3.55 11 8.73 5.24 10 11.50 4.60 12 8.67 3.34 LSAS-Observation by Others Fear (0-12) Pre 14 3.07 1.38 16 2.31 2.50 14 2.29 2.23 15 3.47 2.07 Post 14 2.79 1.37 13 1.92 2.29 13 2.38 2.43 15 2.80 2.21 FU 12 2.67 2.02 11 2.18 2.36 10 3.10 2.81 12 2.58 1.31 LSAS-Observation by Others Avoidance (0-9) Pre 14 1.79 2.19 16 1.38 1.75 14 1.21 2.04 15 1.60 1.64 Post 14 1.64 1.60 13 1.31 2.02 13 1.31 2.10 15 1.33 1.54 FU 12 1.17 1.19 11 1.09 1.45 10 1.50 1.51 12 1.00 1.35 LSAS-Eating and Drinking in Public Fear (0-6) Pre 14 1.00 1.11 16 0.94 1.34 14 1.07 1.33 15 0.93 1.44 Post 14 0.93 1.14 13 0.69 0.95 13 1.00 1.35 15 0.60 1.12 FU 12 0.67 0.78 11 0.64 0.67 10 0.80 1.03 12 0.67 1.15 LSAS-Eating and Drinking in Public Avoidance (0-6) Pre 14 0.86 1.17 16 0.88 1.15 14 0.71 0.83 15 0.53 1.30 Post 14 0.79 1.19 13 1.08 1.80 13 0.62 0.87 15 0.47 1.13 FU 12 0.42 0.67 11 0.36 0.67 10 0.70 0.95 12 0.42 1.16
60 5.4.2 Between-Group Effects
The overall MANOVA on public speaking indices was significant, F(3, 55)=4.69, p<0.05. Follow-up univariate analyses revealed that the differential treatment effect was qualified by significant group differences in avoidance of public speaking, as measured by the LSAS-Public Speaking Avoidance, F(3, 53)=3.64, p<0.05. A-priori planned Helmert contrasts (i.e, EO, EVA, EVS vs. DAVID; EO vs. EVA, EVS; and EVA vs. EVS) were not significant. Since within-group analyses suggested a possible advantage of the EVS condition, the MANOVA was followed up by Tukey HSD post-hoc comparisons. Results revealed a statistically significant advantage of EVS over DAVID and EVA in avoidance of public speaking, ps<0.05. No other significant between- group differences were observed.
5.4.3 High End-State Functioning
Figure 4 presents the HEF rates at posttreatment for each of the four conditions. Of 59 participants, 42 (36%) achieved HEF status at posttreatment. HEF rates were 21%, 38%, 29%, and 53% for the DAVID, EO, EVA, and EVS groups, respectively. These differences did not reach statistical significance.
61 PL 60% 53% EO 50% EVA EVS 38% 40%
29% 30% 21% 20%
10%
0% HEF at Posttreatment
Figure 4. High End-State functioning at posttreatment
5.5 Effects at Follow-up
Changes in public speaking outcome indices and general social anxiety indices are presented in Table 3 and Table 4, respectively.
5.5.1 Within-Group Effects
Results of repeated measures MANOVAs revealed significant pretreatment to follow-up improvements in public speaking anxiety for all groups
(see Figure 5). Univariate analyses showed that the EVS and EO conditions showed significant improvements on all indices (ps<0.0001, and <0.01, respectively). The DAVID condition showed significant improvements on Subjective Peak Fear, F(1, 13)=22.01, p<0.001, and LSAS-Public Speaking Fear, F(1, 13)=5.73, p<0.05. The EVA condition showed only significant improvement on Subjective Peak Fear, F(1, 13)=18.37, p<0.01.
62 Repeated measures MANOVAs for the general social anxiety measures were only significant for the EVS and EVA conditions (ps<0.05). Follow-up univariate ANOVAs revealed that the EVA condition reported significant improvement on the LSAS-Social Interaction Avoidance (ES: 0.61, F (1,13)=21.33, p<0.05), while changes on the other scales did not reach significance (ES ranging from -0.09 to 0.37). Participants in the EVS condition showed significant improvements on the LSAS-Social Interaction Fear and Avoidance (ES: 0.75, F (1, 14)=9.92, p<0.01 and 1.14, F (1, 14)=37.44, p<0.001 respectively), as well as on the LSAS-Observation by Others Fear (ES: 0.57, F (1, 14)=8.07, p<0.05). Changes on the other scales were not significant (ES ranging from 0.17 to 0.32)
PL 2.50 EO 2.02 EVA 2.00 EVS 1.501.491.48
S 1.45 1.50 1.08
Pre-FU E Pre-FU 1.00 0.69 0.75 0.61 0.51 0.41 0.50 0.30
0.00 Speech Fear LSAS-PS Fear LSAS-PS Avoidance
Figure 5. Pretreatment to follow-up effect sizes for public speaking anxiety indices.
63 5.5.2 Between-Group Effects
MANOVAs on public speaking anxiety indices and general social anxiety indices revealed no significant differences between groups.
5.5.3 Relapse and High End-State Functioning
Ten (17%) participants showed relapse during the follow-up period, with no statistical differences among the conditions (see Figure 6). The percentage of participants achieving high end-state functioning at follow-up was 29%, 44%, 43% and 47% for the DAVID, EO, EVA and EVS groups respectively. These differences were not statistically significant.
PL 50% 47% 44% EO 45% 43% EVA 40% EVS 35% 29% 30% 25% 19% 20% 20% 14% 14% 15% 10% 5% 0% Relapse HEF at Follow-Up
Figure 6. Relapse and High End-State functioning at follow-up.
64 5.6 Mediator Analyses
We hypothesized that the EVS condition would show increased efficacy because they were provided information during treatment that highlighted their tendency to overestimate the likelihood that they appear anxious while speaking in front of others. This change in likelihood appraisals was expected to be mediated by improved performance ratings. On the other hand, we hypothesized that the EVA condition would experience increased benefits because they were provided information during treatment that highlighted the tendency to overestimate the negative consequences associated with appearing anxious in from of others. The level of baseline speech-related appraisals was consistent with expectations. Participants provided low performance ratings (M=26.72, SD=20.38), and high likelihood (M=86.34, SD=14.71) and cost (M=65.52, SD=22.55) ratings, with no significant differences among the treatment conditions. Pre- to posttreatment changes in speech-related appraisals are depicted in Figure 7.
65 PL 2.50 2.09 EO 2.00 EVA EVS 1.51 1.55 1.50 1.25 1.13 1.10 0.950.99 1.00 0.840.86
Pre-Post ES Pre-Post 0.72 0.44 0.50
0.00 Performance Likelihood of Cost of Anxious Anxious Appearance Appearance
Figure 7. Pre- to posttreatment effect sizes for speech-related appraisals.
5.6.1 Effect of Treatment on Mediator (Step 1)
In Step 1, I tested the effects of treatment on the proposed mediators by performing ANOVAs with the treatment group as the grouping factor and the residualized (pre-post) change score for the mediator as the dependent variable. To test the specific mediator hypotheses, three separate ANOVAs were performed: (1) performance residualized change scores as the DV and EVS vs. EVA, DAVID, EO as the treatment grouping variable; (2) likelihood residualized change scores as the DV and EVS vs. EVA, DAVID, EO as the treatment grouping variable; and (3) cost residualized change scores as the DV and EVA vs. DAVID, EO, EVS as the treatment grouping variable. Results revealed that EVS showed greater improvement in performance appraisals, F(1, 56)=4.50, p<0.05, and a greater reduction likelihood estimates, F(1, 56)=4.80, p<0.05, compared to the other conditions. Changes in cost
66 estimates observed in the EVA condition were not significantly different from those observed in the other conditions. These data confirm that the first condition for mediation was met with respect to performance and likelihood appraisals, but not for cost estimates. No further analyses were performed for cost estimates.
5.6.2 Effect of Treatment on Outcome (Step 2)
In Step 2, I tested for the presence of a treatment effect by performing a MANOVA with treatment group (EVS vs. EVA, DAVID, EO) as the grouping factor and residualized change scores of the three public speaking anxiety indices (i.e., Peak Fear during Speech, LSAS-Public Speaking Anxiety Fear, LSAS- Public Speaking Anxiety Avoidance) as the dependent variables. The MANOVA was followed up by univariate tests for each measure. EVS-treated participants displayed significantly greater improvement in public speaking anxiety relative to the non-EVS treated participants, F(3, 54)=3.55, p<0.05. Follow-up univariate tests revealed significant treatment effects on all public speaking outcome measures (all ps<0.05). These data confirm that the second condition for mediation was met.
5.6.3 Relationship between Mediator and Outcome (Step 3)
In Step 3, the relationship between the proposed mediators and outcome was examined. To test whether were an improvement in performance estimates was associated with reductions in likelihood estimates, residualized likelihood scores were subjected to an ANCOVA with treatment (EVS vs. EVA, DAVID, EO) as the grouping factor, and with residualized performance scores as the covariate. Similarly, to test whether the improvement in likelihood estimates was
67 associated with the improvement in public speaking anxiety , a MANCOVA was performed with treatment (EVS vs. EVA, DAVID, EO) as the grouping factor, residualized change scores of the three public speaking anxiety indices as the DVs, and the likelihood residualized change score as the covariate. Significant multivariate effects of the covariate followed up by univariate tests for each of the three public speaking measures. Results revealed that changes in performance estimates were associated with changes in likelihood estimates, F(1, 55)=8.03, p<0.01. There was also a significant association between the likelihood residualized change score and residualized change in public speaking anxiety, F(3, 53)=5.72, p<0.01. Follow-up univariate tests for each of public speaking outcome measures yielded significant results for Peak Fear during Speech, F(1, 56)=9.16, p<0.01, and LSAS-Public Speaking Anxiety Avoidance: F(1, 56)=3.83, p<0.05, but not for LSAS-Public Speaking Anxiety Fear. These data confirm that the third necessary condition for mediation was met. However, changes in likelihood ratings did not qualify as a mediator for public speaking fear as measured by the LSAS-Public Speaking Anxiety Fear scale.
5.6.4 Effect of Treatment after Controlling for Mediator (Step 4)
The final step tested the relationship between treatment and outcome after controlling for the effects of the proposed mediator. According to Baron and Kenny (1986), evidence for full mediation exists when the relationship between treatment and outcome is no longer significant after controlling for the effects of the mediator; whereas evidence for partial mediation exists when the relationship
68 between treatment and outcome is significantly attenuated (but still significant) after controlling for mediator effects. This step was tested by comparing the effect of treatment in the second step to the effect of treatment in the third step. The effects of EVS on likelihood ratings were reduced considerably after controlling for changes in performance estimates, F(1, 55)=2.14, p>0.1. Entering performance residual change scores for the mediator variable into the model led to a 58% reduction in the variance explained by EVS. These findings indicate that the effect of performance feedback treatment on likelihood ratings is fully mediated by improvement in performance estimates. Similarly, the association between EVS and public speaking anxiety reduction was reduced considerably after controlling for changes in likelihood ratings, F(3, 53)=1.90, p>0.10. Including likelihood residualized change scores into the model led to a 55% reduction in the variance explained by EVS. Follow- up univariate measures revealed that changes in changes in likelihood ratings fully mediated the effects of treatment on Fear during Speech, F(1, 55)=2.27, p>0.1, and partially mediated reductions in LSAS-Public Speaking Anxiety Fear, F(1, 55)=2.60, p>0.1. Inclusion of the mediator resulted in a 68% reduction in the variance in Fear during Speech explained by EVS, while a reduction of 43% was observed for LSAS-Public Speaking Anxiety Avoidance.
5.7 Moderator Analyses
No significant interactions were observed between any of the potential moderators and the treatment factor (EVS vs. DAVID, EO, EVA) (all ps>0.3).
69 This finding suggests that the increased efficacy observed in the EVS condition compared to the non-EVS conditions was not moderated by gender, community status (student vs. community), or social phobia status (GSP vs non-GSP).
5.8 Clinically Meaningful Differences and Statistical Power
Given the current study has modest cell sizes relative to most clinical trials, I investigated whether the apparent lack of statistical differences in outcome was due to limitations of statistical power. A simulation data set was analyzed using SPSS in order to determine the statistical power to detect clinically meaningful differences (controlled ES>0.50). Setting the alpha level at 0.05, it was found that 15 participants per cell allows for insufficient power (0.24) to detect a medium effect size, where the pretreatment and posttreatment /follow-up measures are significantly correlated (rs range from .30 to .80). Therefore, the current sample size does not allow us to reject the hypothesis that there are clinically meaningful differences between the different treatment conditions. Controlled effect sizes were calculated in order to get an impression of the magnitude of effects for the most relevant contrasts. As shown in Figure 8, the hypothesized clinical superiority for EVS over EO was apparent at posttreatment assessment (i.e, ES>.50 for major outcome measures). It was found that 45 participants per cell would allow for sufficient power (0.80) to detect theses clinically meaningful differences at posttreatment. However, no clinically meaningful differences were observed at follow-up assessment. A minimum of
70 300 participants per cell is necessary to detect these small differences at follow- up.
Subjective Peak Fear LSAS-PS-Fear LSAS-PS-Avoidance HEF
0.60 0.53 0.52 0.51 0.49 0.50 0.45 0.41 0.40 0.30 0.22 0.22 0.20
0.10 0.00
Controlled Effect Sizesat Post EVS. vs. EO EO vs. DAVID
Subjective Peak Fear LSAS-PS-Fear LSAS-PS-Avoidance HEF 0.58 0.60 0.50 0.39 0.40 0.34 0.30 0.27 0.19 0.19 0.20 0.15
0.10 0.01 0.00 Controlled Effect Sizes at FU EVS. vs. EO EO vs. DAVID
Figure 8. Clinically meaningful differences.
71 CHAPTER 6: DISCUSSION
6.1 Efficacy
The primary objective of the current study was to investigate whether adding videotape feedback procedures to exposure treatment would improve treatment efficacy. Participants suffering from social phobia were randomized to receive non-pill placebo, exposure without videotape feedback, exposure with audience videotape feedback, or exposure with performance videotape feedback. As predicted, exposure was associated with significant pre- to posttreatment reductions on public speaking anxiety measures. The magnitude of improvement was moderate, as 38% EO participants moved into the non-clinical range at posttreatment. Contrary to expectation, having participants focus on external stimuli by showing videotaped audience responses did not enhance the effect of exposure to public speaking. While EVA participants showed a significant improvement in speech task fear ratings, the posttreatment HEF rate observed in the EVA condition (29%) was no different from that observed in the EO condition. In contrast, videotaped performance feedback did facilitate the acute effect of repeated exposure to public speaking exercises. Over half the participants (53%) who received performance feedback (EVS) showed marked reduction in public speaking anxiety at posttreatment, and qualified for HEF status. Participants who received our credible placebo treatment (D.A.V.I.D) reported significant improvement as reflected by medium pre-to posttreatment
72 effect sizes. However, treatment with the D.A.V.I.D. was associated with the lowest rate of HEF (21%). Controlled effect sizes across the major outcome measures underscore the superiority of performance videotape feedback in the acute phase (EVS vs. EO controlled ESs ranged from 0.41 to 0.51), and the limited advantage of exposure only over placebo (EO vs. DAVID controlled ESs ranged from 0.22 to 0.52). Due to insufficient power, these between-group differences were, however, not statistically significant. The pattern of results changed considerably at the 1-month follow-up period. Relapse rates were consistent with rates commonly observed in CBT trials for social phobia (Gould et al., 1995), and there were no differences among the conditions. HEF analyses revealed that the advantage of performance feedback had dissipated during the follow-up period. All conditions, with the exception of the EVS condition, showed an increase in HEF rates at follow-up (29%, 44%, 43%, and 47% for the DAVID, EO, EVA groups, respectively). Controlled effect sizes for changes in the major outcome measures at follow-up were in line with the lack of statistical differences among the conditions (EVS vs. EO controlled ESs ranged from 0.15 to 0.34; and EO vs. DAVID controlled ESs ranged from 0.01 to 0.58). In summary, the results indicate that the addition of videotape feedback procedures does not enhance the effects of exposure treatments consisting of repeated speech exercises. Consistent with preliminary experiments (Harvey et al., 2000; Rapee & Hayman, 1996), having participants contrast their imagined
73 performance to their actual performance did facilitate changes in perceptions of their performance. Moreover, these changes were associated with an accelerated speed of improvement in public speaking anxiety. However, despite the increased rate of improvement, performance feedback was not associated with better outcome at 1-month follow-up. At first glance, these results seem to contradict findings reported by Clark and colleagues (Clark et al., in press). They recently examined the efficacy of a “new” cognitive therapy program that included videotape feedback procedures, among other new treatment components (e.g., safety behavior fading). Based on pre- to posttreatment effect sizes, their new treatment was more effective than traditional cognitive behavioral treatment. Since they did not include a comparison treatment (i.e, cognitive therapy without new techniques), it is impossible to determine the relative efficacy of these new techniques. The lack of additional benefits in the audience videotape feedback condition seems to be inconsistent with recent data suggesting that instructing patients to focus on external stimuli rather than on the self increases the effects of exposure (Wells & Papergeorgiou, 1998). A possible explanation for this discrepancy is that the EVA participants did not show a decrease in self-focused attention. We did not measure this construct, and can therefore not ascertain that this change occurred. Wells and Papergeorgiou argue that it is the decrease in self-focused attention, and not the increase in focus on external stimuli per se, that enhances the effects of exposure.
74 How do the effects of our brief exposure-based intervention (three 45-min. sessions compare to the effects of full cognitive-behavioral treatment packages? Meta-analyses of cognitive-behavioral treatments for social phobia report larger pre- to posttreatment effect sizes (between 0.80 and 1.08) for social phobia indices (Federoff & Taylor, 2001; Feske & Chambless, 1995; Taylor, 1996). Moreover, multi-site randomized clinical trials have reported larger response rates (around 75%; Heimberg et al., 1998; Liebowitz et al., 1999). Several factors may account for the lower efficacy rates observed in the present trial. First, the present study specifically targeted public speaking anxiety as opposed to general performance and social interaction anxiety. Public speaking anxiety typically appears high on the hierarchy of feared situations. Improvements reported in other clinical trials reflect reductions in performance and social interaction situations, and it is possible that the changes in public speaking anxiety are not of the same magnitude. Second, participants in the present trial received substantially less therapist contact compared to participants in other clinical trials for social phobia. Traditional CBT packages consist of 12 90-min. sessions, providing patients an amount of therapist contact that is approximately ten-fold what patients received in the current trial. Third, most therapists in the current trial were undergraduate students. Although rated as competent to conduct the interventions, their level of therapy experience was minimal relative to therapists involved in other published randomized clinical trials. Therefore, it is possible that therapists in the current trial were less skillful in conducting treatment. Lastly, in contrast to traditional CBT protocols, participants in the
75 current study were not provided homework assignments. The increased efficacy derived from skill building exercises has been well-documented (Edelman & Chambless, 1995; Leung & Heimberg, 1996). The findings with respect to our placebo condition seem to provide further insight. Consistent with a previous study with claustrophobic individuals (Powers, Smits, & Telch, in press), the DAVID placebo condition was associated with low HEF rates. However, Powers et al. reported HEF rates for brief exposure for claustrophobia that were four-fold the rate of the DAVID condition, whereas our brief intervention consisting of repeated exposure exercises hardly outperformed the DAVID condition. The difference in the pattern of results of these two studies underscores that social phobia is a severe form of anxiety pathology that warrants a high dose of treatment.
6.2 Mechanisms of Change
Attempts to identify factors that may moderate the effects of performance feedback on public speaking anxiety reduction were unsuccessful; none of the potential moderators examined including gender, subtype, and community status influenced the magnitude or direction of the videotape feedback effects. Our findings are in line with previous studies, which have demonstrated that social phobia subtype does not affect the rate of improvement derived from CBT (Brown et al., 1995; Hope, Heimberg, & Bruch, 1995; van Velzen, Emmelkamp, & Scholing, 1997). Contemporary psychosocial theories of social phobia emphasize the role of threat appraisals in the maintenance of the disorder (Clark & Wells, 1995;
76 Rapee & Heimberg, 1997). Specifically, it has been demonstrated that social phobic sufferers form a mental representation of his/her appearance from an audience perspective when they enter a social situation (Wells, Clark, & Ahmad, 1998). Empirical evidence further suggests that the observer’s image taken by social phobics is distorted, as they typically overpredict the degree to which people notice their symptoms of anxiety (Alden & Wallace, 1995), and they tend to significantly underrate their performance (Rapee & Lim, 1992). Baseline data from the current study are consistent with these observations, as participants tended to rate their performance as low, and the likelihood that people would notice their anxiety symptoms as high. Recent studies have indicated that the provision of videotape feedback can help socially anxious persons address their distorted perceptions. Rapee and Hayman, (1996) compared performance ratings of participants who received performance videotape feedback following a speech to performance ratings of participants who did not receive videotape feedback. The results showed a greater match between self-ratings and ratings made by independent observers among persons who watched a taped version of their speech, compared to participants who did not receive feedback. Harvey, Clark, Ehlers, and Rapee (2000) found that cognitive preparation can increase the beneficial effects of videotape feedback. Half of the participants who received videotape feedback were instructed to contrast their imagined performance to their actual performance, which was provided by the videotape feedback. The other half of the sample did not receive any specific instructions. Greater performance ratings were observed among
77 participants who received cognitive preparation compared to those who were not given cognitive preparation. These findings suggest that adding cognitive preparation to videotape feedback procedures leads to greater disconfirmation of appearance/performance concerns. We examined whether changes in perceptions (e.g., performance, and the appraisals of the degree to which people notice their symptoms of anxiety) mediated the accelerated rate of improvement observed in the EVS condition. We found that EVS participants reported greater improvement in performance ratings and in the estimates of the degree to which they think people notice their anxiety (i.e., likelihood ratings). Moreover, the improvement in self-ratings of performance mediated the reduction in likelihood ratings. Further, the mediation analyses provided support for the hypothesis that performance feedback exerts its effects on public speaking anxiety symptoms by decreasing people’s concerns about the likelihood that they will appear anxious while performing in front of others. These mediational effects were observed across two of the three major public speaking anxiety symptoms (i.e., fear during a speech task, and avoidance of “real life” public speaking anxiety situations). Our results are consistent with contemporary theories that social phobics have a tendency to make excessively negative predictions about how they appear to others in the maintenance of social phobia. Our findings, however, suggest that this likelihood bias may only have little relevance for the maintenance of social phobia. Since we did not monitor the changes in other proposed maintaining
78 factors4, we were unable to determine the relative weight of appearance concerns in the maintenance of social anxiety. Studies that determine the interrelationships between the proposed maintaining factors and changes in social anxiety will guide the development of more potent treatment techniques. It is recommended that these future studies include both placebo-, and waitlist-control groups.
6.3 Clinical Implications
What implications do our findings have for clinicians who treat social phobia patients? Exposure treatment often receives negative attention, because it causes significant distress to the patient in the process. Our data suggests that exposure treatment is a procedure that is well-tolerated among social phobia sufferers. We found only a 9% attrition rate during the acute phase of treatment. These findings are in line with meta-analyses that have reported low mean attrition rates for CBT (10.7%; Gould et al., 1997). Most clinical trials for anxiety disorders employ highly trained psychologists and multifaceted demanding protocols. For example, therapists in the most recent clinical trial for social phobia (Clark et al., in press) were four clinical psychologists, who were experienced in the use of cognitive-behavioral treatments for anxiety. They used techniques such as (a) developing with patients a personal version of Clark & Wells’ (1995) model using their own thoughts, images, anxiety symptoms, safety behaviors, and attentional strategies; (b)
4 Clark & Wells (1995) and Rapee & Heimberg (1997) highlight four maintenance processes: 1) an increase in self-focused attention and reduction in observation of other people and their responses; 2) the use of misleading internal information to make excessively negative inferences about how one appears to others; 3) extensive use of safety behaviors; and 4) negatively biased anticipatory and post-event processing. 79 identifying safety behaviors, and demonstrate their adverse effects with experiential exercises; and (c) identifying dysfunctional assumptions and modify these by behavioral experiments and by cognitive restructuring techniques. Is expertise then a requirement for efficacy? Our data provides only limited support for the idea that exposure treatment can be effective when individuals, who do not have extensive experience with people suffering from anxiety disorders, deliver it. Therapists in the present study were undergraduate research assistants, with little to no formal clinical training, who only received brief education in anxiety disorders and an exposure treatment manual. Although, exposure was associated with significant improvements, less than half of the participants moved into the normal range. Again, a direct comparison between specialists and non-specialists is needed to test this hypothesis. Although effective, exposure treatment for public speaking anxiety requires substantial resources (i.e, audience members). Most therapists do not have access to confederates, and public speaking exposures are not feasible homework assignments. While uncontrolled comparisons suggest that group therapy is not associated with better outcome (see Gould et al., 1997), it is a form of treatment that facilitates the practice of in-session exposure exercises. Therefore, group therapy offers an alternative for exposure-based treatment in real-world clinical practice. However, for individuals and therapists who are unwilling to participate in group therapy, conducting exposure to public speaking remains difficult. Preliminary evidence implicates Virtual Reality (VR) as a potential alternative for in vivo exposure to public speaking situations. For
80 example, Anderson, Rothbaum and Hodges (2003) completed a case study (N=2) with patients meeting diagnostic criteria for social phobia with significant public speaking fears. Treatment consisted of repeated exposure to a virtual audience. Results indicated that treatment was successful, as reflected by decreases in self- reported measures of public speaking anxiety that were comparable to controlled clinical outcome trials for social phobia. Finally, should clinicians use videotape feedback procedures in their clinical practice? Our findings suggest that the application of performance feedback accelerates the rate of improvement in symptoms, but it does not lead to better long-term outcome. An examination of the attrition rates suggests that speedier recovery may positively affect the commitment to treatment. Although not statistically significant, the EVS condition showed lower drop out rates during the acute phase of treatment compared to the EO condition (0% vs. 18%). Thus, videotape feedback procedures may be helpful for patients who express concerns about treatment commitment. This hypothesis warrants further research.
6.4 Limitations
Several limitations deserve comment. First, as mentioned above, the present trial is underpowered because of the modest cell sizes. Consequently, it was probable that our statistical analyses missed meaningful effects. Commensurate with guidelines put forth by Kraemer and colleagues (Kraemer, Wilson, Fairburn, & Agras, 2002), it was decided to base our conclusions with respect to the outcome analyses on controlled effect sizes rather than effects that reached statistical significance. Since the current sample size allowed for
81 insufficient power to detect medium main effects, the study was obviously underpowered to detect interactions. Therefore, our moderator analyses should be interpreted with caution, and require reexamination in a study with a sufficient sample size. Second, the lack of a waitlist-control condition is a shortcoming of the present study. Although the course of social phobia is typically chronic when untreated (Gould et al., 1997), we cannot rule out that the changes observed over time simply reflect regression to the mean. Further, inclusion of a waitlist condition would allow us to make inferences with respect of the effect of repeated behavioral assessment (pretreatment, posttreatment, and follow-up), thereby potentially helping to clarify the effects observed in the placebo condition. Third, the follow-up period of four weeks is perhaps too brief to make inferences about the stability of treatment effects over the long-term. It should be noted that the overall HEF rate at follow-up was greater than the HEF rate observed at posttreatment. The pattern of responses on the public speaking measures, coupled with the nature of the illness, is potentially indicative of greater long-term response rates. Specifically, participants generally reported greater pretreatment to follow-up decreases in public speaking avoidance compared to public speaking fear. Following the CBT model of anxiety, it is expected that a decrease in avoidance initially impedes fear reduction, but eventually results in greater fear reduction. This delayed treatment effect has been well documented in the CBT panic disorder (see Gould, Otto, and Pollack, 1995). However, large clinical trials for social phobia show considerable relapse rates for CBT (17%;
82 Liebowitz et al., 1999). Whether the current treatment protocols are associated with delayed (differential) treatment effects, can only be determined by studies that include more frequent and longer follow-up assessments. Fourth, we did not assess and monitor other relevant factors that may have impacted the rate of improvement observed in the current study. Because of feasibility considerations, participants were not assessed for comorbid psychiatric conditions such as major depression, and axis-II disorders. Some clinical trials for anxiety disorders have identified these conditions as moderators of treatment response (Turner, Beidel, Wolff, Spaulding, & Jacob, 1996; van Velzen et al., 1997) . It should be noted that this factor is not likely to explain the differential treatment effects or the lack thereof, because of the randomized assignment that was employed in the current study. Inclusion of these potentially relevant measures in future studies can only aid the development of treatment matching strategies. Fifth, the pretreatment attrition in the current study was considerable, although not different from large recently published trials for anxiety disorders (Coles, Turk, Jindra, & Heimberg, 2004; Hoffman et al., 1998; Huppert, Franklin, Foa, & Davidson, 2003). One of the criticisms of randomized controlled trials is that they may be limited in terms of generalizability because they suffer from selection bias. Specifically, it has been hypothesized that patients who agree to participate are more likely to respond to treatment than those who also meet criteria, but decline participation (Chambless & Ollendick, 2001). Coles and colleagues (Coles et al., 2004) recently provided some data from a specialized
83 anxiety treatment clinic that does not support this hypothesis. They found that the two subgroups did not differ on demographic variables, symptom severity, or quality of life. However, since the current study failed to assess the reasons for reluctance to take part in the study, we could not directly test this hypothesis. Sixth, although the current findings are consistent with the hypothesis that reductions in performance/anxious appearance concerns mediate the effects of performance feedback on public speaking anxiety, our design does not allow us to rule out the possibility that the change in these appraisals was a consequence as opposed to a cause of anxiety reduction. Future studies examining the validity of this mediator focus on demonstrating temporal precedence of the mediator. Finally, our study design does not address the important question of whether the mediational effects of appraisals on social anxiety reduction are specific to psychosocial treatments. Preliminary evidence suggests that changes in anxiety sensitivity (proposed maintaining factor for panic disorder) not only operate in CBT (Smits, Powers, Cho, & Telch, 2004) but also in pharmacotherapy (Simon, Otto, Smits, Colette Nicolaou, Reese, & Pollack, 2004) for panic disorder. Therefore, it is recommended that future studies of alternative treatments with established efficacy include measures that tap the proposed maintaining factors.
84 REFERENCES
Alden, L.E., & Wallace, S.T. (1995). Social phobia and social appraisal in successful and unsuccessful social interactions. Behaviour Research and Therapy , 33(5), 497-505. American Psychiatric Association. (1980). Diagnostic and statistical manual of mental disorders (3rd edition). Washington, DC: Author. American Psychiatric Association. (1987). Diagnostic and statistical manual of mental disorders (3rd-revised edition). Washington, DC: Author. American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders (4th edition). Washington, DC: Author. Amies, P.L., Gelder, M.G., & Shaw, P.M. (1983). Social phobia: a comparative clinical study. British Journal of Psychiatry, 142, 174-9. Amir, N., Foa E.B., & Coles, M.E. (1998). Negative interpretation bias in social phobia. Behaviour Research and Therapy, 36, 945-957 Anderson, P., Rothbaum, B.O., & Hodges, L.F. (2003). Virtual reality exposure in the treatment of social anxiety: two case reports. Cognitive and Behavioral Practice, 16, 200-205. Baker, S.L., Heinrichs, N., Kim, H.J., & Hofmann, S.G. (2002). The liebowitz social anxiety scale as a self-report instrument: a preliminary psychometric analysis. Behaviour Research and Therapy, 40(6), 701-15. Bandura, A. (1977). Self-efficacy: toward a unifying theory of behavioral change. Psychological Review, 84(2), 191-215. Bandura, A., Adams, N.E., & Beyer, J. (1977). Cognitive processes mediating behavioral change. Journal of Personality and Social Psychology, 35(3), 125-39. Baron, R.M., & Kenny, D.A. (1986). The moderator-mediator variable distinction in social psychological research: conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51(6), 1173-1182. Barlow, D.H. (1988). Anxiety and its disorders. New York, N.Y.: Guilford. Barlow, D.H., DiNardo, P.A., Vermilyea, B.B., Vermilyea, J., & Blanchard, E.B. (1986). Co-morbidity and depression among the anxiety disorders. Issues in diagnosis and classification. Journal of Nervous and Mental Disease, 174(2), 63-72. 85 Beck, A.T., & Clark, D.A. (1997). An information processing model of anxiety: automatic and strategic processes. Behaviour Research and Therapy, 35(1), 49-58. Beck, A.T., Emery, G., & Greenberg, R. (1985). Anxiety disorders and phobias: a cognitive perspective. NewYork, New York: Basic Books.
Beidel, D.C., Turner, S.M., & Dancu, C.V. (1985). Physiological, cognitive and behavioral aspects of social anxiety. Behaviour Research and Therapy, 23(2), 109-17. Birbaumer, N., Grodd, W., Diedrich, O., Klose, U., Erb, M., Lotze, M., Schneider, F., Weiss, U., & Flor, H. (1998). fMRI reveals amygdala activation to human faces in social phobics. Neuroreport, 9(6), 1223-6. Blanco, C., Antia, S.X., & Liebowitz, M.R. (2002). Pharmacotherapy of social anxiety disorder. Biological Psychiatry, 51(1), 109-20. Boone, M.L., McNeil, D.W., Masia, C.L., Turk, C.L., Carter, L.E., Ries, B.J., & Lewin, M.R. (1999). Multimodal comparisons of social phobia subtypes and avoidant personality disorder. Journal of Anxiety Disorders, 13(3), 271-92. Borkovec, T.D., & Nau, S.D. Credibility of analogue therapy rationales. Journal of Behavior Therapy and Experimental Psychiatry, 3, 257-260. Brown, E.J., Heimberg, R.G., & Juster, H.R. (1995). Social phobia subtype and avoidant personality disorder: Effect on severity of social phobia, impairment, and outcome of cognitive behavioral treatment. Behavior Therapy, 26(3), 467-486. Caldirola, D., Perna, G., Arancio, C., Bertani, A., & Bellodi, L. (1997). The 35% CO2 challenge test in patients with social phobia. Psychiatry Research, 71(1), 41-48. Chambless, D.L., & Ollendick, T.H. (2001) Empirically supported psychological interventions: controversies and evidence. Annual Review of Psychology, 52, 685–716. Chartier, M.J., Hazen, A.L., & Stein, M.B. (1998). Lifetime patterns of social phobia: a retrospective study of the course of social phobia in a nonclinical population. Depression and Anxiety, 7(3), 113-121. Clark, D. M., Ehlers, A., McManus, F., Hackmann, A., Fennell, M., Campbell, H., Flower, T. (In Press). Cognitive therapy vs fluoxetine in generalized social phobia: a randomized placebo controlled trial. Journal of Consulting and Clinical Psychology, 67, 583-589. Clark, D. M., Salkovskis, P. M., Hackmann, A., Middleton, H., Anastasiades, P., 86 & Gelder, M. G. (1994). A comparison of cognitive therapy, applied relaxation and imipramine in the treatment of panic disorder. British Journal of Psychiatry, 164, 759-769. Clark, D. M., Salkovskis, P. M., Hackmann, A., Wells, A., Ludgate, J., & Gelder, M. (1999). Brief cognitive therapy for panic disorder: a randomized controlled trial. Journal of Consulting and Clinical Psychology, 67, 583-589. Clark, D.M., & Wells, A. (1995). A cognitive model of social phobia. In R. G. Heimberg, M. R. Liebowitz, S. A. Hope, & F. R. Schneier (Eds.), Social Phobia: Diagnosis, Assessment, and Treatment. (pp. 69-93). New York, NY: Guilford Press. Cohen, J. (1988). Statistical power Analysis for the Behavioral Sciences. 2nd edition. Hillsdale, N.J: Lawrence Erlbaum Associates. Coles, M.E., Turk, C.L., Jindra, L., & Heimberg, R.G. (In press). The path from initial inquiry to initiation of treatment for social anxiety disorder in an anxiety disorders specialty clinic. Journal of Anxiety Disorders. Davidson, J.R., Hughes, D.C., George, L.K., & Blazer, D.G. (1994). The boundary of social phobia. Exploring the threshold. Archives of General Psychiatry, 51(12 ), 975-983. De Ruiter, C., Rijken, H., Garssen, B., Van Schaik, A., & Kraaimaat, F. (1989). Comorbidity among the anxiety disorders. Journal of Anxiety Disorders, 3, 57-68. DeRubeis, R.J., Evans, M.D., Hollon, S.D., Garvey, M.J., Grove, W.M., & Tuason, V.B. (1990). How does cognitive therapy work? Cognitive change and symptom change in cognitive therapy and pharmacotherapy for depression. Journal of Consulting and Clinical Psychology, 58(6), 862-9. Dimberg, U., & Ohman, A. (1983). The effects of directional facial cues on electrodermal conditioning to facial stimuli. Psychophysiology, 20(2), 160-7. Edelman, R.E., & Chambless, D.L. (1995). Adherence during sessions and homework in cognitive-behavioral group treatment of social phobia. Behaviour Research and Therapy, 33(5), 573-7. Elkin, I., Shea, M.T., Watkins, J.T., Imber, S.D., Sotsky, S.M., Collins, J.F., Glass, D.R., Pilkonis, P.A., Leber, W.R., Docherty, J.P., & et, a.l. (1989). National Institute of Mental Health Treatment of Depression Collaborative Research Program. General effectiveness of treatments. Archives of General Psychiatry , 46(11), 971-983. Eng, W., Coles, M.E., Heimberg, R.G., & Safren, S.A. (2001). Quality of life following cognitive behavioral treatment for social anxiety disorder:
87 preliminary findings. Depression and Anxiety, 13(4), 192-193. Fairburn, C.G., Jones, R., Peveler, R.C., Carr, S.J., Solomon, R.A., O'Connor, M.E., Burton, J., & Hope, R.A. (1991). Three psychological treatments for bulimia nervosa. A comparative trial. Archives of General Psychiatry, 48(5), 463-469. Fedoroff, I.C., & Taylor, S. (2001). Psychological and pharmacological treatments of social phobia: a meta- analysis. Journal of Clinical Psychopharmacology, 21(3), 311-324. Feske, U. & Chambless, D. L. (1995). Cognitive-behavioral versus exposure only treatment for social phobia: A meta-analysis. Behavior Therapy, 26, 695-720. Foa, E.B., Franklin, M.E., Perry, K.J., & Herbert, J.D. (1996). Cognitive biases in generalized social phobia. Journal of Abnormal Psychology, 105(3), 433- 439. Foa, E.B., & Kozak, M.J. (1986). Emotional processing of fear: exposure to corrective information. Psychological Bulletin, 99(1), 20-35. Furmark, T., Tillfors, M., Marteinsdottir, I., Fischer, H., Pissiota, A., Langstrom, B., & Fredrikson, M. (2002). Common changes in cerebral blood flow in patients with social phobia treated with citalopram or cognitive-behavioral therapy. Archives of General Psychiatry, 59(5), 425-433. Fyer, A.J., Mannuzza, S., Chapman, T.F., Liebowitz, M.R., & Klein, D.F. (1993). A direct interview family study of social phobia. Archives of General Psychiatry, 50(4), 286-293. Gabbard, G.O. (1992). Psychodynamics of panic disorder and social phobia. Bulletin of the Menninger Clinic, 56(2 Suppl A), A3-13. Gelernter, C.S., Uhde, T.W., Cimbolic, P., Arnkoff, D.B., Vittone, B.J., Tancer, M.E., & Bartko, J.J. (1991). Cognitive-behavioral and pharmacological treatments of social phobia. A controlled study. Archives of General Psychiatry, 48(10 ), 938-45. Gilkinson, H. (1942). Social fears as reported by students in college speech classes. Speech Monographs, 9, 141-160. Gorman, J.M., Papp, L.A., Martinez, J., Goetz, R.R., Hollander, E., Liebowitz, M.R., & Jordan, F. (1990). High-dose carbon dioxide challenge test in anxiety disorder patients. Biological Psychiatry, 28(9), 743-57. Gould, Buckminster, S., Pollack, M.H., Otto, M.W., Liang, Y. (1997). Cognitive- behavioral and pharmacological treatment for social phobia; a meta-analysis. Clinical Psychology: Science and Practice, 4, 291-306.
88 Gould, R.A., Otto, M.W., & Pollack, M.H. (1995). A meta-analysis of treatment outcome for panic disorder. Clinical Psychology Review, 15, 819-844. Greenberg, P.E., Sisitsky, T., Kessler, R.C., Finkelstein, S.N., Berndt, E.R., Davidson, J.R.T., Ballenger, J.C., & Fyer, A.J. (1999). The economic burden of anxiety disorders in the 1990s. Journal of Clinical Psychiatry, 60(7), 427- 435. Harvey, A.G., Clark, D.M., Ehlers, A., & Rapee, R.M. (2000). Social anxiety and self-impression: cognitive preparation enhances the beneficial effects of video feedback following a stressful social task. Behaviour Research and Therapy, 38(12), 1183-92. Heimberg, R.G. (2001). Current status of psychotherapeutic interventions for social phobia. Journal of Clinical Psychiatry, 62 Suppl 1, 36-42. 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(3), 172-9. Heimberg, R.G., Horner, K.J., Juster, H.R., Safren, S.A., Brown, E.J., Schneier, F.R., & Liebowitz, M.R. (1999). Psychometric properties of the Liebowitz Social Anxiety Scale. Psychological Medicine, 29(1), 199-212. Heimberg, R.G., Juster, H.R., Hope, D.A., & Mattia J.I. (1995). Cognitive behavioral group treatment: description, case presentation, and empirical support. In M. B. Stein (Ed.), Social Phobia: Clinical and Research Perspectives. (pp. 293-321). Washington, DC: American Psychiatric Press. Heimberg, R.G., Liebowitz, M.R., Hope, D.A., Schneier, F.R., Holt, C.S., Welkowitz, L.A., Juster, H.R., Campeas, R., Bruch, M.A., Cloitre, M., Fallon, B., & Klein, D.F. (1998). Cognitive behavioral group therapy vs phenelzine therapy for social phobia: 12-week outcome. Archives of General Psychiatry, 55(12), 1133-41. Heimberg, R.G., Stein, M.B., Hiripi, E., & Kessler, R.C. (2000). Trends in the prevalence of social phobia in the United States: a synthetic cohort analysis of changes over four decades. European Psychiatry, 15(1), 29-37. Herbert, J.D., Hope, D.A., & Bellack, A.S. (1992). Validity of the distinction between generalized social phobia and avoidant personality disorder. Journal of Abnormal Psychology, 101(2), 332-9. Hofmann, S.G. (2000). Self-focused attention before and after treatment of social phobia. Behaviour Research and Therapy, 38(7), 717-25. Hofmann, S.G., Barlow, D.H., Papp, L.A., Detweiler, M.F., Ray, S.E., Shear,
89 M.K., Woods, S.W., & Gorman, J.M. (1998). Pretreatment attrition in a comparative treatment outcome study on panic disorder. American Journal of Psychiatr, 155, 43–47. Hofmann, S.G., & DiBartolo, P.M. (2000). An instrument to assessment self- statements during public speaking: scale development and preliminary psychometric properties. Behavior Therapy, 31, 499-515. Hofmann, S.G., Ehlers, A., & Roth, W.T. (1995). Conditioning theory: a model for the etiology of public speaking anxiety? Behaviour Research and Therapy, 33(5), 567-71. Hofmann, S.G., Newman, M.G., Ehlers, A., & Roth, W.T. (1995). Psychophysiological differences between subgroups of social phobia. Journal of Abnormal Psychology, 104( 1), 224-31. Hofmann, S.G. (2000). Treatment of social phobia: Potential mediators and moderators. Clinical Psychology: Science & Practice, 7(1), 3-16. Hope, D.A., Heimberg, R.G., & Bruch, M.A. (1995). Dismantling cognitive- behavioral group therapy for social phobia. Behaviour Research and Therapy, 33(6), 637-50. Huppert, J.D., Franklin, M.E., Foa, E.B., Davidson, J.R (2003). Study refusal and exclusion from a randomized treatment study of generalized social phobia. Journal of Anxiety Disorders, 17, 683-693. Jacobson, N. S., & Truax, P. (1991). Clinical significance: A statistical approach to defining meaningful change in psychotherapy research. Journal of Consulting and Clinical Psychology, 59, 12-19 Katzelnick, D.J., Kobak, K.A., DeLeire, T., Henk, H.J., Greist, J.H., Davidson, J.R., Schneier, F.R., Stein, M.B., & Helstad, C.P. (2001). Impact of generalized social anxiety disorder in managed care. American Journal of Psychiatry, 158(12), 1999-2007. Kendall, P. C., & Grove, W. M. (1988). Normative comparisons in therapy outcome. Behavioral Assessment, 10, 147-158. Kendler, K.S., Myers, J., Prescott, C.A., & Neale, M.C. (2001). The genetic epidemiology of irrational fears and phobias in men. Archives of General Psychiatry, 58(3), 257-65. Kendler, K.S., Neale, M.C., Kessler, R.C., Heath, A.C., & Eaves, L.J. (1992). The genetic epidemiology of phobias in women. The interrelationship of agoraphobia, social phobia, situational phobia, and simple phobia. Archives of General Psychiatry, 49(4), 273-81. Kessler, R.C., McGonagle, K.A., Zhao, S., Nelson, C.B., Hughes, M., Eshleman, 90 S., Wittchen, H.U., & Kendler, K.S. (1994). Lifetime and 12-month prevalence of DSM-III-R psychiatric disorders in the United States. Results from the National Comorbidity Survey. Archives of General Psychiatry, 51(1), 8-19. Kessler, R.C., Stein, M.B., & Berglund, P. (1998). Social phobia subtypes in the National Comorbidity Survey. American Journal of Psychiatry, 155(5), 613- 9. King, N.J., Clowes-Hollins, V., & Ollendick, T.H. (1997). The etiology of childhood dog phobia. Behaviour Research and Therapy, 35(1), 77 Kraemer, H.C., Wilson, G.T., Fairburn, C.G., & Agras, W.S. (2002). Mediators and moderators of treatment effects in randomized clinical trials. Archives of General Psychiatry, 59, 877-883. Lang, P.J. (1979). Presidential address, 1978. A bio-informational theory of emotional imagery. Psychophysiology, 16 (6), 495-512. Leary, M. (1983). A brief version of the Fear of Negative Evaluation Scale. Personality and Social Psychology Bulletin, 9, 371-375. Leung, A.W., & Heimberg, R.G. (1996). Homework compliance, perceptions of control, and outcome of cognitive- behavioral treatment of social phobia. Behaviour Research and Therapy, 34(5-6), 423-32. Liebowitz, M.R. (1987). Social phobia. Modern Problems of Pharmacopsychiatry, 22, 141-73. Liebowitz, M.R., Heimberg, R.G., Fresco, D.M., Travers, J., & Stein, M.B. (2000). Social phobia or social anxiety disorder: what's in a name? Archives of General Psychiatry, 57(2), 191-2. Liebowitz, M.R., Heimberg, R.G., Schneier, F.R., Hope, D.A., Davies, S., Holt, C.S., Goetz, D., Juster, H.R., Lin, S.H., Bruch, M.A., Marshall, R.D., & Klein, D.F. (1999). Cognitive-behavioral group therapy versus phenelzine in social phobia: long-term outcome. Depression and Anxiety, 10(3), 89-98. Lipsitz, J.D., Markowitz, J.C., Cherry, S., & Fyer, A.J. (1999). Open trial of interpersonal psychotherapy for the treatment of social phobia. American Journal of Psychiatry, 156(11), 1814-6. Magee, W.J., Eaton, W.W., Wittchen, H.U., McGonagle, K.A., & Kessler, R.C. (1996). Agoraphobia, simple phobia, and social phobia in the National Comorbidity Survey. Archives of General Psychiatry, 53(2), 159-68. Mannuzza, S., Schneier, F.R., Chapman, T.F., Liebowitz, M.R., Klein, D.F., & Fyer, A.J. (1995). Generalized social phobia. Reliability and validity. Archives of General Psychiatry, 52(3), 230-7. 91 Mazumdar, S., Liu, K.S., Houck, P.R., Reynolds, C.F. (1999). Intent-to-treat analysis for longitudinal clinical trials: Coping with the challenge of missing values. Journal of Psychiatric Research, 33(2), 87-95. McNeil, D.W., Ries, B.J., Taylor, L.J., Boone, M.L., Carter, L.E., Turk, C.L., & Lewin, M.R. (1995). Comparison of Social Phobia Subtypes Using Stroop Tests. Journal of Anxiety Disorders, 9(1), 47-57. Melchior, M.A., & Cheek, J.M. (1990). Shyness and anxious self-preoccupation during a social interaction. Journal of Social Behavior and Personality, 5, 117-130. Mennin, D.S., Fresco, D.M., Heimberg, R.G., Schneier, F.R., Davies, S.O., & Liebowitz, M.R. (2002). Screening for social anxiety disorder in the clinical setting: using the Liebowitz Social Anxiety Scale Journal of Anxiety Disorders, 16(4), 661-673. Mennin, D.S., Heimberg, R.G., & Jack, M.S. (2000). Comorbid generalized anxiety disorder in primary social phobia: symptom severity, functional impairment, and treatment response. Journal of Anxiety Disorders, 14(4), 325-43. Merckelbach, H., de Ruiter, C., van den Hout, M.A., & Hoekstra, R. (1989). Conditioning experiences and phobias. Behaviour Research and Therapy, 27(6), 657-62. Mowrer, O.H. (1939). A stimulus response theory. Psychological Review, 46, 553-565. Neufeld, K.J., Swartz, K.L., Bienvenu, O.J., Eaton, W.W., & Cai, G. (1999). Incidence of DIS/DSM-IV social phobia in adults. Acta Psychiatrica Scandinavica , 100(3), 186-92. Newman, M.G., Hofmann, S.G., Trabert, W., Roth, W.T., & et al. (1994). Does behavioral treatment of social phobia lead to cognitive changes? Behavior Therapy , 25(3), 503-517. Oakman, J., van Ameringen, M., Mancini, C., &Farvolden, P. (2003). A confirmatory factor analyses of a self-report version of the Liebowitz Social Anxiety Scvale. Journal of Clinical Psychology, 59, 149-161. Ohman, A., & Dimberg, U. (1978). Facial expressions as conditioned stimuli for electrodermal responses: a case of "preparedness"? Journal of Personality and Social Psychology, 36(11), 1251-8. Ost, L.G. (1987). Applied relaxation: description of a coping technique and review of controlled studies. Behaviour Research and Therapy, 25(5), 397- 409.
92 Ost, L.G., & Hugdahl, K. (1981). Acquisition of phobias and anxiety response patterns in clinical patients. Behaviour Research and Therapy, 19(5), 439-47. Papp, L.A., Klein, D.F., Martinez, J., Schneier, F., Cole, R., Liebowitz, M.R., Hollander, E., Fyer, A.J., Jordan, F., & Gorman, J.M. (1993). Diagnostic and substance specificity of carbon-dioxide-induced panic. American Journal of Psychiatry, 150( 2), 250-7. Patel, A., Knapp, M., Henderson, J., & Baldwin, D. (2002). The economic consequences of social phobia. Journal of Affective Disorders, 68(2-3), 221- 233. Paul, G. (1966). Insight vs. desensitization in psychotherapy. Stanford, CA: Stanford University Press. Pavlov, I.P. (1927). Lectures on conditioned reflexes. New York: Liveright Publishing Corporation. Peters, L., & Andrews, G. (1995). Procedural validity of the computerized version of the Composite International Diagnostic Interview (CIDI-Auto) in the anxiety disorders. Psychological Medicine, 25(6), 1269-80. Pine, D.S., Klein, R.G., Coplan, J.D., Papp, L.A., Hoven, C.W., Martinez, J., Kovalenko, P., Mandell, D.J., Moreau, D., Klein, D.F., & Gorman, J.M. (2000). Differential carbon dioxide sensitivity in childhood anxiety disorders and nonill comparison group. Archives of General Psychiatry, 57(10), 960-7. Pollard, C.A., & Henderson, J.G. (1988). Four types of social phobia in a community sample. Journal of Nervous and Mental Disease, 176(7), 440-5. Powers, M.B., Smits, J.A.J., & Telch, M.J. (In Press). Disentangling the effects of safety-behavior utilization and availability during exposure-based treatment: A placebo-controlled study. Journal of Consulting & Clinical Psychology. Rachman, S. (1977). The conditioning theory of fear-acquisition: a critical examination. Behaviour Research and Therapy, 15(5), 375-87. Rachman, S. (1980). Emotional processing. Behaviour Research and Therapy, 18(1), 51-60. Rapee, R.M., & Hayman, K. (1996). The effects of video feedback on the self- evaluation of performance in socially anxious subjects. Behaviour Research and Therapy, 34(4), 315-22. Rapee, R.M., & Heimberg, R.G. (1997). A cognitive-behavioral model of anxiety in social phobia. Behaviour Research and Therapy, 35(8), 741-56. Rapee, R.M., & Lim, L. (1992). Discrepancy between self- and observer ratings of performance in social phobics. Journal of Abnormal Psychology, 101(4),
93 728-31. Reich, J., Goldenberg, I., Goisman, R., Vasile, R., & Keller, M. (1994). A prospective, follow-along study of the course of social phobia: II. Testing for basic predictors of course. Journal of Nervous and Mental Disease, 182(5), 297-301. Safren, S.A., Heimberg, R.G., Horner, K.J., Juster, H.R., Schneier, F.R., & Liebowitz, M.R. (1999). Factor structure of social fears: the liebowitz social anxiety scale. Journal of Anxiety Disorders, 13, 253-270. Safren, S.A., Heimberg, R.G., Brown, E.J., & Holle, C. (1997). Quality of life in social phobia. Depression and Anxiety, 4(3), 126-33. Safren, S.A., Heimberg, R.G., & Juster, H.R. (1997). Clients' expectancies and their relationship to pretreatment symptomatology and outcome of cognitive- behavioral group treatment for social phobia. Journal of Consulting and Clinical Psychology, 65(4), 694-8. Salkovskis, P.M. (1991). The importance of behaviour in the maintenance of anxiety and panic: A cognitive account. Special Issue: The changing face of behavioural psychotherapy. Behavioural Psychotherapy, 19(1), 6-19. Sanderson, W.C., DiNardo, P.A., Rapee, R.M., & Barlow, D.H. (1990). Syndrome comorbidity in patients diagnosed with a DSM-III-R anxiety disorder. Journal of Abnormal Psychology, 99(3), 308-12. Schneier, F.R., Heckelman, L.R., Garfinkel, R., Campeas, R., Fallon, B.A., Gitow, A., Street, L., Del Bene, D., & Liebowitz, M.R. (1994). Functional impairment in social phobia. Journal of Clinical Psychiatry, 55(8), 322-31. Schneier, F.R., Johnson, J., Hornig, C.D., Liebowitz, M.R., & Weissman, M.M. (1992). Social phobia. Comorbidity and morbidity in an epidemiologic sample. Archives of General Psychiatry, 49(4), 282-8. Schneier, F.R., Liebowitz, M.R., Abi-Dargham, A., Zea-Ponce, Y., Lin, S.H., & Laruelle, M. (2000). Low dopamine D(2) receptor binding potential in social phobia. American Journal of Psychiatry, 157(3), 457-9. Seligman, M.E.P. (1971). Phobias and preparedness. Behavior Therapy, 2, 307- 320. Sinha, S., Papp, L.A., & Gorman, J.M. (2000). How study of respiratory physiology aided our understanding of abnormal brain function in panic disorder. Journal of Affective Disorders, 61(3), 191-200. Simon, N.M., Otto, M.W., Smits, J.A.J., Colette Nicolaou, Reese, H.E., & Pollack, M.H. (In Press) Changes in anxiety sensitivity with pharmacotherapy for panic disorder. Journal of Psychiatric Research. 94 Sloan, T., & Telch, M.J. (2002). The effects of safety-seeking behavior and guided threat reappraisal on fear reduction during exposure: An experimental investigation. Behaviour Research & Therapy, 40(3), 235-251. Smits J.A.J., Powers M., Cho, Y., & Telch M. (In Press). Mechanism of change in cognitive-behavioral treatment of panic disorder: Evidence for the fear of fear mediational hypothesis. Journal of Consulting and Clinical Psychology. Smits, J.A.J., Telch, M.J., & Randall, P.K. (2002). An examination of the decline of fear and disgust during exposure-based treatment. Behaviour Research & Therapy, 40 (11), 1243-1253. Stein, M.B. (1998). Neurobiological perspectives on social phobia: from affiliation to zoology. Biological Psychiatry, 44(12), 1277-85. Stein, M.B., Chartier, M.J., Hazen, A.L., Kozak, M.V., Tancer, M.E., Lander, S., Furer, P., Chubaty, D., & Walker, J.R. (1998). A direct-interview family study of generalized social phobia. American Journal of Psychiatry, 155(1), 90-7. Stein, M.B., Chartier, M.J., Lizak, M.V., & Jang, K.L. (2001). Familial aggregation of anxiety-related quantitative traits in generalized social phobia: clues to understanding "disorder" heritability? American Journal of Medical Genetics, 105(1), 79-83. Stein, M.B., & Chavira, D.A. (1998). Subtypes of social phobia and comorbidity with depression and other anxiety disorders. Journal of Affective Disorders , 50 Suppl 1, S11-6. Stein, M.B., Fyer, A.J., Davidson, J.R., Pollack, M.H., & Wiita, B. (1999). Fluvoxamine treatment of social phobia (social anxiety disorder): a double- blind, placebo-controlled study. American Journal of Psychiatry, 156(5), 756-60. Stein, M.B., Jang, K.L., & Livesley, W.J. (2002). Heritability of social anxiety- related concerns and personality characteristics: a twin study. Journal of Nervous and Mental Disease, 190(4), 219-24. Stopa, L., & Clark, D.M. (1993). Cognitive processes in social phobia. Behaviour Research and Therapy, 31(3), 255-67. Stopa L, Clark DM. (2000). Social phobia and interpretation of social events. Behaviour Research and Therapy, 38, 273-283 Taylor, S. (1996). Meta-analysis of cognitive-behavioral treatments for social phobia. Journal of Behavior Therapy and Experimental Psychiatry, 27(1), 1- 9. Telch, M.J., Ilai, D., Valentiner, D., & Craske, M.G. (1994). Match-mismatch of 95 fear, panic and performance. Behaviour Research and Therapy, 32(7), 691- 700. Telch, M.J., Valentiner, D., & Bolte, M. (1994). Proximity to safety and its effects on fear prediction bias. Behaviour Research and Therapy, 32(7), 747- 51. Telch, M.J., Valentiner, D.P., Ilai, D., Petruzzi, D., & Hehmsoth, M. (2000). The facilitative effects of heart-rate feedback in the emotional processing of claustrophobic fear. Behaviour Research and Therapy, 38(4), 373-87. Thompson, R.F., & Spencer, W.A. (1966). Habituation: a model phenomenon for the study of neuronal substrates of behavior. Psychological Review, 73(1), 16-43. Tiihonen, J., Kuikka, J., Bergstrom, K., Lepola, U., Koponen, H., & Leinonen, E. (1997). Dopamine reuptake site densities in patients with social phobia. American Journal of Psychiatry, 154(2), 239-42. Tillfors, M., Furmark, T., Marteinsdottir, I., Fischer, H., Pissiota, A., Langstrom, B., & Fredrikson, M. (2001). Cerebral blood flow in subjects with social phobia during stressful speaking tasks: a PET study. American Journal of Psychiatry, 158( 8), 1220-6. Turk, C.L., Heimberg, R.G., Orsillo, S.M., Holt, C.S., Gitow, A., Street, L.L., Schneier, F.R., & Liebowitz, M.R. (1998). An investigation of gender differences in social phobia. Journal of Anxiety Disorders, 12(3), 209-23. Turner, S.M., Beidel, D.C., Borden, J.W., Stanley, M.A., & Jacob, R.G. (1991). Social phobia: Axis I and II correlates. Journal of Abnormal Psychology, 100(1 ), 102-6. Turner, S.M., Beidel, D.C., Cooley, M.R., Woody, S.R., & Messer, S.C. (1994). A multicomponent behavioral treatment for social phobia: social effectiveness therapy. Behaviour Research and Therapy, 32(4), 381-90. 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(4), 389-94. Turner, S.M., Beidel, D.C., & Jacob, R.G. (1994). Social phobia: a comparison of behavior therapy and atenolol. Journal of Consulting and Clinical Psychology , 62(2), 350-8. Turner, S.M., Beidel, D.C., & Larkin, K.T. (1986). Situational determinants of social anxiety in clinic and nonclinic samples: physiological and cognitive correlates. Journal of Consulting and Clinical Psychology, 54(4), 523-7. Turner, S.M., Beidel, D.C., & Townsley, R.M. (1992). Social phobia: a 96 comparison of specific and generalized subtypes and avoidant personality disorder. Journal of Abnormal Psychology, 101(2), 326-31. Turner, S.M., Beidel, D.C., Wolff, P.L., Spaulding, S., & Jacob, R.G. (1996). Clinical features affecting treatment outcome in social phobia. Behaviour Research and Therapy, 34(10), 795-804. Valentiner, D.P., Telch, M.J., Ilai, D., & Hehmsoth, M.M. (1993). Claustrophobic fear behavior: a test of the expectancy model of fear. Behaviour Research and Therapy, 31(4), 395-402. Van Velzen, C.J., Emmelkamp, P.M., & Scholing, A. (1997). The impact of personality disorders on behavioral treatment outcome for social phobia. Behaviour Research and Therapy, 35(10), 889-900. Watson, D., & Friend, R. (1969). Measurement of social-evaluative anxiety. Journal of Consulting and Clinical Psychology, 33(4), 448-57. Watson, J.B.&.R.R. (1920). Conditioned emotional responses. Journal of Experimental Psychology, 3, 1-14. Wells, A., Clark, D.M., & Ahmad, S. (1998). How do I look with my minds eye: perspective taking in social phobic imagery. Behaviour Research and Therapy, 36(6), 631-4. Wells, A., & Papageorgiou, C. (1998). Social phobia: Effects of external attention on anxiety, negative beliefs, and perspective taking. Behavior Therapy, 29(3), 357-370. Williams, S.L., Dooseman, G., & Kleinfield, E. (1984). Comparative effectiveness of guided mastery and exposure treatments for intractable phobias. Journal of Consulting and Clinical Psychology, 52(4), 505-18. Williams, S.L., Kinney, P.J., & Falbo, J. (1989). Generalization of therapeutic changes in agoraphobia: the role of perceived self-efficacy. Journal of Consulting and Clinical Psychology, 57(3), 436-42. Williams, S.L., Turner, S.M., & Peer, D.F. (1985). Guided mastery and performance desensitization treatments for severe acrophobia. Journal of Consulting and Clinical Psychology, 53(2), 237-47. Wittchen, H.U., Stein, M.B., & Kessler, R.C. (1999). Social fears and social phobia in a community sample of adolescents and young adults: prevalence, risk factors and co-morbidity. Psychological Medicine, 29(2), 309-23. Woody, S.R., Chambless, D.L., & Glass, C.R. (1997). Self-focused attention in the treatment of social phobia. Behaviour Research and Therapy, 35(2), 117- 29.
97 World Health Organization. (1997). Composite International Diagnostic Interview (CIDI) researcher's manual. Washington DC: American Psychiatric Press Inc. Zvolensky, M.J., & Eifert, G.H. (2001). A review of psychological factors/processes affecting anxious responding during voluntary hyperventilation and inhalations of carbon dioxide-enriched air. Clinical Psychology Review, 21(3), 375-400.
98 VITA
Jasper Antonius Smits was born in Nijmegen, the Netherlands on April 28, 1974, the son of Marianne Smits-van Grinsven en Anton Smits. After completing high school at Elshof College, Nijmegen, the Netherlands, he entered the University of Amsterdam in 1994. During spring and fall semesters in 1998, he attended the University of Texas at Austin as a psychology research intern. He received the degree of Doctorandus (Drs.) in Clinical Psychology from the University of Amsterdam in December 1998. In September 1999, he entered the Graduate School of the University of Texas at Austin, where he received his Master’s Degree (M.A.) in Psychology in August, 2001. During the academic year of 2003-2004, he completed a pre-doctoral clinical internship at Harvard Medical School and Massachusetts General Hospital.
Permanent Address: Southern Methodist University, Department of Psychology, Dedman College, P.O. Box 750442, Dallas, Texas 75275.
This dissertation was typed by the author.
99