Relations Among Symptoms of Social Phobia Subtypes, Avoidant Personality Disorder, Panic, and Depression

Available online at www.sciencedirect.com

Behavior Therapy 41 (2010) 2 – 13

www.elsevier.com/locate/bt

Relations Among Symptoms of Social Phobia Subtypes, Avoidant Personality Disorder, Panic, and Depression Shawn A. Carter Kevin D. Wu Northern Illinois University

This study's primary goal was to examine relations between symptoms of specific social phobia (SSP), generalized social phobia (GSP), avoidant personality disorder (APD), and panic and depression. Past research has suggested a single social phobia continuum in which SSP displays less symptom severity than GSP or APD. We found SSP symptoms correlated less strongly with depression but more strongly with panic relative to both GSP and APD symptoms. These findings challenge a unidimensional model of social phobia, suggesting a multidimensional model may be more appropriate. These findings also inform current research aimed at classifying mood and anxiety disorders more broadly by identifying that the different factors of fear versus distress appear to underlie different subtypes of social phobia.

SOCIAL PHOBIA AND AVOIDANT personality disorder (APD) were introduced as disorders in the third edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-III; American Psychiatric Association [APA], 1980). Although originally conceived as distinct conditions, the lines between these disorders have been blurred in subsequent DSM editions, starting with DSM-III-R's (APA, 1987) inclusion of a generalized subtype of social phobia (GSP) and revision of the APD criteria to

Address correspondence to Kevin D. Wu, Ph.D., 311 PsychologyComputer Science Bldg, Department of Psychology, Northern Illinois University, DeKalb, IL 60115; e-mail: [email protected]. 0005-7894/08/002–013$1.00/0 © 2008 Association for Behavioral and Cognitive Therapies. Published by Elsevier Ltd. All rights reserved.

reflect social fears and avoidance. In DSM-IV-TR (APA, 2000), APD and severe GSP are so closely related that the following statement summarizes their differential diagnosis: “There appears to be a great deal of overlap between Avoidant Personality Disorder and Social Phobia, Generalized Type, so much so that they may be alternative conceptualizations of the same or similar conditions” (APA, 2000, p. 720). In addition to significant overlap between social phobia and APD, researchers have found broad overlap among all of the anxiety disorders and also between anxiety and mood disorders (e.g., T. A. Brown, Campbell, Lehman, Grisham, & Mancill, 2001; T. A. Brown, Chorpita, & Barlow, 1998 Clark & Watson, 1991; Mineka, Watson, & Clark, 1998; Watson, 2005). This led to the proposal of a tripartite model in which general distress explains the relation between depression and anxiety, whereas physiological hyperarousal and anhedonia differentiate anxiety from depression, respectively (Clark & Watson, 1991). More recently there has been support for a model that divides internalizing disorders across two factors – fear and distress (Cox, Clara, & Enns, 2002; Kendler, Prescott, Myers, & Neale 2003; Krueger, 1999; Slade & Watson, 2006; Vollebergh et al., 2001; Watson, 2005). In this latest model, major depressive disorder, dysthymic disorder, generalized anxiety disorder, and posttraumatic stress disorder load primarily on the distress factor, whereas panic disorder, agoraphobia, obsessive-compulsive disorder, specific phobia, and social phobia (conceptualized as a single construct, as opposed to separate generalized and nongeneralized subtypes) load primarily on the fear factor. One clear lesson from this literature has been that in order to fully conceptualize a condition such as social phobia,

social phobia one must understand its relations with other disorders.

Social Phobia and Avoidant Personality Disorder Substantial research on social phobia has focused on ways to best classify its proposed subtypes and relations with APD. However, this research has not always been consistent across studies. Whereas some researchers have followed DSM by defining social phobia subtypes based on the number of feared social situations, others have defined subtypes based on type of feared situations. Further, there has been an inconsistent use of terminology in describing individuals who suffer from social anxiety but do not meet full criteria for GSP. Different names for this construct that have appeared throughout the research literature include nongeneralized (E. J. Brown, Heimberg, & Juster, 1995; Holt, Heimberg, & Hope, 1992), circumscribed (Boone et al., 1999; Ries et al., 1998), discrete (Levin et al., 1993; Schneier, Spitzer, Gibbon, Fyer, & Liebowitz, 1991), and specific social phobia (SSP; Stemberger, Turner, Beidel, & Calhoun, 1995; Tran & Chambless, 1995; Turner, Beidel, Townsley, 1992). Consistent with Hook and Valentiner (2002), we use the term SSP to indicate a fear of a limited number of social situations. In an early study that examined differences between social phobia subtypes, Heimberg, Hope, Dodge, and Becker (1990) identified individuals who either feared (a) certain circumscribed social situations or (b) most or all social situations. Although the number of feared situations formed the basis of this distinction, it was found that the group who only feared circumscribed situations consisted predominantly of individuals whose only social fear was public speaking. This finding is consistent with other research that has considered that in addition to number of situations feared these two subtypes also differ as to the nature of the feared situations (Schneier et al., 1991). In fact, Turner et al. (1992) argued that SSP is defined by fears of performance situations, such as public speaking, eating, or writing, whereas GSP is defined by fears of interpersonal interactions, although individuals with GSP often display fears of performance situations as well. Although this classification is not consistent with the DSM, there is evidence, as outlined below, that in addition to a simple severity distinction there may be other meaningful differences between proposed social phobia subtypes (e.g., Hook & Valentiner, 2002; Stein & Chavira, 1998; Stemberger et al., 1995). Turning to the relation between social phobia and APD, researchers have proposed a unidimensional

3

model (also referred to as a social phobia continuum) with SSP on the least severe end, followed by GSP, and then APD on the most severe end (Boone et al., 1999; Herbert, Hope, & Bellack, 1992; Hofmann, Newman, Becker, Taylor, & Roth, 1995; Turner et al., 1992). Therefore, the unidimensional model suggests that SSP, GSP, and APD are similar conditions with symptom severity and number of situations feared being the primary criteria for classifying individuals suffering from social anxiety. Several studies support this possibility. For example, researchers consistently have found that individuals with either GSP or comorbid GSP and APD display greater amounts of overall distress, social anxiety, social impairment, fear of negative evaluation, and depression than individuals with SSP (Boone et al., 1999; E. J. Brown et al., 1995; Hofmann, Newman, Ehlers, & Roth, 1995; Holt et al., 1992; Tillfors, Furmark, Ekselius, & Fredrikson, 2004; Tran & Chambless, 1995; Turner et al., 1992). Conversely, individuals with GSP and individuals with GSP plus an additional diagnosis of APD have not consistently differed on these same measures across studies (Boone et al., 1999; E. J. Brown et al., 1995; Herbert et al., 1992; Huppert, Strunk, Ledley, Davidson, & Foa, 2008; Turner et al., 1992; Tillfors et al., 2004; Tran & Chambless, 1995; van Velzen, Emmelkamp, & Scholing, 2000). In a recent study using cluster analyses, a diagnosis of APD was found to significantly correlate with membership in a cluster displaying more severe symptomatology among individuals diagnosed with GSP with or without APD (Chambless, Fydrich, & Rodebaugh, 2008). However, once controlling for severity of social phobia symptoms, an APD diagnosis no longer correlated significantly with cluster membership, leading the researchers to conclude that APD is a more severe form of GSP, as opposed to a qualitatively distinct construct. Another study used confirmatory factor analysis to examine GSPdiagnosed individuals with or without APD and found that, even after removing two APD items that showed strong overlap with GSP, a two-factor GSP versus APD model displayed similar fit to a combined GSP/APD single-factor model (Huppert et al., 2008). Collectively, these studies suggest that GSP and APD are more pathological than SSP and, as suggested by DSM, potentially are redundant diagnoses. In addition to considerations of symptom severity, there are several other differences that suggest SSP and GSP/APD differ in their etiology and course and therefore may be distinct conditions. Previous studies have found that both GSP and APD generally have an earlier age of onset than SSP (E. J. Brown et al., 1995; Holt et al., 1992;

4

carter & wu

Oosterbaan, van Balkom, Spinhoven, de Meij, & van Dyck, 2002; Schneier et al., 1991); appear to show a higher degree of genetic heritability than SSP (Mannuzza et al., 1995; Tillfors, Furmark, Ekselius, & Fredrikson, 2001); are more often present in highly shy individuals than SSP (Chavira, Stein, & Malcarne, 2002); and are defined by a pattern of both increased neuroticism and decreased extraversion relative to SSP (Stemberger et al., 1995; Widiger & Costa, 1994). In fact, Stemberger et al. (1995) found that individuals with SSP did not differ from nonclinical controls on a measure of extraversion. At the same time, individuals with SSP, but not GSP, have reported more traumatic social experiences than adult nonpatients (Stemberger et al., 1995). However, this study did not report whether the traumatic events were experienced before or after the onset of SSP, leaving questions of causality. Together these finding suggest that GSP and APD are more genetically and personality-based, whereas SSP may be more environmentally influenced. As further evidence that a single severity continuum does not fully capture the nature of these conditions, numerous studies have found that individuals with SSP show an increased heart rate during a public speaking challenge (Boone et al., 1999; Heimberg et al., 1990; Hofmann, Newman, Ehlers et al., 1995; Levin et al., 1993). Conversely, individuals with GSP alone or comorbid GSP and APD do not differ from nonclinical controls on heart rate; however, they report elevated subjective anxiety during this task. Therefore, SSP has shown a distinct symptom profile compared to GSP and APD in at least one domain (i.e., elevated somatic fear response). Hook and Valentiner (2002) argued that these findings, in particular, were inconsistent with the unidimensional model, and that when considered in light of the above-described differences between SSP and GSP/ APD, a multidimensional model of social phobia may better explain the relations between these constructs. In this model, SSP is more closely related to the other anxiety disorders, particularly panic disorder and the specific phobias, whereas GSP is more closely related to APD and other personality disorders (Hook & Valentiner, 2002).

Relations between Social Phobia Subtypes and Other Disorders Consistent with Hook and Valentiner's (2002) multidimensional model of social phobia, Hofmann, Heinrichs, and Moscovitch (2004) theorized, but did not empirically test, that fearfulness (i.e., fear) and anxiousness (i.e., distress), can differentiate SSP from GSP and APD. In this model, fearfulness is defined by autonomic arousal in the presence of a phobic

stimulus and is hypothesized to be more closely related to SSP than GSP and APD. Anxiousness, on the other hand, is defined as an anticipatory cognitive response most often thought of as worry or distress, which is believed to be more closely related to GSP and APD than to SSP. In other words, SSP is predicted to be more related to other fear-based disorders, such as panic disorder, whereas GSP and APD are predicted to be more related to distress-based disorders, such as major depressive disorder. Evidence for this model can be inferred from Hughes et al. (2006), which sought to examine relations between social phobia and factors of the tripartite model of anxiety and depression (Clark & Watson, 1991) in social phobia patients. They found that two measures of performance anxiety (i.e., SSP) – the Social Phobia Scale (SPS; Mattick & Clarke, 1998) and the Fear of Performance subscale of the Liebowitz Social Anxiety Scale (LSAS; Liebowitz, 1987) – were more highly correlated with measures of panic – the Anxiety Sensitivity Index (ASI; Peterson & Reiss, 1992) and the Mood and Anxiety Symptom Questionnaire: Anxious Arousal (MASQ-AA; Watson & Clark, 1991) – than were measures of interaction anxiety (i.e., GSP) –the Social Interaction Anxiety Scale (SIAS; Mattick & Clarke, 1998) and the Fear of Social Interaction subscale of the LSAS (Liebowitz, 1987). Conversely, the interaction anxiety measures correlated higher than the performance anxiety measures with a measure of depression – the Mood and Anxiety Symptom Questionnaire: Anhedonic Depression (MASQ-AD; Watson & Clark, 1991). Although APD was not specifically assessed in this study, since GSP and APD are highly related if not identical constructs, it seems reasonable to predict that a measure of APD may converge with measures of interaction anxiety to produce the same pattern of results.

The Current Study The unidimensional model holds that both GSP and APD are simply more severe forms of social phobia than SSP and thus should display more severe symptomatology across all domains, including distress, depression, and anxiety. However, according to Hook and Valentiner (2002), a serious threat to the unidimensional model would be incurred if SSP was found to correlate with more severe symptoms than GSP/APD or produced a symptom profile that was distinct from GSP/APD on grounds other than simply relatively lower severity. The purpose of the current study was to examine for such possibilities in a large student sample. Although it is important to target clinical samples, student data are useful when examining relations among dimensional constructs

social phobia because they capture a broad range of symptom severity. That is, a student sample is likely to include individuals at points all along the severity continuum, including those who report symptoms in the clinically-elevated range. This approach allowed us to determine if the conclusions drawn from Hughes et al.'s (2006) social phobia sample replicated within a quite different and broader sample. Another goal of this study was to extend the findings of Hughes et al. (2006) by assessing APD symptoms in order to determine where they fit within this model. Following from the model, symptoms of SSP were predicted to correlate more strongly than GSP and APD symptoms with symptoms of panic, whereas both GSP and APD symptoms were predicted to correlate more strongly than SSP symptoms with symptoms of depression. It was further hypothesized that after controlling for general distress, panic symptoms would predict SSP symptoms but not GSP or APD symptoms, whereas symptoms of depression would predict both GSP and APD symptoms but not SSP symptoms. Together such findings would support a model of social phobia in which SSP and GSP/APD display different patterns of symptom severity that are not captured by a single social phobia continuum. One critical implication of such a finding is that support for a multidimensional model would challenge the practice of defining social phobia as a single construct as opposed to two distinct subtypes. Within state-of-the-art classification models that delineate the mood and anxiety disorders across fear/distress factors, such data would suggest that the two social phobia subtypes may be better conceptualized as falling on different sides of this model.

Methods participants Participants were 490 undergraduate students at Northern Illinois University who completed this study as partial fulfillment of a research exposure requirement. The sample was 60.6% female and 72.7% Caucasian (9.1% Black, 6.9% Asian, 6.1% Hispanic, .4% Native American, and 4.9% “other”). Ages ranged from 18 to 40 (Median = 18.0, M = 18.7, SD = 1.6). measures The social phobia scale and social interaction anxiety scale The SPS and SIAS (Mattick & Clarke, 1998) were designed in parallel to measure anxiety related to being observed by others and anxiety experienced while interacting with others, respectively. Each

5

measure contains 20 items rated on a 5-point scale ranging from 0 (not at all characteristic or true of me) to 4 (extremely characteristic or true of me). Previous research has shown these scales to demonstrate good internal consistency as measured by Cronbach's alphas ranging from .89 to .94 for the SPS and .88 to .94 for the SIAS in both clinical and nonclinical samples (Mattick & Clarke, 1998). Tests of convergent validity have shown the SPS and SIAS to be highly correlated (r = .72; Mattick & Clarke, 1998). Similarly, the SPS has been shown to correlate highly with the Fear of Performance subscale of the LSAS (r = .75); the SIAS was found to correlate highly with the Fear of Social Interaction subscale of the LSAS (r = .77; Hughes et al., 2006). The SPS and SIAS both have been shown to discriminate between individuals with social phobia and nonclinical controls (Heimberg, Mueller, Holt, Hope, & Liebowitz, 1992) and between individuals with social phobia and individuals with other anxiety disorders (Peters, 2000). Both instruments also have been shown to be sensitive to treatment change following cognitive-behavioral therapy for individuals diagnosed with SSP, GSP, and GSP with comorbid APD (Ries et al., 1998). Whereas Osman, Gutierrez, Barrios, Kopper, and Chiros (1998) found that a two-factor model (in which items from each measure represent separate factors) displayed good fit in a student sample, Safren, Turk, and Heimberg (1998) found that the same two-factor solution provided a poor fit in a social phobia patient sample. Instead they proposed a three-factor solution based on the results of an itemlevel factor analysis. In this solution a majority of the items from the SIAS produced a fear of social interactions factor, whereas items from the SPS split across two factors: 1) fears of being observed by others and 2) fears of displaying anxiety symptoms in front of others. Recently, it has been suggested that the three reverse-scored items on the SIAS comprise their own factor and are distinct from the other 17 items of this instrument (Rodebaugh, Woods, & Heimberg, 2007). However, given (a) that the 17-item SIAS correlates near unity with the full 20-item version (r = .97 in the current sample) and (b) to remain consistent with the bulk of previously published research, the full 20-item versions of both the SIAS and SPS were used in the current study. The schedule for nonadaptive and adaptive personality version 2 The SNAP-2 (Clark, Simms, Wu, & Casillas, 2008) was designed to measure both normal-range personality traits and symptoms related to DSM-IV personality disorders. The full version of the SNAP-

6

carter & wu

2 contains 390 true-false items. These items represent 15 trait and temperament scales, 12 diagnostic scales, and 6 validity scales. The SNAP-2 scales generally have been shown to be internally consistent, stable over intervals of up to 14 months, and display good convergent and discriminant validity with related measures of personality and personality pathology in both student and psychiatric patient samples (Clark et al., 2008; Melley, Oltmanns, Turkheimer, 2002). For the current study, the 19 items comprising the Avoidant Personality Disorder (APD) scale were used to measure symptoms of APD. This scale has been shown to have good internal consistency (alphas = .81–.89) in both clinical and nonclinical samples and good stability over a mean retest period of 49 days (r = .84) in a nonclinical sample (Clark et al., 2008; Wu, Clark, & Watson, 2006). The SNAP−2: APD scale correlated highly with the original version (rs = .84–.85) in samples of community adults and psychiatric patients and with the Structured Interview for DSM-IV Personality (SIDP-IV; Pfohl, Blum, & Zimmerman, 1997; r = .76) in a patient sample (Clark et al., 2008). The mood and anxiety symptom questionnaire The MASQ (Watson & Clark, 1991) was designed to measure both the differences and similarities between symptoms of anxiety and depression. The 62-item short form of the MASQ (Watson & Walker, 1996) contains four subscales. Two of these subscales – Anxious Arousal (AA) and Anhedonic Depression (AD) – were designed to differentiate between symptoms of panic and depression, respectively. The other two subscales – General Distress: Anxiety (GD: A), and General Distress: Depression (GD:D) – measure general distress related primarily to either anxiety or depression. The differences between the two anxiety scales are that whereas the AA scale focuses on somatic sensations (e.g., Was trembling or shaking), the GD:A scale focuses more on anxious mood and nonspecific anxiety symptoms (e.g., Felt afraid). Similarly, the GD:D scale focuses more on depressed mood and nonspecific depression symptoms (e.g., Felt hopeless), whereas the AD scale focuses on low positive affect (e.g., Felt like there wasn't anything interesting or fun to do). Respondents rate to what degree they have experienced these symptoms during the past week on a 5-point scale ranging from 1 (not at all) to 5 (extremely). These scales have been shown to be internally consistent (alphas= .78–.93) in both clinical and nonclinical samples (Watson et al., 1995a, 1995b). Correlations among the subscales across all samples also revealed that the general distress subscales correlated highly with each other (rs = .61–.86) suggesting they are measuring similar constructs. In contrast, the AA and AD subscales were

more modestly related (rs = .25–.49) and therefore provide better discriminant validity. The body sensations questionnaire The BSQ (Chambless, Caputo, Bright, & Gallagher, 1984) was designed to measure fear of panic symptoms. The BSQ contains 17 items that are rated on a 5-point scale from 1 (not at all frightened by this sensation) to 5 (extremely frightened by this sensation). The BSQ is scored by calculating the mean response across all 17 items. Chambless et al. validated the BSQ using a clinical sample of 175 agoraphobic (with panic attacks) patients. The BSQ demonstrated high internal consistency as shown by a Cronbach's alpha of .87 and relative stability over a pretreatment median interval of 31 days (r= .67). The BSQ demonstrated good convergent validity when it was shown to correlate .67 with the Agoraphobic Cognitions Questionnaire (ACQ; Chambless et al.). The BSQ also showed adequate discriminant validity, correlating .36 with the Beck Depression Inventory (BDI; Beck, Rush, Shaw, & Emery, 1979). The anxiety sensitivity index revised version The ASI-R (Taylor & Cox, 1998) was designed to measure fear of anxiety-related symptoms. The ASI-R contains 36 items rated on a 5-point scale from 0 (very little) to 4 (very much). It was developed as a multidimensional measure of anxiety sensitivity, but studies have shown inconsistent factor structures for this instrument and there is no evidence to suggest that one set of subscales is superior to the others (Deacon & Abramowitz, 2006; Deacon, Abramowitz, Wood, & Tolin, 2003; Taylor & Cox, 1998; Zvolensky et al., 2003). Therefore, in the current study only the ASI-R total score is used. Previous research has found the ASI-R total score to be practically identical to the original ASI (r = .94) and internally consistent (alphas = .94–.95) in both clinical and nonclinical samples (Deacon & Abramowitz, 2006; Deacon et al., 2003; Taylor & Cox, 1998).

procedure After obtaining written informed consent for this IRB-approved protocol, students completed the questionnaires in small groups (generally 5–25 students per group). Administration took on average 45–60 minutes. Anonymity was ensured by the use of random 4-digit identification numbers that could not be linked to students' identifying information.

Results descriptive data Descriptive statistics for each scale are presented in Table 1. Both raw and standardized (T) values are

7

social phobia Table 1 Descriptive Statistics and Internal Consistency Reliability Scale

# of Items

Possible Range

M (SD)

T (SD)

% of Ss T N 65

Coefficient Alpha

Social Phobia Scale Social Interaction Anxiety Scale SNAP-2 Avoidant Personality Disorder MASQ General Distress: Anxiety MASQ General Distress: Depression MASQ Anhedonic Depression MASQ Anxious Arousal Body Sensations Questionnaire Anxiety Sensitivity Index – Revised

20 20 19 11 12 22 17 17 36

0–80 0–80 0–19 11–55 12–60 22–110 17–86 1–5 0–144

17.3 (13.1) 23.4 (12.9) 5.6 (4.8) 21.4 (7.6) 25.5 (9.9) 56.9 (14.2) 28.0 (10.5) 2.0 (0.8) 28.8 (22.4)

53.1 54.3 49.3 48.5 50.4 51.6 50.8 51.4 51.7

18.8% 18.0% 15.1% 11.0% 12.0% 9.1% 12.4% 14.9% 13.9%

.92 .91 .89 .86 .92 .90 .90 .94 .95

(12.8) (12.8) (12.0) (12.1) (11.3) (10.3) (12.0) (13.4) (11.7)

Note. N = 484–490. SNAP = Schedule for Nonadaptive and Adaptive Personality. MASQ = Mood and Anxiety Symptom Questionnaire. T scores for the current sample were calculated using Ms and SDs from previously published undergraduate samples – Social Phobia Scale and Social Interaction Anxiety Scale (Mattick & Clarke, 1998); SNAP (Wu et al., 2006); MASQ (Watson et al., 1995a); Body Sensations Questionnaire (Schmidt, Kotov, Lerew, Joiner, & Ialongo, 2005); and Anxiety Sensitivity Index – Revised (Deacon et al., 2003).

social phobia and APD measures confirms that all three are highly related. The SPS-SIAS correlation (r = .75) is the highest value between any two instruments in the matrix. Using Meng, Rosenthal, and Rubin's (1992) equation for testing the significance of a contrast between two correlated correlation coefficients, the SPS-SIAS correlation was significantly higher than the SPS-APD correlation (z = 5.17, p b .001) and the SIAS-APD correlation (z = 3.94, p b .001). However, note that the SPS and SIAS both displayed their second highest correlation with APD (rs = .62 and .65, respectively); these values were not significantly different (z = 1.27, p = .10). Turning to intercorrelations among MASQ subscales, the results showed the expected pattern. The two general distress scales (GD:A and GD:D)

presented. T-score means for the current sample ranged from 48.5 to 54.3, indicating that all of the mean scores fell within .5 SD of the respective normative means (i.e., within 5 T-score points of 50). However, across all measures a substantial percentage of participants (9.1 – 18.8%) displayed scores greater than 1.5 SD above the mean of the normative sample (i.e., TN 65). Regarding reliability, all of the instruments showed good internal consistency with coefficient alphas ranging from .86 to .95. These values are consistent with previous research reporting these measures and support their reliability in the current sample.

zero-order correlations Zero-order correlations among all measures are presented in Table 2. Correlations among the Table 2 Zero-order Correlations Among Measures Scale

1

2

3

4

5

6

7

Measures of Social Phobia Subtypes and APD 1. Social Phobia Scale 2. Social Interaction Anxiety Scale 3. SNAP-2 Avoidant Personality Disorder

––– .75 .62a

––– .65a

–––

Mood and Anxiety Symptom Questionnaire Scales 4. General Distress: Anxiety 5. General Distress: Depression 6. Anhedonic Depression 7. Anxious Arousal

.50 .50 .38c .47a

.38 .42 .43b .33b

.48 .60 .59a .42a

––– .79 .38 .81

––– .58 .66

––– .33

–––

Additional Measures of Panic 8. Body Sensations Questionnaire 9. Anxiety Sensitivity Index- Revised

.45a .60a

.32b .46b

.36b .46b

.57 .60

.50 .53

.24 .33

.55 .62

Note. N = 484–490. All correlations significant, p b .001. Different superscripts indicate a significant difference within-row across columns 1, 2, and 3 (correlated correlations z-test, p b .05). APD = Avoidant Personality Disorder. SNAP = Schedule for Nonadaptive and Adaptive Personality.

8

––– .72

8

carter & wu

were highly correlated (r = .79). Anxious Arousal correlated higher with GD:A (r = .81) than with GD:D (r = .66; z = 8.15, p b .001) and Anhedonic Depression correlated higher with GD:D (r = .58) than with GD:A (r = .38; z = 7.87, p b .001). Importantly, the lowest intercorrelation among the MASQ scales was between AA and AD (r = .33), reflecting the primary strength of this instrument. A series of correlated correlations z-tests (provided by Meng et al., 1992) was used to test the hypotheses that (a) the SPS was more strongly related to panic symptoms than were the SIAS and APD and (b) the SIAS and APD were more strongly related to depression symptoms than was the SPS. Regarding panic, results were that compared to the SIAS, the SPS correlated significantly higher with all three panic measures, the MASQ-AA (z = 4.80, p b .001), the BSQ (z = 4.44, p b .001), and the ASI-R (z = 5.29, p b .001). Compared to APD, the SPS correlated significantly higher with two of the three panic measures, the BSQ (z = 2.54, p b .01) and ASIR (z = 4.35, p b .001); the contrast with MASQ-AA did not reach significance (z = 1.44, p = .07). Conversely, there was not a significant difference between the SIAS and APD on two of the three panic measures, the BSQ (z = 1.13, p = .13) and ASI-R (z = 0.00, p = .50); APD correlated significantly higher with MASQ-AA than did the SIAS (z = 2.58, p b .01). Overall it was found that the correlation between the SPS and panic symptoms in general, as calculated by a mean correlation with the BSQ, ASI-R, and MASQ-AA (M = .51), was significantly higher than the correlations between panic and either the SIAS (M = .37; z = 4.92, p b .001) or APD (M = .41; z = 2.91, p b .01), whereas there was not a significant difference between panic and the SIAS or

APD (z = 1.16, p = .12). Turning to relations with depression, the SPS correlated significantly lower with MASQ-AD than did the SIAS (z = 1.72, p b .05) and APD (z = 6.26, p b .001). In fact, the SPS showed its lowest correlation in Table 2 with depression. Worth noting, APD correlated significantly higher with MASQ-AD than did the SIAS (z = 5.05, p b .001). To further highlight this pattern of relations, the SPS correlated significantly higher with MASQAA than with MASQ-AD (z = 1.96, p b .05); the SIAS and APD both showed significantly stronger relations with MASQ-AD than with MASQ-AA (z = 2.11, p b .05 and z = 3.99, p b .001, respectively).

partial correlations Because all Table 2 symptoms involve general distress (rs ranged from .38 to .81), and because such variance may obscure more specific associations, we next examined partial correlations controlling for the MASQ general distress scales (Table 3). The overall pattern remained similar to Table 2, but this method highlighted correlations from Table 2 that extend beyond the influence of only or primarily general distress. Again using Meng et al. (1992), the main findings were that (a) the SPS correlated significantly higher than the SIAS with MASQ-AA (z = 2.52, p b .01), BSQ (z = 2.83, p b .01), and ASI-R (z = 3.92, p b .001); (b) the SPS correlated significantly higher than APD with the BSQ (z = 2.81, p b .01) and ASI-R (z = 4.57, p b .001), whereas the contrast with MASQ-AA did not reach significance (z = 1.50, p = .07); (c) there was no significant difference between the SIAS and APD on MASQ-AA (z = 0.46, p = .32) or BSQ (z = 0.70, p = .24), whereas the contrast with the ASI-R approached significance (z = 1.68, p = .05); (d) the

Table 3 Correlations Among Measures Partialling MASQ General Distress Scales Scale

1

2

3

Measures of Social Phobia Subtypes and APD 1. Social Phobia Scale 2. Social Interaction Anxiety Scale 3. SNAP-2 Avoidant Personality Disorder

4

5

––– .69 .47b

––– .55a

–––

Mood and Anxiety Symptoms Questionnaire Scales 4. Anhedonic Depression 5. Anxious Arousal

.16c .12a

.27b .03b

.38a .05a,b

––– .02

–––

Additional Measures of Panic 6. Body Sensations Questionnaire 7. Anxiety Sensitivity Index- Revised

.21a .42a

.11b .29b

.08b .22b

−.02 .10

.19 .28

Note. N = 484. Bolded correlations significant, p b .01. Different superscripts indicate a significant difference within-row across columns 1, 2, and 3 (correlated correlations z-test, p b .05). MASQ = Mood and Anxiety Symptom Questionnaire. APD = Avoidant Personality Disorder. SNAP = Schedule for Nonadaptive and Adaptive Personality.

6

––– .57

9

social phobia SPS correlated significantly higher with the mean of the three panic markers (M = .25) than did the SIAS (M = .14; z = 3.13, p b .001) or APD (M = .12; z = 2.83, p b .01), whereas there was not a significant difference between panic and the SIAS or APD (z = 0.47, p = .32); and (e) the SPS correlated significantly lower with MASQ-AD than did the SIAS (z = 3.50, p b .001) and APD (z = 5.80, p b .001), whereas APD correlated significantly higher than the SIAS with MASQ-AD (z = 3.08, p b .01). Furthermore, with general distress controlled, four correlations involving either SIAS or APD and panic symptoms (SIAS/MASQ-AA, APD/MASQ-AA, SIAS/BSQ, and APD/BSQ) no longer were significantly different from zero. Thus, it was not simply that the correlations were merely lower than those between SPS and panic, but were in fact low in an absolute sense. Another finding from Table 3 is that with general distress controlled, the SIAS showed a significantly stronger correlation with APD than did the SPS (z = 2.68, p b .01). This offers a direct correlational test of the suggestion that interaction anxiety (i.e., GSP) and APD are more closely related than are performance anxiety (i.e., SSP) and APD, despite the very strong correlation between the available measures of these subtypes.

hierarchical regressions To determine the degree to which the two types of general distress (anxiety and depression), and symptoms of panic and depression predicted the SPS, SIAS, and APD after controlling for general distress, a series of hierarchical regressions was conducted (Table 4). For the SPS, the two MASQ general distress scales were entered in step 1. This produced a significant result (F2,481 = 94.94, p b .001) and an R2 of .28. Individually both MASQGD:A (β = .28) and MASQ-GD:D (β = .28) significantly predicted SPS. The specific panic (MASQ-AA) and depression (MASQ-AD) scales were entered in step 2. Again, this produced a significant result (F2,479 = 9.59, p b .001) and an R2 change of .03. Both the MASQ-AA (β = .16) and MASQ-AD (β = .17) significantly predicted SPS. For the SIAS, the two MASQ general distress scales were entered in step 1 and produced a significant result (F2,481 = 53.99, p b .001) and an R2 of .18. Individually, only MASQ-GD:D (β = .32), but not MASQ-GD:A (β = .12), significantly predicted SIAS. MASQ-AA and MASQ-AD were entered in step 2 and also produced a significant result (F2,479 = 19.76, p b .001) with an R2 change of .06. However, only MASQ-AD (β = .31), but not MASQ-AA (β = .04), was a significant predictor of SIAS.

Table 4 Hierarchical Regressions Predicting Social Anxiety Symptoms B

SE B

Predicting Social Phobia Scale Step 1: MASQ-GD:A .48 .11 MASQ-GD:D .37 .08 Step 2: MASQ-AA .20 .08 MASQ-AD .15 .04

β

ΔR2

Test statistic

.28

F(2,481) = t(481) = t(481) = F(2,479) = t(479) = t(479) =

94.94⁎⁎ 4.43⁎⁎ 4.49⁎⁎ 9.59⁎⁎ 2.51⁎ 3.54⁎⁎

F(2,481) = t(481) = t(481) = F(2,479) = t(479) = t(479) =

53.99⁎⁎ 1.83ns 4.82⁎⁎ 19.76⁎⁎ 0.66ns 6.24⁎⁎

.28 .28 .03 .16 .17

Predicting Social Interaction Anxiety Scale Step 1: .18 MASQ-GD:A .21 .11 .12 MASQ-GD:D .42 .08 .32 Step 2: .06 MASQ-AA .05 .08 .04 MASQ-AD .28 .04 .31

Predicting SNAP-2 Avoidant Personality Disorder Step 1: .36 F(2,481) = MASQ-GD:A .01 .04 .01 t(481) = MASQ-GD:D .29 .03 .59 t(481) = Step 2: .09 F(2,479) = MASQ-AA .03 .03 .06 t(479) = MASQ-AD .13 .01 .38 t(479) =

134.30⁎⁎ 0.20ns 9.88⁎⁎ 40.48⁎⁎ 1.04ns 8.91⁎⁎

Note. N = 484. ns = Not Significant, p N .05. ⁎ p b .05. ⁎⁎p b .001. MASQ = Mood and Anxiety Symptom Questionnaire. GD:A = General Distress: Anxiety. GD:D = General Distress: Depression. AA = Anxious Arousal. AD = Anhedonic Depression. SNAP = Schedule for Nonadaptive and Adaptive Personality.

Turning to the APD scale, the two MASQ general distress scales were entered in step 1. This result was significant (F2,481 = 134.30, p b .001) with an R2 of .36. Individually, only MASQ-GD:D (β = .59), but not MASQ-GD:A (β = .01), significantly predicted APD. The specific panic and depression scales were entered in step 2 and produced a significant result (F2,479 = 40.48, p b .001) and an R2 change of .09. However, as for the SIAS, only MASQ-AD (β = .38), but not MASQ-AA (β = .06) significantly predicted APD.

Discussion Given the high correlation between the SPS and SIAS (r = .75 in the current sample), it would seem that finding any differential relations between these measures and other constructs would be unlikely. Moreover, given that SSP is thought to display less severe pathology than GSP or APD according to the unidimensional model of social phobia, one would expect that any observed differences would reflect lower levels of distress or symptomatology related to the SPS versus the SIAS or APD. However, not only

10

carter & wu

did we find differential relations involving SSP and GSP/APD with measures of panic and depression symptoms, but consistent with our hypotheses, the results were contrary to that predicted by the unidimensional model. Whereas APD showed the strongest association with depression symptoms followed by the SIAS and then the SPS, which is consistent with both the unidimensional and multidimensional model, the SPS showed an overall pattern of stronger relations with panic symptoms than did the SIAS or APD, which is consistent with a multidimensional model but not a unidimensional model. Furthermore, in a series of hierarchical regressions, although all four MASQ scales significantly predicted SPS scores, only the two depression scales (MASQ-GD:D and MASQ-AD) but not the two anxiety scales (MASQ-GD:A and MASQ-AA) significantly predicted SIAS and APD scores. This last finding generally confirmed our hypotheses; however, we did not expect MASQ-AD to significantly predict SPS. Still, the magnitude of association between the SPS and MASQ-AD (β = .17) was much weaker than between MASQ-AD and either the SIAS or APD (β = .31 and .38, respectively), which is consistent with our other predictions. Together these findings suggest that GSP/APD have stronger relations with depression than with panic and may be surprising given the current classification of GSP as an anxiety disorder. Implications of these findings are that social phobia may not be best conceptualized as a construct defined by a single social anxiety continuum. Instead, different types of social anxiety symptoms (i.e., performance and interaction anxiety) relate differentially to symptoms reflecting other disorders. Specifically, symptoms of performance anxiety or SSP display a distinct relation with panic whereas symptoms of interaction anxiety or GSP are more closely related to depression. Moreover, APD symptoms were found to be more closely related to GSP symptoms than SSP symptoms, not only in that the APD scale correlated significantly higher with the SIAS than the SPS once general distress had been controlled, but that both the SIAS and APD displayed a pattern of relations with symptoms of panic and depression that were more similar to one another than either was to the pattern of relations between the SPS and those constructs. Although the zero-order SIAS-SPS correlation was higher than the SIAS-APD correlation, this finding may be influenced by the fact that the SIAS and SPS (a) were designed to measure closely related symptoms of social anxiety, (b) were designed in parallel by the same authors, and (c) likely share a certain amount of method variance. Still, the overall pattern of our findings is more

consistent with a multidimensional model of social phobia in which SSP is related to fear-based disorders (e.g., panic), whereas GSP and APD are more related to distress-based disorders (e.g., depression), than a unidimensional model of social phobia in which GSP and APD are associated simply with greater symptomatology across all markers of pathology in comparison to SSP. In light of current attempts to classify the mood and anxiety disorders using fear versus distress factors, these results suggest that social phobia itself may be better conceptualized as two distinct symptom dimensions that split across the factors. Unfortunately, the available structural research in this area has (a) defined social phobia as a single construct and (b) not included APD. Therefore, further testing of this model is needed in which APD and the two social phobia subtypes are measured as separate variables in factor analyses to determine if GSP and APD load on the distress factor whereas SSP loads on the fear factor. The current findings are consistent with this possibility. Although the current study was successful in addressing its primary goals, it has limitations. First, participants included only students. As previously mentioned, given that the goal of this study was to examine symptom dimensions of social anxiety that encompass a wide range of severity, one of the benefits of student data is that they allow for the analysis of relations among these symptoms along the entire severity dimension of each of the constructs examined. This was certainly the case in the current sample. Some participants reported little to no symptomatology, whereas across all measures there was a substantial percentage of participants who scored in a clinically elevated range. In comparison, patient-only samples often suffer from range restriction due to severely negatively-skewed distributions and only allow for the examination of symptom relations at the upper end of the severity continuum (Olatunji, Williams, Haslam, Abramowitz, & Tolin, 2008). Still the use of student-only data has limitations in that college students possess characteristics that generally are not associated with individuals diagnosed with social phobia (e.g., higher school achievement, less functional impairment). To reduce these confounds, future studies may attempt replication of the current findings utilizing a community sample that is more demographically similar to patients with social phobia and/or APD. Future research would also benefit from reexamining these relations in a mixed nonclinical/clinical sample, including individuals with SSP, GSP and APD, to determine if they are consistent within a sample that has a more equal distribution of

social phobia individuals who possess both lower and higher degrees of symptomatology. Second, our data were collected using questionnaires. Future studies would benefit from including different methods, such as interviews or behavioral observation, to reduce concern over shared method variance. A related issue is that our method assessed symptom dimensions as opposed to formal diagnoses of specific clinical disorders. Extant research, including the current study, must be clear to note what phenomena are under investigation. That is, this method applies self-report questionnaires as surrogate markers of specific disorders because they assess symptom dimensions that map onto relevant DSM diagnostic criteria. However, no questionnaire or combination of questionnaires used in this study captures the full diagnostic criteria of the formal disorders. Therefore, one should not assume that responses on a dimensional symptom questionnaire are a direct substitute for a clinical diagnosis, especially given that in the current nonclinical sample a majority of participants would not meet full diagnostic criteria for any mental disorder. Said differently, a dimensional score—even a markedly elevated score—on (e.g.) the Social Phobia Scale is not synonymous with a diagnosis of (e.g.) social phobia. For the current study, it was precisely these dimensions that were under investigation and of primary interest; future research may focus on clinical diagnoses and comorbidity rates to suit different purposes or address different research questions. Still, questionnaires offer the benefits of collecting much larger samples and asking many more questions that cover a broader range of symptom severity than diagnoses alone. Therefore, another goal for future research will be to determine if these results are replicated using different questionnaires from those used in the current study. Despite our selection of well-validated instruments, different markers of distress, panic, and depression may perform differently and it will be important to know how robust these findings are across the available measures. Of particular note, we used the MASQ to assess three different constructs. The MASQ was chosen on the basis of its strong psychometric properties and specific ability to measure panic and depression separately; however, considering again the issue of shared method variance, it will be important to replicate these findings using separate instruments for each of the relevant constructs – provided that they demonstrate adequate levels of discriminant validity as the MASQ scales do. Lastly, given the sample selected and the use of self-report questionnaires, the current study was

11

limited in the type and number of constructs that could be examined. Future studies are needed to examine the possibility that there may be additional differential relations between these constructs and other disorders. For example, if SSP is a fear-based disorder, as the current study and others studies suggest, then one would also expect it to show stronger relations than GSP or APD with agoraphobia and the specific phobias. Conversely, if GSP and APD are distress-based disorders, then one would expect these constructs to show stronger relations than SSP with generalized anxiety disorder and dysthymic disorder. Future research should assess for additional relevant experiences/symptoms using multiple methods. Such research will help to examine possible differential relations between the social phobia subtypes and a host of other disorders, and in doing so, provide a fuller picture of the structure of SSP, GSP, and APD. References American Psychiatric Association. (1980). Diagnostic and statistical manual of mental disorders, (3rd ed.) Washington, DC: Author. American Psychiatric Association. (1987). Diagnostic and statistical manual of mental disorders, (3rd ed. rev.) Washington, DC: Author. American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disorders, (4th ed. text rev.) Author: Washington, DC. Beck, A. T., Rush, A. J., Shaw, B. F., & Emery, G. (1979). Cognitive therapy of depression. New York: Guilford Press. Boone, M. L., McNeil, D. W., Masia, C. L., Turk, C. L., Carter, L. E., Ries, B. J., et al. (1999). Multimodal comparisons of social phobia subtypes and avoidant personality disorder. Journal of Anxiety Disorders, 13, 271–292. 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, 467–486. Brown, T. A., Campbell, L. A., Lehman, C. L., Grisham, J. R., & Mancill, R. B. (2001). Lifetime comorbidity of DSM-IV anxiety and mood disorders in a large clinical sample. Journal of Abnormal Psychology, 110, 585–599. Brown, T. A., Chorpita, B. F., & Barlow, D. H. (1998). Structural relationships among dimensions of the DSM-IV anxiety and mood disorders and dimensions of negative affect, positive affect, and autonomic arousal. Journal of Abnormal Psychology, 107, 179–192. Chambless, D. L., Caputo, G. C., Bright, P., & Gallagher, R. (1984). Assessment of fear of fear in agoraphobics: The Body Sensations Questionnaire and the Agoraphobic Cognitions Questionnaire. Journal of Consulting and Clinical Psychology, 52, 1090–1097. Chambless, D. L., Fydrich, T., & Rodebaugh, T. L. (2008). Generalized social phobia and avoidant personality disorder: Meaningful distinction or useless duplication? Depression and Anxiety, 25, 8–19. Chavira, D. A., Stein, M. B., & Malcarne, V. L. (2002). Scrutinizing the relationship between shyness and social phobia. Journal of Anxiety Disorders, 16, 585–598.

12

carter & wu

Clark, L. A., Simms, L. J., Wu, K. D., & Casillas, A. (2008). Manual for the Schedule for Adaptive and Nonadaptive Personality—II. Minneapolis, MN: Univ. of Minnesota Press, 45(3), 249–254. Clark, L. A., & Watson, D. (1991). Tripartite model of anxiety and depression: Psychometric evidence and taxonomic implications. Journal of Abnormal Psychology, 100, 316–336. Cox, B. J., Clara, I. P., & Enns, M. W. (2002). Posttraumatic stress disorder and the structure of common mental disorders. Depression and Anxiety, 15, 168–171. Deacon, B., & Abramowitz, J. (2006). Anxiety sensitivity and its dimensions across the anxiety disorders. Journal of Anxiety Disorders, 20, 837–857. Deacon, B. J., Abramowitz, J. S., Woods, C. M., & Tolin, D. F. (2003). The Anxiety Sensitivity Index – Revised: Psychometric properties and factor structure in two nonclinical samples. Behaviour Research and Therapy, 41, 1427–1449. Heimberg, R. G., Hope, D. A., Dodge, C. S., & Becker, R. E. (1990). DSM-III-R subtypes of social phobia: Comparisons of generalized subtypes and public speaking phobics. Journal of Nervous and Mental Disease, 178, 172–179. Heimberg, R. G., Mueller, G. P., Holt, C. S., Hope, D. A., & Liebowitz, M. R. (1992). Assessment of anxiety in social interaction and being observed by others: The Social Interaction Anxiety Scale and the Social Phobia Scale. Behavior Therapy, 23, 53–73. 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, 332–339. Hofmann, S. G., Heinrichs, N., & Moscovitch, D. A. (2004). The nature and expression of social phobia: Toward a new classification. Clinical Psychology Review, 24, 769–797. Hofmann, S. G., Newman, M. G., Becker, E., Taylor, C. B., & Roth, W. T. (1995). Social phobia with and without avoidant personality disorder: Preliminary behavior therapy outcome findings. Journal of Anxiety Disorders, 9, 427–438. Hofmann, S. G., Newman, M. G., Ehlers, A., & Roth, W. T. (1995). Psychophysiological differences between subgroups of social phobia. Journal of Abnormal Psychology, 104, 224–231. Holt, C. S., Heimberg, R. G., & Hope, D. A. (1992). Avoidant personality disorder and the generalized subtype of social phobia. Journal of Abnormal Psychology, 101, 318–325. Hook, J. N., & Valentiner, D. P. (2002). Are specific and generalized social phobias qualitatively distinct? Clinical Psychology: Science and Practice, 9, 379–395. Hughes, A. A., Heimberg, R. G., Coles, M. E., Gibb, B. E., Liebowitz, M. R., & Schneier, F. R. (2006). Relations of the factors of the tripartite model of anxiety and depression to types of social anxiety. Behaviour Research and Therapy, 44, 1629–1641. Huppert, J. D., Strunk, D. R., Ledley, D. R., Davidson, J. R. T., & Foa, E. B. (2008). Generalized social anxiety disorder and avoidant personality disorder: Structural analysis and treatment outcome. Depression and Anxiety, 25, 441–448. Kendler, K. S., Prescott, C. A., Myers, J., & Neale, M. C. (2003). The structure of genetic and environmental risk factors for common psychiatric and substance use disorders in men and women. Archives of General Psychiatry, 60, 929–937. Krueger, R. F. (1999). The structure of common mental disorders. Archives of General Psychiatry, 56, 921–926. Levin, A. P., Saoud, J. B., Strauman, T., Gorman, J. M., Fyer, A. J., Crawford, R., et al. (1993). Responses of “generalized” and “discrete” social phobics during public speaking. Journal of Anxiety Disorders, 7, 207–221.

Liebowitz, M. R. (1987). Social phobia. Modern Problems in Pharmacopsychiatry, 22, 141–173. 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, 230–237. Mattick, R. P., & Clarke, J. C. (1998). Development and validation of measures of social phobia scrutiny fear and social interaction anxiety. Behaviour Research and Therapy, 36, 455–470. Melley, A. H., Oltmanns, T. F., & Turkheimer, E. (2002). The Schedule of Nonadaptive and Adaptive Personality (SNAP): Temporal stability and predictive validity of the diagnostic scales. Assessment, 9, 181–187. Meng, X. L., Rosenthal, R., & Rubin, D. B. (1992). Comparing correlated correlation coefficients. Psychological Bulletin, 111, 172–175. Mineka, S., Watson, D., & Clark, L. A. (1998). Comorbidity of anxiety and unipolar mood disorders. Annual Review of Psychology, 49, 377–412. Olatunji, B. O., Williams, B. J., Haslam, N., Abramowitz, J. S., & Tolin, D. F. (2008). The latent structure of obsessivecompulsive symptoms: A taxometric study. Depression and Anxiety, 25, 956–968. Oosterbaan, D. B., van Balkom, A. J. L. M., Spinhoven, P., de Meij, T. G. J., & van Dyck, R. (2002). The influence on treatment gain of comorbid avoidant personality disorder in patients with social phobia. Journal of Nervous and Mental Disease, 190, 41–43. Osman, A., Gutierrez, P. M., Barrios, F. X., Kopper, B. A., & Chiros, C. E. (1998). The Social Phobia and Social Interaction Anxiety Scales: Evaluation and psychometric properties. Journal of Psychopathology and Behavioral Assessment, 20, 249–264. Peters, L. (2000). Discriminant validity of the Social Phobia and Anxiety Inventory (SPAI), the Social Phobia Scale (SPS), and the Social Interaction Anxiety Scale (SIAS). Behaviour Research and Therapy, 38, 943–950. Peterson, R. A., & Reiss, S. (1992). Anxiety Sensitivity Index revised test manual. Worthington, OH: International Diagnostic Systems. Pfohl, B., Blum, N., & Zimmerman, M. (1997). Structured Interview for DSM-IV Personality. Washington DC: American Psychiatric Press. Ries, B. J., McNeil, D. W., Boone, M. L., Turk, C. L., Carter, L. E., & Heimberg, R. G. (1998). Assessment of contemporary social phobia verbal report instruments. Behaviour Research and Therapy, 36, 983–994. Rodebaugh, T. L., Woods, C. M., & Heimberg, R. G. (2007). The reverse of social anxiety is not always the opposite: The reverse-scored items of the Social Interaction Anxiety Scale do not belong. Behavior Therapy, 38, 192–206. Safren, S. A., Turk, C. L., & Heimberg, R. G. (1998). Factor structure of the Social Interaction Anxiety Scale and the Social Phobia Scale. Behaviour Research and Therapy, 36, 443–453. Schmidt, N. B., Kotov, R., Lerew, D. R., Joiner, T. E., & Ialongo, N. S. (2005). Evaluating latent discontinuity in cognitive vulnerability to panic: A taxometric investigation. Cognitive Therapy and Research, 29, 673–690. Schneier, F. R., Spitzer, R. L., Gibbon, M., Fyer, A. J., & Liebowitz, M. R. (1991). The relationship of social phobia subtypes and avoidant personality disorder. Comprehensive Psychiatry, 32, 496–502. Slade, T., & Watson, D. (2006). The structure of common DSMIV and ICD-10 mental disorders in the Australian general population. Psychological Medicine, 36, 1593–1600.

social phobia Stein, M. B., & Chavira, D. A. (1998). Subtypes of social phobia and comorbidity with depression and other anxiety disorders. Journal of Affective Disorders, 50, S11–S16. Stemberger, R., Turner, S. M., Beidel, D. C., & Calhoun, K. S. (1995). Social phobia: An analysis of possible developmental factors. Journal of Abnormal Psychology, 104, 526–531. Taylor, S., & Cox, B. J. (1998). An expanded Anxiety Sensitivity Index: Evidence for a hierarchic structure in a clinical sample. Journal of Anxiety Disorders, 12, 463–483. Tillfors, M., Furmark, T., Ekselius, L., & Fredrikson, M. (2001). Social phobia and avoidant personality disorder as related to parental history of social anxiety: A general population study. Behaviour Research and Therapy, 39, 289–298. Tillfors, M., Furmark, T., Ekselius, L., & Fredrikson, M. (2004). Social phobia and avoidant personality disorder: One spectrum disorder? Nordic Journal of Psychiatry, 58, 147–152. Tran, G. Q., & Chambless, D. L. (1995). Psychopathology of social phobia: Effects of subtype and of avoidant personality disorder. Journal of Anxiety Disorders, 9, 489–501. Turner, S. M., Beidel, D. C., & Townsley, R. M. (1992). Social phobia: A comparison of specific and generalized subtypes and avoidant personality disorder. Journal of Abnormal Psychology, 101, 326–331. van Velzen, C. J. M., Emmelkamp, P. M. G., & Scholing, A. (2000). Generalized social phobia versus avoidant personality disorder: Differences in psychopathology, personality traits, and social and occupational functioning. Journal of Anxiety Disorders, 14, 395–411. Vollebergh, W. A. M., Iedema, J., Bijl, R. V., de Graaf, R., Smit, F., & Ormel, J. (2001). The structure and stability of

13

common mental disorders: The NEMESIS study. Archives of General Psychiatry, 58, 597–603. Watson, D. (2005). Rethinking the mood and anxiety disorders: A quantitative hierarchical model for DSM-V. Journal of Abnormal Psychology, 114, 522–536. Watson, D., & Clark, L. A., (1991). The Mood and Anxiety Symptom Questionnaire. Unpublished manuscript, University of Iowa, Department of Psychology, Iowa City. Watson, D., Clark, L. A., Weber, K., Assenheimer, J. S., Strauss, M. E., & McCormick, R. A. (1995). Testing a tripartite model: II. Evaluating the symptom structure of anxiety and depression in student, adult, and patient samples. Journal of Abnormal Psychology, 104, 15–25. Watson, D., & Walker, L. M. (1996). The long-term stability and predictive validity of trait measures of affect. Journal of Personality and Social Psychology, 70, 567–577. Watson, D., Weber, K., Assenheimer, J. S., Clark, L. A., Strauss, M. E., & McCormick, R. A. (1995). Testing a tripartite model: I. Evaluating the convergent and discriminant validity of anxiety and depression symptom scales. Journal of Abnormal Psychology, 104, 3–14. Widiger, T. A., & CostaJr., P. T. (1994). Personality and personality disorders. Journal of Abnormal Psychology, 103, 78–91. Wu, K. D., Clark, L. A., & Watson, D. (2006). Relations between obsessive-compulsive disorder and personality: Beyond axis I and axis II comorbidity. Journal of Anxiety Disorders, 20, 695–717. Zvolensky, M. J., Arrindell, W. A., Taylor, S., Bouvard, M., Cox, B. J., Stewart, S. H., et al. (2003). Anxiety sensitivity in six countries. Behaviour Research and Therapy, 41, 841–859. R E C E I V E D : June 23, 2008 A C C E P T E D : October 19, 2008 Available online 31 October 2008