- Email: [email protected]
Clinical predictors of diagnostic status in individuals with social anxiety disorder
Available online at www.sciencedirect.com
ScienceDirect Comprehensive Psychiatry 55 (2014) 1906 – 1913 www.elsevier.com/locate/comppsych
Clinical predictors of diagnostic status in individuals with social anxiety disorder Antonina S. Farmer a, b,⁎, Daniel F. Gros a, b , Randi E. McCabe c, d , Martin M. Antony e a Mental Health Service, Ralph H. Johnson Veterans Affairs Medical Center, Charleston, SC, USA Department of Psychiatry and Behavioral Sciences, Medical University of South Carolina, Charleston, SC, USA c Anxiety Treatment and Research Centre, St. Joseph's Healthcare, Hamilton, Ontario, Canada d Department of Psychiatry and Behavioural Neurosciences, McMaster University, Hamilton, Ontario, Canada e Department of Psychology, Ryerson University, Toronto, Ontario, Canada
b
Abstract Objective: In psychiatric patients, comorbidity tends to be the rule, rather than the exception. This is especially true for patients with social anxiety disorder (SAD), but research on the implications of diagnostic status has been limited. This study aimed to examine the frequency of SAD as: (1) the only diagnosis, (2) a principal diagnosis with comorbid conditions, or (3) a comorbid condition when another diagnosis is principal in a treatmentseeking population. The study also sought to identify clinical features that distinguish people in these diagnostic groups. Method: Our sample included 684 adult participants presenting for treatment in a specialty clinic for anxiety disorders. We established diagnoses with semistructured clinical interviews, and participants completed self-report measures of social anxiety, associated transdiagnostic symptoms, general distress, and impairment due to psychological difficulties. We analyzed group differences and investigated predictors of principal SAD diagnosis. Results: Over half of participants reported symptoms that met criteria for a SAD diagnosis (51.8%). Of these, 8.8% had SAD only (no comorbid psychiatric diagnoses), 48.6% had multiple conditions with SAD as the principal diagnosis, and 42.7% had multiple conditions with SAD as an additional diagnosis. SAD-only was associated with less severe impairment and transdiagnostic symptoms. Among participants with comorbid conditions, greater fear of negative evaluation, behavioral avoidance, and coping with substances predicted a principal SAD diagnosis, whereas SAD as an additional diagnosis was more likely when participants presented with greater anxiety sensitivity, intolerance of uncertainty, and thought avoidance. Conclusions: Our findings suggest that principal diagnosis of SAD is common in a treatment-seeking population and is associated with more severe disorder-specific symptoms and less severe transdiagnostic features related to anxiety. Implications for assessment and treatment planning in clinical practice are discussed. © 2014 Elsevier Inc. All rights reserved.
1. Introduction Social anxiety disorder (SAD) is the persistent, disproportionate fear of social or performance situations in which a This study is supported by Department of Veteran Affairs Clinical Sciences Research and Development Career Development Award CX000845 (PI: Gros). This material is the result of work supported with resources and the use of facilities at the Ralph H. Johnson VAMC. The views expressed in this article are those of the authors and do not necessarily reflect the position or policy of the Department of Veterans Affairs or the United States government. Submitted for publication to Comprehensive Psychiatry (May 2014). ⁎ Corresponding author at: Department of Psychology, Randolph-Macon College, Ashland, Virginia 23005. Tel.: +1 804 752 3734. E-mail address: [email protected] (A.S. Farmer). http://dx.doi.org/10.1016/j.comppsych.2014.07.019 0010-440X/© 2014 Elsevier Inc. All rights reserved.
person may be scrutinized or evaluated [1]. After major depressive disorder (MDD), alcohol dependence, and specific phobia, this condition is the fourth most prevalent psychiatric disorder, occurring in approximately 12.1% of people in their lifetimes [2]. However, this disorder very frequently co-occurs with other psychiatric conditions, and researchers are just beginning to understand reasons for this co-occurrence and implications of comorbidity for the presentation of symptoms, treatment response, and prognosis. Epidemiological studies estimate about 80% [3–5] of people with SAD also experience a comorbid psychiatric condition, most commonly other anxiety disorders and MDD. In fact, in a sample of people seeking treatment in an outpatient psychiatric setting, researchers found that SAD was the
A.S. Farmer et al. / Comprehensive Psychiatry 55 (2014) 1906–1913
principal diagnosis for only 4% of the 640 participants with a diagnosis of SAD in the sample [6]. Notably, this study defined principal diagnosis as the primary reason for the individual seeking treatment. This finding suggests that despite the symptoms of SAD being significantly impairing, few people with SAD actually describe it as their chief complaint. A number of retrospective and prospective studies have suggested that SAD typically precedes the development of other psychiatric conditions, like depression [e.g., 7–10]. Yet, we know little about differences among people for whom SAD is the most severe and impairing disorder and people who have a more severe psychiatric condition with SAD as an additional diagnosis. Although a number of studies have investigated the frequency of comorbid conditions in SAD, only a few have examined clinical correlates of SAD with comorbid conditions compared to SAD only. People who have multiple psychiatric disorders usually report more severe symptoms, as well as greater distress and impairment compared to those who have SAD only [11,12]. This may be in part because those with co-occurring conditions may remain undiagnosed and thus untreated longer, leading to greater distress and impairment from social fears [13]. Additionally, people with SAD and another psychiatric condition are more likely to seek help and take medications to control their symptoms [14]. However, a study examining primary care setting interactions found people with SAD to mostly seek help for their comorbid psychiatric conditions, particularly depression [10]. Only 5.6% of people with SAD and without depression even mentioned psychological symptoms to their providers. Overall, psychiatric comorbidity has important implications for treatment seeking and provision of clinical care. Given that the vast majority of people with SAD present for treatment with other psychiatric conditions, it is surprising that we know little about the clinical presentation of SAD when not the principal diagnosis. Several studies have examined differences in people for whom SAD precedes or follows another diagnosis temporally [e.g., 9,15]. However, precedence does not necessarily suggest greater severity or impairment. We sought to address this important gap in our knowledge of whether and how clinical features vary across cases where SAD is one of several diagnoses versus the principal diagnosis. The present categorical diagnostic system described in the Diagnostic and Statistical Manual of Mental Disorders [1,16] has been criticized due to the significant overlap in symptoms, particularly across the anxiety and mood disorders [17,18]. Researchers have proposed hybrid models of disorders that propose disorders to have shared symptom dimensions as well as disorder-specific symptoms. The hybrid model of SAD [19,20] suggests that this condition shares some symptoms with other anxiety and depressive disorders (e.g., avoidance of unpleasant thoughts, anhedonia, using substances for coping, and functional impairment) but the severity of SAD would be determined by relatively
1907
unique features (e.g., behavioral avoidance of situations, fear of negative evaluation). According to this model, which has been replicated across analog and clinical samples [19,20], patients with SAD and comorbid diagnoses should present with high levels of nonspecific symptoms, whereas patients with a principal diagnosis of SAD should present with the highest levels of disorder-specific symptoms. The aim of the present study was to compare symptom characteristics and demographic features of adults with (1) SAD only (no comorbid psychiatric diagnoses), (2) SAD as a principal diagnosis with co-occurring psychological conditions, and (3) SAD as an additional diagnosis when another disorder is principal. In particular, we sought to examine how diagnostic status relates to self-reported symptom dimensions related to SAD, symptoms associated with frequently comorbid conditions (e.g., depression), and transdiagnostic symptoms that are closely associated with other anxiety disorders. Specifically, we assessed for anxiety sensitivity, the tendency for people to misinterpret physiological symptoms of anxiety as dangerous; this construct is closely related to panic disorder [21]. We also measured intolerance of uncertainty, a discomfort with lack of certainty about future events; this construct is closely related to worry and generalized anxiety [22,23]. Based on available literature and clinical experience, we hypothesized that participants with comorbid conditions would exhibit greater levels of nonspecific psychological symptoms and impairment than those with only SAD. We also expected that participants with principal SAD would have higher levels of SADspecific symptoms compared to those with SAD as an additional diagnosis.
2. Method 2.1. Participants and procedure Participants included 684 psychiatric outpatients presenting for treatment in a Canadian university hospital clinic specializing in the assessment and treatment of anxiety disorders. The sample consisted of 263 men (38.5%) and 420 women (61.5%), ranging in age from 15 to 74 years (M = 36.30, SD = 12.71). Most participants self-identified as White (60.6%), followed by Asian (34.9%), NativeCanadian (1.8%), Black (0.7%), Hispanic (0.4%), and other ethnicities (1.6%). The marital status of the participants was primarily single (44.9%), followed by married (36.8%), cohabitating (9.7%), divorced (4.9%), separated (3.1%), and other (0.7%). The education status of the sample was as follows: less than high school degree (14.7%); high school degree or equivalent (40.5%); college degree (37.4%); at least some graduate or professional education (7.4%). Notably, several participants were missing demographic data about sex (n = 1), ethnicity (n = 7), education status (n = 8), and marital status (n = 4).
1908
A.S. Farmer et al. / Comprehensive Psychiatry 55 (2014) 1906–1913
After receiving a patient referral from the participant's physician (usually their primary care doctor or psychiatrist), each participant was called to schedule an intake interview prior to initiating treatment. During this session, patients completed a series of self-report questionnaires and underwent a clinical interview to determine diagnoses based on the Diagnostic and Statistical Manual, Fourth Edition (DSM-IV) [1]. All participants provided informed, voluntary, written consent for their data to be included in the study. The procedures and measures used in this study were approved by the local institutional review board and incorporated standard assessment practices in the clinic. 2.2. Assessment tools 1. Structured Clinical Interview for DSM-IV Axis I Diagnoses (SCID-I) [24]. This semistructured diagnostic interview was used to assess for the presence of current and lifetime Axis I disorders. This instrument was administered by doctoral-level psychologists, postdoctoral fellows, or advanced graduate students (with thorough training for diagnostic reliability). When criteria for multiple disorders were met, the current disorder causing the most distress and/or functional impairment was identified as the principal diagnosis via a discussion between the patient and clinician, and informed by self-report measures. The interrater reliability of assessors in our clinic was good (kappa = 0.89). 2. Multidimensional Assessment of Social Anxiety (MASA) [19,25]. This 38-item self-report measure was used to assess specific and nonspecific symptom dimensions related to social anxiety, consistent with the hybrid model of social anxiety. The measure consists of six subscales (αs = .72–.91) that assess a specific factor for the diagnostic category of social anxiety (behavioral avoidance, MASA-BA), and five nonspecific factors for related symptoms of anxiety: physiological arousal and avoidance (MASA-PA) thought avoidance (MASA-TA), anhedonia (MASA-ANH), functional impairment (MASA-FI), and coping with substances (MASA-CS). These scales have demonstrated adequate internal consistency, convergent and discriminant validity, and test–retest reliability in clinical and nonclinical samples [19,20,25]. 3. Brief Fear of Negative Evaluation scale (BFNE) [26]. As an additional measure of social anxiety-specific fears, this 12-item self-report questionnaire (α = .97) assessed apprehension about being negatively evaluated due to loss of social approval. This scale has demonstrated adequate reliability, internal consistency, and convergent validity with other measures of social anxiety [26,27]. 4. Depression Anxiety Stress Scales – 21-item version (DASS-21) [28]. This 21-item questionnaire measured
psychological symptoms more broadly, yielding three psychometrically distinct subscales (αs = .82–.91): depression (e.g., dysphoria, hopelessness, inertia, and self-depreciation); anxiety (e.g., autonomic arousal, anxious affect, and situational anxiety); and stress (e.g., difficulty relaxing, nervous tension, irritability, and agitation). Prior literature has found the DASS-21 scales to demonstrate good internal consistency, factor structure, concurrent validity, and reliability [28–30] and scores were doubled to be comparable with the 42-item version [31]. 5. Anxiety Sensitivity Index (ASI) [32]. This 16-item self-report instrument (α = .90) assessed the degree to which a person believes that physiological anxiety symptoms have negative social, cognitive, and/or physical consequences. This scale has good psychometric properties [33]. 6. Intolerance of Uncertainty Scale – 12 item version (IUS-12) [34]. This 12-item self-report measure (α = .93) was used to assess how much a person identifies with the idea that uncertainty about future events is unacceptable and problematic and the extent to which such uncertainty inhibits action. Psychometric properties have been demonstrated in prior research [34]. 7. Illness Intrusiveness Ratings Scale (IIRS) [35]. As a measure of functional impairment due to psychiatric symptoms, this 13-item instrument (α = .87) assessed perceived interference of any symptoms across domains considered integral to quality of life (e.g., health, work, relationships). The IIRS scale has demonstrated adequate reliability, validity, and sensitivity to change in various clinical populations [36].
2.3. Statistical analyses We first conducted statistical analyses to examine differences in demographic and symptomological variables among the diagnostic groups. For categorical variables (e.g., gender, comorbid diagnosis occurrence), we summarized data with absolute (n) and relative (%) values, and we investigated group differences with Pearson chi-squared tests. For continuous variables (e.g., age, subscale scores), we summarized measures with means and standard deviations, and we investigated group differences in symptom measures with one-way analyses of variance (ANOVAs), with Tukey HSD posthoc tests for bivariate comparisons. Lastly, we conducted a backward stepwise logistic regression analysis to identify predictors of a principal SAD diagnosis when participants present comorbid diagnoses, allowing us to examine relative contribution of variables. We began by including all variables in the full model for which diagnostic status had a significant effect and then used a backwards elimination approach by systematically removing nonsignificant variables [37]. All statistical analyses were performed with IBM SPSS 19.0 software.
A.S. Farmer et al. / Comprehensive Psychiatry 55 (2014) 1906–1913
1909
frequency of specific phobia, depressive disorders, or bipolar disorder between diagnostic groups.
3. Results 3.1. Prevalence of SAD and comorbid diagnoses In this treatment-seeking sample, over half of participants had a diagnosis of SAD (n = 354, 51.8%). Of these, 31 participants had no comorbidities (i.e., SAD-only; 4.5% of total sample), while 323 participants had comorbid diagnoses. SAD was the principal diagnosis for 172 participants who received more than one diagnosis (i.e., principal SAD; 25.1%). SAD was an additional diagnosis for 151 participants who had another condition as the principal diagnosis (i.e., additional SAD; 22.1%). Table 1 describes the demographic characteristics by diagnostic group. With regard to demographic variables, groups significantly differed in age and sex, as well as a marginally significant effect of relationship status. Posthoc tests revealed that the SAD-only group tended to be significantly younger than the additional SAD group (p = .034), but the principal and additional SAD groups did not differ (p = .54). Participants in the SAD-only group were also more likely to be male compared to the additional SAD group (p = .029), while participants with additional SAD were marginally more likely to be married or cohabiting than those with principal SAD or SAD-only (ps b .05). Groups did not differ in ethnicity, education attainment, or age of onset of SAD symptoms. Table 1 also lists the most frequent comorbid Axis I diagnoses (N 5%) in participants with a SAD diagnosis and chi-squared statistics comparing frequencies among participants with principal SAD and additional SAD. Overall, participants with additional SAD were more likely to report symptoms meeting criteria for comorbid anxiety disorders, including panic disorder with or without agoraphobia, obsessive-compulsive disorder, and generalized anxiety disorder. There were no significant differences in the
3.2. Clinical profiles Table 2 describes group means and diagnostic group differences in specific social anxiety symptoms, nonspecific and related symptom dimensions, transdiagnostic symptoms relevant to other anxiety disorders, as well as the interference of psychiatric symptoms. The effect of diagnostic group was statistically significant for all clinical measures except for marginal significance for intolerance of uncertainty (p = .051). Posthoc analyses were conducted to examine bivariate differences. As predicted, the SAD-only group presented with less severe nonspecific symptoms associated with social anxiety, transdiagnostic symptoms related to other anxiety conditions, and symptoms of broad anxiety and depression. With regard to mood and anxiety symptoms broadly, the SADonly group had significantly lower scores on DASS-21Depression (ps b .001) and DASS-21-Anxiety (ps b .01), but higher scores on DASS-21-Stress (ps b .001), compared with both the principal and additional SAD groups. The SAD-only group also had lower scores on the ASI and IIRS than participants in both comorbid groups (ps b .05). However, differences were less consistent with nonspecific symptoms. The SAD-only group had lower scores on MASA-PA (p = .004) and IUS-12 (p = .039), and marginally lower scores on MASA-TA (p = .051) than the additional SAD group but not the principal SAD group. The SAD-only group also had lower scores on MASA-ANH (p = .003) and MASA-CS (p = .048) than the principal SAD group but not the additional SAD group. Among SAD participants with comorbid diagnoses, the principal and additional groups significantly differed on a
Table 1 Diagnostic status group differences in demographic characteristics of patients with social anxiety disorder (SAD). SAD-only n = 31 Age – mean (SD) Age of SAD onset Sex (% female) Ethnicity/race (% white) Marital status (% married/cohabitating) Education (% with college degree) Comorbid Axis I Diagnoses Panic disorder Obsessive-compulsive disorder Generalized anxiety disorder Specific phobia Posttraumatic stress disorder Major depressive disorder Bipolar disorder Substance use disorder
a
29.60 (9.99) 12.93 (7.70) 15 (48.4%) a 17 (54.8%) 10 (32.3%) a 17 (54.8%)
Principal SAD n = 172 a,b
33.99 (11.88) 13.17 (7.85) 98 (57.0%) a,b 108 (63.5%) 54 (32.0%) a 65 (38.0%) 41 (23.8%) 37 (21.5%) 75 (43.6%) 42 (24.4%) 11 (6.4%) 109 (63.4%) 18 (10.5%) 32 (18.6%)
Additional SAD n = 151 b
35.38 (11.87) 14.01 (10.19) 103 (68.7%) b 93 (62.4%) 67 (4.4%) 62 (41.6%) 66 55 92 46 18 98 10 18
(43.7%) (36.4%) (60.9%) (30.5%) (11.9%) (64.9%) (6.6%) (11.9%)
Difference Statistic F[2,351] = 3.18 F[2,328] = 3.18 χ 2 = 6.96 χ 2 = 0.85 χ 2 = 5.63 χ 2 = 3.11 χ2 χ2 χ2 χ2 χ2 χ2 χ2 χ2
= = = = = = = =
a
.043 n.s. .031 n.s. .060 n.s. b.001 .003 .002 n.s. .083 n.s. n.s. .098
4.33 8.78 9.66 1.48 3.00 0.08 1.50 2.75
Summary statistics are means (with standard deviations) for age, and number (with percent) of participants meeting criteria for each criterion. identified with same letter are not significantly different from each other (p b .05) in posthoc tests.
p-value
a,c
Group means
1910
A.S. Farmer et al. / Comprehensive Psychiatry 55 (2014) 1906–1913
number of social anxiety specific and transdiagnostic measures. As predicted, the principal SAD group reported more severe symptoms specific to SAD. They tended to have higher scores on the BFNE (p = .014), social avoidance (MASA-BA; p b .001), and functional impairment (MASAFI; p = .015) compared to the additional SAD group. In contrast, the additional SAD group had higher scores on nonspecific symptom measures: MASA-TA (p =.001), DASS-21-Anxiety ( p = .032), and ASI ( p = .022). Surprisingly, principal and additional groups did not differ on the IIRS (p = .621), suggesting similar levels of psychological symptom intrusiveness of into the quality of life. 3.3. Predicting principal SAD diagnosis Lastly, we conducted an analysis to determine predictors of a principal SAD diagnosis in the presence of comorbid conditions. In a logistic regression predicting diagnostic status (principal = 1, additional = 0), six symptom dimensions remained significantly associated with a principal diagnosis of SAD (no demographic remained significant). These results (summarized in Table 3) determined that participants were more likely to have a principal diagnosis of SAD when they reported greater fear of negative evaluation, social avoidance, and coping with substances; participants were more likely to have additional SAD if they reported greater thought avoidance, anxiety sensitivity, or intolerance of uncertainty. For example, all else being equal, for every point increase on the MASA-BA scale, participants with comorbid diagnoses were 10% more likely to have a principal diagnosis of SAD. Based on comparison of log likelihood ratios, the predictive model was significantly better at predicting SAD diagnostic status than chance (χ 2 = 74.20, p b .001), accurately categorizing 70.2% of participants. However, this is only a 17.5% improvement over the null model, suggesting that other factors also play a role in diagnostic status.
4. Discussion The present study addressed the clinical presentation of comorbidity in SAD among people presenting for treatment to an anxiety disorders specialty clinic. Similar to previous studies in epidemiological and clinical samples, we found SAD only (i.e., in the absence of comorbid disorders) to account for a small percentage of SAD diagnoses (8.8% of participants with SAD). However, contrary to past reports of very low occurrence of SAD as the principal diagnosis when comorbid conditions were present [6], we found SAD to be approximately equally likely to be the principal diagnosis and an additional diagnosis. Notably, previous studies defined the principal diagnosis as the primary reason for treatment. Our use of severity and impairment as the criterion for principal diagnosis suggests that SAD symptoms are as likely to be acutely problematic as symptoms of comorbid conditions when they co-occur. It is important to highlight that our sample was drawn from a tertiary service provider. Since people with SAD have been found to be more likely to seek help initially for problems other than SAD, it makes sense that our sample has a higher rate of comorbidity than the general population. Furthermore, the 8.8% SAD-only rate is lower than epidemiological estimates of 15–20% [3–5]. Since our participants were presenting to an anxiety disorders clinic, it is probable that their anxiety conditions were more likely to be the most severe diagnosis compared to people presenting to less specialized providers. This finding has potential significance to clinical practice by informing identification of this often underdiagnosed condition. Although the majority of patients with SAD speak to their provider about comorbid psychiatric symptoms first, about 75% of them express interest in treatment for SAD when offered directly [38]. However, providers would need to know to ask about and recognize symptoms of SAD, which may be more difficult when comorbid symptoms are
Table 2 Diagnostic Status group differences in symptom profiles of patients with social anxiety disorder (SAD). SAD-only n = 31 Brief Fear of Negative Evaluation Multidimensional Assessment of Social Anxiety Behavioral avoidance Physiological arousal and avoidance Anhedonia Functional impairment Thought avoidance Coping with substances Depression Anxiety Stress Scales – 21 items Depression Anxiety Stress Intolerance of Uncertainty Scale – 12 items Anxiety Sensitivity Scale Illness Intrusiveness Ratings Scale
40.90 (6.89)
a,b
Principal SAD n = 172 41.02 (8.12)
a
Additional SAD n = 151 38.27 (9.62)
b
Difference statistic F[2,349] = 4.25
p-value .015
39.93 (7.28) a,b 6.84 (3.53) a 16.16 (5.72) a 18.29 (3.48) a,b 19.87 (6.31) a,b 7.81 (4.31) a
42.06 (7.89) a 8.39 (3.53) a,b 20.31 (6.50) b 19.17 (3.86) a 20.22 (6.72) a 10.09 (5.22) b
37.93 (8.17) b 9.08 (3.69) b 19.03 (6.28) a,b 17.87 (4.63) b 22.92 (6.55) b 8.36 (4.69) a,b
F[2,350] F[2,350] F[2,350] F[2,350] F[2,350] F[2,350]
= = = = = =
10.80 5.44 6.08 3.95 7.58 6.20
b.001 .005 .003 .020 .001 .002
12.84 (9.59) 9.55 (7.51) 16.19 (7.68) 37.74 (9.06) a 24.54 (10.79) 50.45 (10.6)
23.20 (10.86) a 15.37 (9.09) 10.09 (5.22) a 41.95 (10.94)a,b 32.81 (11.97) 59.36 (14.4) a
21.17 (10.62) a 17.94 (9.38) 8.36 (4.69) a 42.87 (10.51) b 36.52 (13.14) 57.80 (15.6) a
F[2,350] F[2,350] F[2,350] F[2,346] F[2,346] F[2,349]
= = = = = =
12.48 11.64 12.39 3.00 12.77 4.86
b.001 b.001 b.001 .051 b.001 .008
Summary statistics are means, with standard deviations in parentheses. a,b,cGroup means identified with same letter are not significantly different from each other (p b .05) based on post-hoc Tukey's HSD tests.
A.S. Farmer et al. / Comprehensive Psychiatry 55 (2014) 1906–1913 Table 3 Demographic and symptom variables predicting principal social anxiety disorder.
Brief Fear of Negative Evaluation MASA – Behavioral avoidance MASA – Thought avoidance MASA – Coping with substances Anxiety Sensitivity Scale Intolerance of Uncertainty Scale
Odds ratio (95% CI)
p-value
1.05 (1.03–1.07) 1.10 (1.08–1.12) 0.93 (0.91–0.95) 1.10 (1.07–1.13) 0.96 (0.95–0.97) 0.97 (0.95–0.98)
.006 b.001 b.001 .001 .001 .050
CI = confidence interval; MASA = Multidimensional Assessment of Social Anxiety.
present, particularly when those symptoms are also severe and impairing. Consistent with our predictions, we found greater severity of associated symptoms (e.g., depression, anxiety sensitivity) and illness intrusiveness in participants with comorbid diagnoses in addition to SAD. Notably, the presence of comorbid diagnoses was not associated with severity of SAD-specific symptoms (e.g., fear of negative evaluation, social avoidance), contrary to past research [5,11]. Compared to earlier epidemiological studies, our sample was seeking treatment for anxiety symptoms; thus, the overall high degree of severity and impairment from psychological symptoms in this population may obscure an effect of comorbidity on severity. This study adds to prior knowledge by examining differences among people with SAD as the principal diagnosis or an additional diagnosis when other conditions co-occur. Participants in the principal SAD group presented with more severe fear of negative evaluation, social avoidance, and use of substances to cope with anxiety, but less severe transdiagnostic symptoms of thought avoidance, anxiety sensitivity, and intolerance of uncertainty compared to the additional SAD group. Despite statistically significant differences, all specific and nonspecific symptoms were elevated across the three groups. These associations provide further support for a hybrid model of SAD [19], as specific symptoms of social avoidance and fear of negative evaluation distinguished people with principal SAD diagnoses, but nonspecific symptoms were present, though to different degrees, across the diagnostic groups. Although we did not further distinguish among specific comorbid diagnoses, it is noteworthy that depression symptoms were not different across the two comorbid groups, and a mood disorder diagnosis was similarly likely. Nonspecific anxiety symptoms and thought avoidance were greatest in the additional SAD group, consistent with their greater likelihood of meeting criteria for panic disorder, obsessive-compulsive disorder, and generalized anxiety disorder. This difference may suggest that comorbid anxiety conditions are more likely to supersede the severity of SAD symptoms. However, because our sample was drawn from an anxiety disorders specialty clinic, this is likely an effect of this population presenting for treatment for primarily anxiety symptoms.
1911
Across psychiatric disorders, comorbidity tends to require more treatment planning and/or more complex therapeutic approaches, which may tax mental health providers' resources with additional need for time and training. Comorbidity may also contribute to poorer responses to treatment or greater likelihood of relapse [7]. Some initial research comparing treatment outcomes of SAD only and SAD with comorbid disorders found poorer prognosis when mood disorders were present, though not with comorbid anxiety disorders [39]. The high frequency of nonspecific mood and anxiety symptoms across our sample (including the SAD-only group) supports the use of recently developed transdiagnostic treatment approaches that target both specific and nonspecific anxiety and mood symptoms [40,41]. Though some evidence suggests greater decreases in comorbidity with such treatments [42], a single study examining equivalence of transdiagnostic and traditional diagnosis-specific treatment in reducing specific social anxiety symptoms was inconclusive [43]. This study also has potential implications for future research on SAD. Often, investigators recruit participants with only principal diagnoses to study underlying dysfunction, response to interventions, and mechanisms of change. With regard to SAD, our results support past evidence that comorbidity tends to be the rule rather than the exception. The small (though significant) differences in social anxiety symptom presentation among the three groups suggest that diagnostic status (e.g., principal vs. additional) may be less important than knowing whether a person's symptoms meet criteria for particular disorders. However, further research is necessary to clarify the nature and implications of symptom overlap between SAD and other conditions. The present study did not include any self-report symptom measures of other conditions in our study, but such measures may help understand phenotypic variation in anxiety disorder cases. Additionally, future research may examine how underlying physiological processes, emotion regulation strategies, and daily experiences may differ across different diagnostic statuses of SAD. With regard to demographic predictors of diagnostic status, we found several interesting differences. Participants with SAD-only tended to be younger but only compared to the additional SAD group. Though this study is crosssectional in nature, this finding lines up with research on SAD typically preceding the development of comorbid conditions since our groups did not differ in age of onset of SAD [11,44]. Longitudinal studies could provide information about how the clinical course of symptoms relates to decision-making regarding principal diagnoses at the outset of treatment. Additionally, prospective studies might elucidate how diagnostic status at the outside of treatment relates to treatment outcome. This study had a number of strengths over previous research on features associated SAD. First, our large treatment-seeking sample provided sufficient data to distinguish among these three diagnostic groups. Second, we used
1912
A.S. Farmer et al. / Comprehensive Psychiatry 55 (2014) 1906–1913
a validated assessment of participants' symptoms with a clinical interview by advanced trainees and mental health professionals. Third, our measures allowed us to examined both diagnosis-specific and transdiagnostic symptom patterns. Nevertheless, a limitation of this study is the largely demographically homogenous sample. Also, in conducting clinical interviews as part of the research study, future studies may also collect data on the participants' specific reason for seeking treatment. This information may help make comparisons with previous findings and investigate corroboration between clinician and patient views on the severity and interference of symptoms across the diagnostic groups.
5. Conclusion The present findings highlight the disorder-specific and nonspecific clinical features associated with the diagnostic status of SAD. Specifically, this pattern supports a hybrid model of SAD [19], with which clinicians can conceptualize a patient's problems on multiple dimensions common across diagnoses while also considering the severity of distinguishing features. Such a model may minimize the reliance on multiple diagnoses to represent the range of patients' symptoms. This study also presents novel information about the frequency of SAD as a principal (most severe, impairing) diagnosis in a treatment-seeking population. Given that people with this diagnosis are likely to be hesitant to mention psychological difficulties, particularly social fears, proper diagnosis may require querying for these concerns even if a patient initially raises other anxiety or mood difficulties.
References [1] American Psychiatric Association. Diagnostic and statistical manual of mental disorders: DSM-IV-TR. Washington, DC: American Psychiatric Publishing, Inc.; 2000. [2] Kessler RC, Berglund P, Demler O, Jin R, Merikangas KR, Walters EE. Lifetime prevalence and age-of-onset distributions of DSM-IV disorders in the National Comorbidity Survey Replication. Arch Gen Psychiatry 2005;62:593-602. [3] Lecrubier Y. Comorbidity in social anxiety disorder: impact on disease burden and management. J Clin Psychiatry 1998;59:33-8. [4] Merikangas K, Angst J. Comorbidity and social phobia: evidence from clinical, epidemiologic, and genetic studies. Eur Arch Psychiatry Clin Neurosci 1995;244:297-303. [5] Ruscio AM, Brown TA, Chiu WT, Sareen J, Stein MB, Kessler RC. Social fears and social phobia in the USA: results from the National Comorbidity Survey Replication. Psychol Med 2008;38:15-28. [6] Zimmerman M, McGlinchey JB, Chelminski I, Young D. Diagnostic co-morbidity in 2300 psychiatric out-patients presenting for treatment evaluated with a semi-structured diagnostic interview. Psychol Med 2008;38:199-210. [7] Angst J. Comorbidity of anxiety, phobia, compulsion and depression. Int Clin Psychopharmacol 1993;8:21-5. [8] Beesdo K, Bittner A, Pine DS, Stein MB, Holfer M, Lieb R, et al. Incidence of social anxiety disorder and the consistent risk for secondary depression in the first three decades of life. Arch Gen Psychiatry 2007;64:903-12.
[9] Perugi G, Simonini E, Savino M, Mengali F, Cassano GB, Akiskal HS. Primary and secondary social phobia: psychopathologic and familial differentiations. Compr Psychiatry 1990;31:245-52. [10] Weiller E, Bisserbe JC, Boyer P, Lepine JP, Lecrubier Y. Social phobia in general health care: an unrecognised undertreated disabling disorder. Br J Psychiatry 1996;168:169-74. [11] Schneier FR, Johnson J, Hornig CD, Liebowitz MR, Weissman MM. Social phobia: comorbidity and morbidity in an epidemiologic sample. Arch Gen Psychiatry 1992;49:282-8. [12] Wittchen HU, Fuetsch M, Sonntag H, Müller N, Liebowitz M. Disability and quality of life in pure and comorbid social phobia. Findings from a controlled study. Eur Psychiatry 2000;15:46-58. [13] Dunner DL. Management of anxiety disorders: the added challenge of comorbidity. Depress Anxiety 2001;13:57-71. [14] Magee WJ, Eaton WW, Wittchen HU, McGonagle KA, Kessler RC. Agoraphobia, simple phobia, and social phobia in the National Comorbidity Survey. Arch Gen Psychiatry 1996;53:159-68. [15] Van Ameringen M, Mancini C, Styan G, Donison D. Relationship of social phobia with other psychiatric illness. J Affect Disord 1991;21:93-9. [16] American Psychiatric Association. Diagnostic and statistical manual of mental disorders. 5th ed. Arlington, VA: American Psychiatric Publishing; 2013. [17] Brown TA, Barlow DH. Dimensional versus categorical classification of mental disorders in the fifth edition of the Diagnostic and statistical manual of mental disorders and beyond: comment on the special section. J Abnorm Psychol 2005;114:551-6. [18] McGlinchey JB, Zimmerman M. Examining a dimensional representation of depression and anxiety disorders' comorbidity in psychiatric outpatients with item response modeling. J Abnorm Psychol 2007;116:464-74. [19] Gros DF, Simms LJ, Antony MM. A hybrid model of social phobia: an analysis of social anxiety and related symptoms of anxiety. J Clin Psychol 2011;67:293-307. [20] Gros DF, McCabe RE, Antony MM. Using a hybrid model to investigate the comorbidity and symptom overlap between social phobia and the other anxiety disorders and unipolar mood disorders. Psychiatry Res 2013;210:188-92. [21] Reiss S. Expectancy model of fear, anxiety, and panic. Clin Psychol Rev 1991;11:141-53. [22] Buhr K, Dugas MJ. Investigating the construct validity of intolerance of uncertainty and its unique relationship with worry. J Anxiety Disord 2006;20:222-36. [23] Dugas MJ, Buhr K, Ladouceur R. The role of intolerance of uncertainty in etiology and maintenance. In: Heimberg RG, Turk CL, & Mennin DS, editors. Generalized anxiety disorder: Advances in research and practice. New York: Guilford Press; 2004. p. 143-63. [24] First MB, Spitzer RL, Williams JB. Structured Clinical Interview for DSM-IV Axis I Disorders: Clinician version, administration booklet. American Psychiatric Pub; 1997. [25] Gros DF, Simms LJ, Antony MM, McCabe RE. Development and Psychometric Evaluation of the Multidimensional Assessment of Social Anxiety (MASA). J Clin Psychol 2012;68:432-47. [26] Leary MR. A brief version of the Fear of Negative Evaluation Scale. Pers Soc Psychol Bull 1983;9:371-5. [27] Carleton RN, McCreary DR, Norton PJ, Asmundson GJ. Brief Fear of Negative Evaluation Scale—revised. Depress Anxiety 2006;23:297-303. [28] Lovibond PF, Lovibond SH. The structure of negative emotional states: comparison of the Depression Anxiety Stress Scales (DASS) with the Beck Depression and Anxiety Inventories. Behav Res Ther 1995;33:335-43. [29] Brown TA, Chorpita BF, Korotitsch W, Barlow DH. Psychometric properties of the Depression Anxiety Stress Scales (DASS) in clinical samples. Behav Res Ther 1997;35:79-89. [30] Clara IP, Cox BJ, Enns MW. Confirmatory factor analysis of the Depression–Anxiety–Stress Scales in depressed and anxious patients. J Psychopathol Behav Assess 2001;23:61-7.
A.S. Farmer et al. / Comprehensive Psychiatry 55 (2014) 1906–1913 [31] Antony MM, Bieling PJ, Cox BJ, Enns MW, Swinson RP. Psychometric properties of the 42-item and 21-item versions of the Depression Anxiety Stress Scales in clinical groups and a community sample. Psychol Assess 1998;10:176-81. [32] Peterson RA, Heilbronner RL. The anxiety sensitivity index: construct validity and factor analytic structure. J Anxiety Disord 1987;1:117-21. [33] Zinbarg RE, Brown TA, Barlow DH, Rapee RM. Anxiety sensitivity, panic, and depressed mood: a reanalysis teasing apart the contributions of the two levels in the hierarchical structure of the Anxiety Sensitivity Index. J Abnorm Psychol 2001;110:372-7. [34] Carleton RN, Norton MAPJ, Asmundson GJG. Fearing the unknown: a short version of the Intolerance of Uncertainty Scale. J Anxiety Disord 2007;21:105-17. [35] Devins GM, Binik YM, Hutchinson TA, Hollomby DJ, Barré PE, Guttmann RD. The emotional impact of end-stage renal disease: importance of patients' perception of intrusiveness and control. Int J Psychiatry Med 1983;13:327-43. [36] Devins GM. Using the Illness Intrusiveness Ratings Scale to understand health-related quality of life in chronic disease. J Psychosom Res 2010;68:591-602. [37] Menard S. Applied logistic regression analysis. Sage; 2002.
1913
[38] Zimmerman M, Chelminski I. Clinician recognition of anxiety disorders in depressed outpatients. J Psychiatr Res 2003;37: 325-33. [39] Erwin BA, Heimberg RG, Juster H, Mindlin M. Comorbid anxiety and mood disorders among persons with social anxiety disorder. Behav Res Ther 2002;40:19-35. [40] Norton PJ. A randomized clinical trial of transdiagnostic cognitivebehavioral treatments for anxiety disorder by comparison to relaxation training. Behav Ther 2012;43:506-17. [41] Barlow DH, Allen LB, Choate ML. Toward a unified treatment for emotional disorders. Behav Ther 2004;35:205-30. [42] Chamberlain LD, Norton PJ. An evaluation of the effects of diagnostic composition on individual treatment outcome within transdiagnostic cognitive–behavioral group therapy for anxiety. Cogn Behav Ther 2013;42:56-63. [43] Norton PJ, Barrera TL. Transdiagnostic versus diagnosis-specific CBT for anxiety disorders: a preliminary randomized controlled noninferiority trial. Depress Anxiety 2012;29:874-82. [44] Chartier MJ, Hazen AL, Stein MB. Lifetime patterns of social phobia: a retrospective study of the course of social phobia in a nonclinical population. Depress Anxiety 1998;7:113-21.
ScienceDirect Comprehensive Psychiatry 55 (2014) 1906 – 1913 www.elsevier.com/locate/comppsych
Clinical predictors of diagnostic status in individuals with social anxiety disorder Antonina S. Farmer a, b,⁎, Daniel F. Gros a, b , Randi E. McCabe c, d , Martin M. Antony e a Mental Health Service, Ralph H. Johnson Veterans Affairs Medical Center, Charleston, SC, USA Department of Psychiatry and Behavioral Sciences, Medical University of South Carolina, Charleston, SC, USA c Anxiety Treatment and Research Centre, St. Joseph's Healthcare, Hamilton, Ontario, Canada d Department of Psychiatry and Behavioural Neurosciences, McMaster University, Hamilton, Ontario, Canada e Department of Psychology, Ryerson University, Toronto, Ontario, Canada
b
Abstract Objective: In psychiatric patients, comorbidity tends to be the rule, rather than the exception. This is especially true for patients with social anxiety disorder (SAD), but research on the implications of diagnostic status has been limited. This study aimed to examine the frequency of SAD as: (1) the only diagnosis, (2) a principal diagnosis with comorbid conditions, or (3) a comorbid condition when another diagnosis is principal in a treatmentseeking population. The study also sought to identify clinical features that distinguish people in these diagnostic groups. Method: Our sample included 684 adult participants presenting for treatment in a specialty clinic for anxiety disorders. We established diagnoses with semistructured clinical interviews, and participants completed self-report measures of social anxiety, associated transdiagnostic symptoms, general distress, and impairment due to psychological difficulties. We analyzed group differences and investigated predictors of principal SAD diagnosis. Results: Over half of participants reported symptoms that met criteria for a SAD diagnosis (51.8%). Of these, 8.8% had SAD only (no comorbid psychiatric diagnoses), 48.6% had multiple conditions with SAD as the principal diagnosis, and 42.7% had multiple conditions with SAD as an additional diagnosis. SAD-only was associated with less severe impairment and transdiagnostic symptoms. Among participants with comorbid conditions, greater fear of negative evaluation, behavioral avoidance, and coping with substances predicted a principal SAD diagnosis, whereas SAD as an additional diagnosis was more likely when participants presented with greater anxiety sensitivity, intolerance of uncertainty, and thought avoidance. Conclusions: Our findings suggest that principal diagnosis of SAD is common in a treatment-seeking population and is associated with more severe disorder-specific symptoms and less severe transdiagnostic features related to anxiety. Implications for assessment and treatment planning in clinical practice are discussed. © 2014 Elsevier Inc. All rights reserved.
1. Introduction Social anxiety disorder (SAD) is the persistent, disproportionate fear of social or performance situations in which a This study is supported by Department of Veteran Affairs Clinical Sciences Research and Development Career Development Award CX000845 (PI: Gros). This material is the result of work supported with resources and the use of facilities at the Ralph H. Johnson VAMC. The views expressed in this article are those of the authors and do not necessarily reflect the position or policy of the Department of Veterans Affairs or the United States government. Submitted for publication to Comprehensive Psychiatry (May 2014). ⁎ Corresponding author at: Department of Psychology, Randolph-Macon College, Ashland, Virginia 23005. Tel.: +1 804 752 3734. E-mail address: [email protected] (A.S. Farmer). http://dx.doi.org/10.1016/j.comppsych.2014.07.019 0010-440X/© 2014 Elsevier Inc. All rights reserved.
person may be scrutinized or evaluated [1]. After major depressive disorder (MDD), alcohol dependence, and specific phobia, this condition is the fourth most prevalent psychiatric disorder, occurring in approximately 12.1% of people in their lifetimes [2]. However, this disorder very frequently co-occurs with other psychiatric conditions, and researchers are just beginning to understand reasons for this co-occurrence and implications of comorbidity for the presentation of symptoms, treatment response, and prognosis. Epidemiological studies estimate about 80% [3–5] of people with SAD also experience a comorbid psychiatric condition, most commonly other anxiety disorders and MDD. In fact, in a sample of people seeking treatment in an outpatient psychiatric setting, researchers found that SAD was the
A.S. Farmer et al. / Comprehensive Psychiatry 55 (2014) 1906–1913
principal diagnosis for only 4% of the 640 participants with a diagnosis of SAD in the sample [6]. Notably, this study defined principal diagnosis as the primary reason for the individual seeking treatment. This finding suggests that despite the symptoms of SAD being significantly impairing, few people with SAD actually describe it as their chief complaint. A number of retrospective and prospective studies have suggested that SAD typically precedes the development of other psychiatric conditions, like depression [e.g., 7–10]. Yet, we know little about differences among people for whom SAD is the most severe and impairing disorder and people who have a more severe psychiatric condition with SAD as an additional diagnosis. Although a number of studies have investigated the frequency of comorbid conditions in SAD, only a few have examined clinical correlates of SAD with comorbid conditions compared to SAD only. People who have multiple psychiatric disorders usually report more severe symptoms, as well as greater distress and impairment compared to those who have SAD only [11,12]. This may be in part because those with co-occurring conditions may remain undiagnosed and thus untreated longer, leading to greater distress and impairment from social fears [13]. Additionally, people with SAD and another psychiatric condition are more likely to seek help and take medications to control their symptoms [14]. However, a study examining primary care setting interactions found people with SAD to mostly seek help for their comorbid psychiatric conditions, particularly depression [10]. Only 5.6% of people with SAD and without depression even mentioned psychological symptoms to their providers. Overall, psychiatric comorbidity has important implications for treatment seeking and provision of clinical care. Given that the vast majority of people with SAD present for treatment with other psychiatric conditions, it is surprising that we know little about the clinical presentation of SAD when not the principal diagnosis. Several studies have examined differences in people for whom SAD precedes or follows another diagnosis temporally [e.g., 9,15]. However, precedence does not necessarily suggest greater severity or impairment. We sought to address this important gap in our knowledge of whether and how clinical features vary across cases where SAD is one of several diagnoses versus the principal diagnosis. The present categorical diagnostic system described in the Diagnostic and Statistical Manual of Mental Disorders [1,16] has been criticized due to the significant overlap in symptoms, particularly across the anxiety and mood disorders [17,18]. Researchers have proposed hybrid models of disorders that propose disorders to have shared symptom dimensions as well as disorder-specific symptoms. The hybrid model of SAD [19,20] suggests that this condition shares some symptoms with other anxiety and depressive disorders (e.g., avoidance of unpleasant thoughts, anhedonia, using substances for coping, and functional impairment) but the severity of SAD would be determined by relatively
1907
unique features (e.g., behavioral avoidance of situations, fear of negative evaluation). According to this model, which has been replicated across analog and clinical samples [19,20], patients with SAD and comorbid diagnoses should present with high levels of nonspecific symptoms, whereas patients with a principal diagnosis of SAD should present with the highest levels of disorder-specific symptoms. The aim of the present study was to compare symptom characteristics and demographic features of adults with (1) SAD only (no comorbid psychiatric diagnoses), (2) SAD as a principal diagnosis with co-occurring psychological conditions, and (3) SAD as an additional diagnosis when another disorder is principal. In particular, we sought to examine how diagnostic status relates to self-reported symptom dimensions related to SAD, symptoms associated with frequently comorbid conditions (e.g., depression), and transdiagnostic symptoms that are closely associated with other anxiety disorders. Specifically, we assessed for anxiety sensitivity, the tendency for people to misinterpret physiological symptoms of anxiety as dangerous; this construct is closely related to panic disorder [21]. We also measured intolerance of uncertainty, a discomfort with lack of certainty about future events; this construct is closely related to worry and generalized anxiety [22,23]. Based on available literature and clinical experience, we hypothesized that participants with comorbid conditions would exhibit greater levels of nonspecific psychological symptoms and impairment than those with only SAD. We also expected that participants with principal SAD would have higher levels of SADspecific symptoms compared to those with SAD as an additional diagnosis.
2. Method 2.1. Participants and procedure Participants included 684 psychiatric outpatients presenting for treatment in a Canadian university hospital clinic specializing in the assessment and treatment of anxiety disorders. The sample consisted of 263 men (38.5%) and 420 women (61.5%), ranging in age from 15 to 74 years (M = 36.30, SD = 12.71). Most participants self-identified as White (60.6%), followed by Asian (34.9%), NativeCanadian (1.8%), Black (0.7%), Hispanic (0.4%), and other ethnicities (1.6%). The marital status of the participants was primarily single (44.9%), followed by married (36.8%), cohabitating (9.7%), divorced (4.9%), separated (3.1%), and other (0.7%). The education status of the sample was as follows: less than high school degree (14.7%); high school degree or equivalent (40.5%); college degree (37.4%); at least some graduate or professional education (7.4%). Notably, several participants were missing demographic data about sex (n = 1), ethnicity (n = 7), education status (n = 8), and marital status (n = 4).
1908
A.S. Farmer et al. / Comprehensive Psychiatry 55 (2014) 1906–1913
After receiving a patient referral from the participant's physician (usually their primary care doctor or psychiatrist), each participant was called to schedule an intake interview prior to initiating treatment. During this session, patients completed a series of self-report questionnaires and underwent a clinical interview to determine diagnoses based on the Diagnostic and Statistical Manual, Fourth Edition (DSM-IV) [1]. All participants provided informed, voluntary, written consent for their data to be included in the study. The procedures and measures used in this study were approved by the local institutional review board and incorporated standard assessment practices in the clinic. 2.2. Assessment tools 1. Structured Clinical Interview for DSM-IV Axis I Diagnoses (SCID-I) [24]. This semistructured diagnostic interview was used to assess for the presence of current and lifetime Axis I disorders. This instrument was administered by doctoral-level psychologists, postdoctoral fellows, or advanced graduate students (with thorough training for diagnostic reliability). When criteria for multiple disorders were met, the current disorder causing the most distress and/or functional impairment was identified as the principal diagnosis via a discussion between the patient and clinician, and informed by self-report measures. The interrater reliability of assessors in our clinic was good (kappa = 0.89). 2. Multidimensional Assessment of Social Anxiety (MASA) [19,25]. This 38-item self-report measure was used to assess specific and nonspecific symptom dimensions related to social anxiety, consistent with the hybrid model of social anxiety. The measure consists of six subscales (αs = .72–.91) that assess a specific factor for the diagnostic category of social anxiety (behavioral avoidance, MASA-BA), and five nonspecific factors for related symptoms of anxiety: physiological arousal and avoidance (MASA-PA) thought avoidance (MASA-TA), anhedonia (MASA-ANH), functional impairment (MASA-FI), and coping with substances (MASA-CS). These scales have demonstrated adequate internal consistency, convergent and discriminant validity, and test–retest reliability in clinical and nonclinical samples [19,20,25]. 3. Brief Fear of Negative Evaluation scale (BFNE) [26]. As an additional measure of social anxiety-specific fears, this 12-item self-report questionnaire (α = .97) assessed apprehension about being negatively evaluated due to loss of social approval. This scale has demonstrated adequate reliability, internal consistency, and convergent validity with other measures of social anxiety [26,27]. 4. Depression Anxiety Stress Scales – 21-item version (DASS-21) [28]. This 21-item questionnaire measured
psychological symptoms more broadly, yielding three psychometrically distinct subscales (αs = .82–.91): depression (e.g., dysphoria, hopelessness, inertia, and self-depreciation); anxiety (e.g., autonomic arousal, anxious affect, and situational anxiety); and stress (e.g., difficulty relaxing, nervous tension, irritability, and agitation). Prior literature has found the DASS-21 scales to demonstrate good internal consistency, factor structure, concurrent validity, and reliability [28–30] and scores were doubled to be comparable with the 42-item version [31]. 5. Anxiety Sensitivity Index (ASI) [32]. This 16-item self-report instrument (α = .90) assessed the degree to which a person believes that physiological anxiety symptoms have negative social, cognitive, and/or physical consequences. This scale has good psychometric properties [33]. 6. Intolerance of Uncertainty Scale – 12 item version (IUS-12) [34]. This 12-item self-report measure (α = .93) was used to assess how much a person identifies with the idea that uncertainty about future events is unacceptable and problematic and the extent to which such uncertainty inhibits action. Psychometric properties have been demonstrated in prior research [34]. 7. Illness Intrusiveness Ratings Scale (IIRS) [35]. As a measure of functional impairment due to psychiatric symptoms, this 13-item instrument (α = .87) assessed perceived interference of any symptoms across domains considered integral to quality of life (e.g., health, work, relationships). The IIRS scale has demonstrated adequate reliability, validity, and sensitivity to change in various clinical populations [36].
2.3. Statistical analyses We first conducted statistical analyses to examine differences in demographic and symptomological variables among the diagnostic groups. For categorical variables (e.g., gender, comorbid diagnosis occurrence), we summarized data with absolute (n) and relative (%) values, and we investigated group differences with Pearson chi-squared tests. For continuous variables (e.g., age, subscale scores), we summarized measures with means and standard deviations, and we investigated group differences in symptom measures with one-way analyses of variance (ANOVAs), with Tukey HSD posthoc tests for bivariate comparisons. Lastly, we conducted a backward stepwise logistic regression analysis to identify predictors of a principal SAD diagnosis when participants present comorbid diagnoses, allowing us to examine relative contribution of variables. We began by including all variables in the full model for which diagnostic status had a significant effect and then used a backwards elimination approach by systematically removing nonsignificant variables [37]. All statistical analyses were performed with IBM SPSS 19.0 software.
A.S. Farmer et al. / Comprehensive Psychiatry 55 (2014) 1906–1913
1909
frequency of specific phobia, depressive disorders, or bipolar disorder between diagnostic groups.
3. Results 3.1. Prevalence of SAD and comorbid diagnoses In this treatment-seeking sample, over half of participants had a diagnosis of SAD (n = 354, 51.8%). Of these, 31 participants had no comorbidities (i.e., SAD-only; 4.5% of total sample), while 323 participants had comorbid diagnoses. SAD was the principal diagnosis for 172 participants who received more than one diagnosis (i.e., principal SAD; 25.1%). SAD was an additional diagnosis for 151 participants who had another condition as the principal diagnosis (i.e., additional SAD; 22.1%). Table 1 describes the demographic characteristics by diagnostic group. With regard to demographic variables, groups significantly differed in age and sex, as well as a marginally significant effect of relationship status. Posthoc tests revealed that the SAD-only group tended to be significantly younger than the additional SAD group (p = .034), but the principal and additional SAD groups did not differ (p = .54). Participants in the SAD-only group were also more likely to be male compared to the additional SAD group (p = .029), while participants with additional SAD were marginally more likely to be married or cohabiting than those with principal SAD or SAD-only (ps b .05). Groups did not differ in ethnicity, education attainment, or age of onset of SAD symptoms. Table 1 also lists the most frequent comorbid Axis I diagnoses (N 5%) in participants with a SAD diagnosis and chi-squared statistics comparing frequencies among participants with principal SAD and additional SAD. Overall, participants with additional SAD were more likely to report symptoms meeting criteria for comorbid anxiety disorders, including panic disorder with or without agoraphobia, obsessive-compulsive disorder, and generalized anxiety disorder. There were no significant differences in the
3.2. Clinical profiles Table 2 describes group means and diagnostic group differences in specific social anxiety symptoms, nonspecific and related symptom dimensions, transdiagnostic symptoms relevant to other anxiety disorders, as well as the interference of psychiatric symptoms. The effect of diagnostic group was statistically significant for all clinical measures except for marginal significance for intolerance of uncertainty (p = .051). Posthoc analyses were conducted to examine bivariate differences. As predicted, the SAD-only group presented with less severe nonspecific symptoms associated with social anxiety, transdiagnostic symptoms related to other anxiety conditions, and symptoms of broad anxiety and depression. With regard to mood and anxiety symptoms broadly, the SADonly group had significantly lower scores on DASS-21Depression (ps b .001) and DASS-21-Anxiety (ps b .01), but higher scores on DASS-21-Stress (ps b .001), compared with both the principal and additional SAD groups. The SAD-only group also had lower scores on the ASI and IIRS than participants in both comorbid groups (ps b .05). However, differences were less consistent with nonspecific symptoms. The SAD-only group had lower scores on MASA-PA (p = .004) and IUS-12 (p = .039), and marginally lower scores on MASA-TA (p = .051) than the additional SAD group but not the principal SAD group. The SAD-only group also had lower scores on MASA-ANH (p = .003) and MASA-CS (p = .048) than the principal SAD group but not the additional SAD group. Among SAD participants with comorbid diagnoses, the principal and additional groups significantly differed on a
Table 1 Diagnostic status group differences in demographic characteristics of patients with social anxiety disorder (SAD). SAD-only n = 31 Age – mean (SD) Age of SAD onset Sex (% female) Ethnicity/race (% white) Marital status (% married/cohabitating) Education (% with college degree) Comorbid Axis I Diagnoses Panic disorder Obsessive-compulsive disorder Generalized anxiety disorder Specific phobia Posttraumatic stress disorder Major depressive disorder Bipolar disorder Substance use disorder
a
29.60 (9.99) 12.93 (7.70) 15 (48.4%) a 17 (54.8%) 10 (32.3%) a 17 (54.8%)
Principal SAD n = 172 a,b
33.99 (11.88) 13.17 (7.85) 98 (57.0%) a,b 108 (63.5%) 54 (32.0%) a 65 (38.0%) 41 (23.8%) 37 (21.5%) 75 (43.6%) 42 (24.4%) 11 (6.4%) 109 (63.4%) 18 (10.5%) 32 (18.6%)
Additional SAD n = 151 b
35.38 (11.87) 14.01 (10.19) 103 (68.7%) b 93 (62.4%) 67 (4.4%) 62 (41.6%) 66 55 92 46 18 98 10 18
(43.7%) (36.4%) (60.9%) (30.5%) (11.9%) (64.9%) (6.6%) (11.9%)
Difference Statistic F[2,351] = 3.18 F[2,328] = 3.18 χ 2 = 6.96 χ 2 = 0.85 χ 2 = 5.63 χ 2 = 3.11 χ2 χ2 χ2 χ2 χ2 χ2 χ2 χ2
= = = = = = = =
a
.043 n.s. .031 n.s. .060 n.s. b.001 .003 .002 n.s. .083 n.s. n.s. .098
4.33 8.78 9.66 1.48 3.00 0.08 1.50 2.75
Summary statistics are means (with standard deviations) for age, and number (with percent) of participants meeting criteria for each criterion. identified with same letter are not significantly different from each other (p b .05) in posthoc tests.
p-value
a,c
Group means
1910
A.S. Farmer et al. / Comprehensive Psychiatry 55 (2014) 1906–1913
number of social anxiety specific and transdiagnostic measures. As predicted, the principal SAD group reported more severe symptoms specific to SAD. They tended to have higher scores on the BFNE (p = .014), social avoidance (MASA-BA; p b .001), and functional impairment (MASAFI; p = .015) compared to the additional SAD group. In contrast, the additional SAD group had higher scores on nonspecific symptom measures: MASA-TA (p =.001), DASS-21-Anxiety ( p = .032), and ASI ( p = .022). Surprisingly, principal and additional groups did not differ on the IIRS (p = .621), suggesting similar levels of psychological symptom intrusiveness of into the quality of life. 3.3. Predicting principal SAD diagnosis Lastly, we conducted an analysis to determine predictors of a principal SAD diagnosis in the presence of comorbid conditions. In a logistic regression predicting diagnostic status (principal = 1, additional = 0), six symptom dimensions remained significantly associated with a principal diagnosis of SAD (no demographic remained significant). These results (summarized in Table 3) determined that participants were more likely to have a principal diagnosis of SAD when they reported greater fear of negative evaluation, social avoidance, and coping with substances; participants were more likely to have additional SAD if they reported greater thought avoidance, anxiety sensitivity, or intolerance of uncertainty. For example, all else being equal, for every point increase on the MASA-BA scale, participants with comorbid diagnoses were 10% more likely to have a principal diagnosis of SAD. Based on comparison of log likelihood ratios, the predictive model was significantly better at predicting SAD diagnostic status than chance (χ 2 = 74.20, p b .001), accurately categorizing 70.2% of participants. However, this is only a 17.5% improvement over the null model, suggesting that other factors also play a role in diagnostic status.
4. Discussion The present study addressed the clinical presentation of comorbidity in SAD among people presenting for treatment to an anxiety disorders specialty clinic. Similar to previous studies in epidemiological and clinical samples, we found SAD only (i.e., in the absence of comorbid disorders) to account for a small percentage of SAD diagnoses (8.8% of participants with SAD). However, contrary to past reports of very low occurrence of SAD as the principal diagnosis when comorbid conditions were present [6], we found SAD to be approximately equally likely to be the principal diagnosis and an additional diagnosis. Notably, previous studies defined the principal diagnosis as the primary reason for treatment. Our use of severity and impairment as the criterion for principal diagnosis suggests that SAD symptoms are as likely to be acutely problematic as symptoms of comorbid conditions when they co-occur. It is important to highlight that our sample was drawn from a tertiary service provider. Since people with SAD have been found to be more likely to seek help initially for problems other than SAD, it makes sense that our sample has a higher rate of comorbidity than the general population. Furthermore, the 8.8% SAD-only rate is lower than epidemiological estimates of 15–20% [3–5]. Since our participants were presenting to an anxiety disorders clinic, it is probable that their anxiety conditions were more likely to be the most severe diagnosis compared to people presenting to less specialized providers. This finding has potential significance to clinical practice by informing identification of this often underdiagnosed condition. Although the majority of patients with SAD speak to their provider about comorbid psychiatric symptoms first, about 75% of them express interest in treatment for SAD when offered directly [38]. However, providers would need to know to ask about and recognize symptoms of SAD, which may be more difficult when comorbid symptoms are
Table 2 Diagnostic Status group differences in symptom profiles of patients with social anxiety disorder (SAD). SAD-only n = 31 Brief Fear of Negative Evaluation Multidimensional Assessment of Social Anxiety Behavioral avoidance Physiological arousal and avoidance Anhedonia Functional impairment Thought avoidance Coping with substances Depression Anxiety Stress Scales – 21 items Depression Anxiety Stress Intolerance of Uncertainty Scale – 12 items Anxiety Sensitivity Scale Illness Intrusiveness Ratings Scale
40.90 (6.89)
a,b
Principal SAD n = 172 41.02 (8.12)
a
Additional SAD n = 151 38.27 (9.62)
b
Difference statistic F[2,349] = 4.25
p-value .015
39.93 (7.28) a,b 6.84 (3.53) a 16.16 (5.72) a 18.29 (3.48) a,b 19.87 (6.31) a,b 7.81 (4.31) a
42.06 (7.89) a 8.39 (3.53) a,b 20.31 (6.50) b 19.17 (3.86) a 20.22 (6.72) a 10.09 (5.22) b
37.93 (8.17) b 9.08 (3.69) b 19.03 (6.28) a,b 17.87 (4.63) b 22.92 (6.55) b 8.36 (4.69) a,b
F[2,350] F[2,350] F[2,350] F[2,350] F[2,350] F[2,350]
= = = = = =
10.80 5.44 6.08 3.95 7.58 6.20
b.001 .005 .003 .020 .001 .002
12.84 (9.59) 9.55 (7.51) 16.19 (7.68) 37.74 (9.06) a 24.54 (10.79) 50.45 (10.6)
23.20 (10.86) a 15.37 (9.09) 10.09 (5.22) a 41.95 (10.94)a,b 32.81 (11.97) 59.36 (14.4) a
21.17 (10.62) a 17.94 (9.38) 8.36 (4.69) a 42.87 (10.51) b 36.52 (13.14) 57.80 (15.6) a
F[2,350] F[2,350] F[2,350] F[2,346] F[2,346] F[2,349]
= = = = = =
12.48 11.64 12.39 3.00 12.77 4.86
b.001 b.001 b.001 .051 b.001 .008
Summary statistics are means, with standard deviations in parentheses. a,b,cGroup means identified with same letter are not significantly different from each other (p b .05) based on post-hoc Tukey's HSD tests.
A.S. Farmer et al. / Comprehensive Psychiatry 55 (2014) 1906–1913 Table 3 Demographic and symptom variables predicting principal social anxiety disorder.
Brief Fear of Negative Evaluation MASA – Behavioral avoidance MASA – Thought avoidance MASA – Coping with substances Anxiety Sensitivity Scale Intolerance of Uncertainty Scale
Odds ratio (95% CI)
p-value
1.05 (1.03–1.07) 1.10 (1.08–1.12) 0.93 (0.91–0.95) 1.10 (1.07–1.13) 0.96 (0.95–0.97) 0.97 (0.95–0.98)
.006 b.001 b.001 .001 .001 .050
CI = confidence interval; MASA = Multidimensional Assessment of Social Anxiety.
present, particularly when those symptoms are also severe and impairing. Consistent with our predictions, we found greater severity of associated symptoms (e.g., depression, anxiety sensitivity) and illness intrusiveness in participants with comorbid diagnoses in addition to SAD. Notably, the presence of comorbid diagnoses was not associated with severity of SAD-specific symptoms (e.g., fear of negative evaluation, social avoidance), contrary to past research [5,11]. Compared to earlier epidemiological studies, our sample was seeking treatment for anxiety symptoms; thus, the overall high degree of severity and impairment from psychological symptoms in this population may obscure an effect of comorbidity on severity. This study adds to prior knowledge by examining differences among people with SAD as the principal diagnosis or an additional diagnosis when other conditions co-occur. Participants in the principal SAD group presented with more severe fear of negative evaluation, social avoidance, and use of substances to cope with anxiety, but less severe transdiagnostic symptoms of thought avoidance, anxiety sensitivity, and intolerance of uncertainty compared to the additional SAD group. Despite statistically significant differences, all specific and nonspecific symptoms were elevated across the three groups. These associations provide further support for a hybrid model of SAD [19], as specific symptoms of social avoidance and fear of negative evaluation distinguished people with principal SAD diagnoses, but nonspecific symptoms were present, though to different degrees, across the diagnostic groups. Although we did not further distinguish among specific comorbid diagnoses, it is noteworthy that depression symptoms were not different across the two comorbid groups, and a mood disorder diagnosis was similarly likely. Nonspecific anxiety symptoms and thought avoidance were greatest in the additional SAD group, consistent with their greater likelihood of meeting criteria for panic disorder, obsessive-compulsive disorder, and generalized anxiety disorder. This difference may suggest that comorbid anxiety conditions are more likely to supersede the severity of SAD symptoms. However, because our sample was drawn from an anxiety disorders specialty clinic, this is likely an effect of this population presenting for treatment for primarily anxiety symptoms.
1911
Across psychiatric disorders, comorbidity tends to require more treatment planning and/or more complex therapeutic approaches, which may tax mental health providers' resources with additional need for time and training. Comorbidity may also contribute to poorer responses to treatment or greater likelihood of relapse [7]. Some initial research comparing treatment outcomes of SAD only and SAD with comorbid disorders found poorer prognosis when mood disorders were present, though not with comorbid anxiety disorders [39]. The high frequency of nonspecific mood and anxiety symptoms across our sample (including the SAD-only group) supports the use of recently developed transdiagnostic treatment approaches that target both specific and nonspecific anxiety and mood symptoms [40,41]. Though some evidence suggests greater decreases in comorbidity with such treatments [42], a single study examining equivalence of transdiagnostic and traditional diagnosis-specific treatment in reducing specific social anxiety symptoms was inconclusive [43]. This study also has potential implications for future research on SAD. Often, investigators recruit participants with only principal diagnoses to study underlying dysfunction, response to interventions, and mechanisms of change. With regard to SAD, our results support past evidence that comorbidity tends to be the rule rather than the exception. The small (though significant) differences in social anxiety symptom presentation among the three groups suggest that diagnostic status (e.g., principal vs. additional) may be less important than knowing whether a person's symptoms meet criteria for particular disorders. However, further research is necessary to clarify the nature and implications of symptom overlap between SAD and other conditions. The present study did not include any self-report symptom measures of other conditions in our study, but such measures may help understand phenotypic variation in anxiety disorder cases. Additionally, future research may examine how underlying physiological processes, emotion regulation strategies, and daily experiences may differ across different diagnostic statuses of SAD. With regard to demographic predictors of diagnostic status, we found several interesting differences. Participants with SAD-only tended to be younger but only compared to the additional SAD group. Though this study is crosssectional in nature, this finding lines up with research on SAD typically preceding the development of comorbid conditions since our groups did not differ in age of onset of SAD [11,44]. Longitudinal studies could provide information about how the clinical course of symptoms relates to decision-making regarding principal diagnoses at the outset of treatment. Additionally, prospective studies might elucidate how diagnostic status at the outside of treatment relates to treatment outcome. This study had a number of strengths over previous research on features associated SAD. First, our large treatment-seeking sample provided sufficient data to distinguish among these three diagnostic groups. Second, we used
1912
A.S. Farmer et al. / Comprehensive Psychiatry 55 (2014) 1906–1913
a validated assessment of participants' symptoms with a clinical interview by advanced trainees and mental health professionals. Third, our measures allowed us to examined both diagnosis-specific and transdiagnostic symptom patterns. Nevertheless, a limitation of this study is the largely demographically homogenous sample. Also, in conducting clinical interviews as part of the research study, future studies may also collect data on the participants' specific reason for seeking treatment. This information may help make comparisons with previous findings and investigate corroboration between clinician and patient views on the severity and interference of symptoms across the diagnostic groups.
5. Conclusion The present findings highlight the disorder-specific and nonspecific clinical features associated with the diagnostic status of SAD. Specifically, this pattern supports a hybrid model of SAD [19], with which clinicians can conceptualize a patient's problems on multiple dimensions common across diagnoses while also considering the severity of distinguishing features. Such a model may minimize the reliance on multiple diagnoses to represent the range of patients' symptoms. This study also presents novel information about the frequency of SAD as a principal (most severe, impairing) diagnosis in a treatment-seeking population. Given that people with this diagnosis are likely to be hesitant to mention psychological difficulties, particularly social fears, proper diagnosis may require querying for these concerns even if a patient initially raises other anxiety or mood difficulties.
References [1] American Psychiatric Association. Diagnostic and statistical manual of mental disorders: DSM-IV-TR. Washington, DC: American Psychiatric Publishing, Inc.; 2000. [2] Kessler RC, Berglund P, Demler O, Jin R, Merikangas KR, Walters EE. Lifetime prevalence and age-of-onset distributions of DSM-IV disorders in the National Comorbidity Survey Replication. Arch Gen Psychiatry 2005;62:593-602. [3] Lecrubier Y. Comorbidity in social anxiety disorder: impact on disease burden and management. J Clin Psychiatry 1998;59:33-8. [4] Merikangas K, Angst J. Comorbidity and social phobia: evidence from clinical, epidemiologic, and genetic studies. Eur Arch Psychiatry Clin Neurosci 1995;244:297-303. [5] Ruscio AM, Brown TA, Chiu WT, Sareen J, Stein MB, Kessler RC. Social fears and social phobia in the USA: results from the National Comorbidity Survey Replication. Psychol Med 2008;38:15-28. [6] Zimmerman M, McGlinchey JB, Chelminski I, Young D. Diagnostic co-morbidity in 2300 psychiatric out-patients presenting for treatment evaluated with a semi-structured diagnostic interview. Psychol Med 2008;38:199-210. [7] Angst J. Comorbidity of anxiety, phobia, compulsion and depression. Int Clin Psychopharmacol 1993;8:21-5. [8] Beesdo K, Bittner A, Pine DS, Stein MB, Holfer M, Lieb R, et al. Incidence of social anxiety disorder and the consistent risk for secondary depression in the first three decades of life. Arch Gen Psychiatry 2007;64:903-12.
[9] Perugi G, Simonini E, Savino M, Mengali F, Cassano GB, Akiskal HS. Primary and secondary social phobia: psychopathologic and familial differentiations. Compr Psychiatry 1990;31:245-52. [10] Weiller E, Bisserbe JC, Boyer P, Lepine JP, Lecrubier Y. Social phobia in general health care: an unrecognised undertreated disabling disorder. Br J Psychiatry 1996;168:169-74. [11] Schneier FR, Johnson J, Hornig CD, Liebowitz MR, Weissman MM. Social phobia: comorbidity and morbidity in an epidemiologic sample. Arch Gen Psychiatry 1992;49:282-8. [12] Wittchen HU, Fuetsch M, Sonntag H, Müller N, Liebowitz M. Disability and quality of life in pure and comorbid social phobia. Findings from a controlled study. Eur Psychiatry 2000;15:46-58. [13] Dunner DL. Management of anxiety disorders: the added challenge of comorbidity. Depress Anxiety 2001;13:57-71. [14] Magee WJ, Eaton WW, Wittchen HU, McGonagle KA, Kessler RC. Agoraphobia, simple phobia, and social phobia in the National Comorbidity Survey. Arch Gen Psychiatry 1996;53:159-68. [15] Van Ameringen M, Mancini C, Styan G, Donison D. Relationship of social phobia with other psychiatric illness. J Affect Disord 1991;21:93-9. [16] American Psychiatric Association. Diagnostic and statistical manual of mental disorders. 5th ed. Arlington, VA: American Psychiatric Publishing; 2013. [17] Brown TA, Barlow DH. Dimensional versus categorical classification of mental disorders in the fifth edition of the Diagnostic and statistical manual of mental disorders and beyond: comment on the special section. J Abnorm Psychol 2005;114:551-6. [18] McGlinchey JB, Zimmerman M. Examining a dimensional representation of depression and anxiety disorders' comorbidity in psychiatric outpatients with item response modeling. J Abnorm Psychol 2007;116:464-74. [19] Gros DF, Simms LJ, Antony MM. A hybrid model of social phobia: an analysis of social anxiety and related symptoms of anxiety. J Clin Psychol 2011;67:293-307. [20] Gros DF, McCabe RE, Antony MM. Using a hybrid model to investigate the comorbidity and symptom overlap between social phobia and the other anxiety disorders and unipolar mood disorders. Psychiatry Res 2013;210:188-92. [21] Reiss S. Expectancy model of fear, anxiety, and panic. Clin Psychol Rev 1991;11:141-53. [22] Buhr K, Dugas MJ. Investigating the construct validity of intolerance of uncertainty and its unique relationship with worry. J Anxiety Disord 2006;20:222-36. [23] Dugas MJ, Buhr K, Ladouceur R. The role of intolerance of uncertainty in etiology and maintenance. In: Heimberg RG, Turk CL, & Mennin DS, editors. Generalized anxiety disorder: Advances in research and practice. New York: Guilford Press; 2004. p. 143-63. [24] First MB, Spitzer RL, Williams JB. Structured Clinical Interview for DSM-IV Axis I Disorders: Clinician version, administration booklet. American Psychiatric Pub; 1997. [25] Gros DF, Simms LJ, Antony MM, McCabe RE. Development and Psychometric Evaluation of the Multidimensional Assessment of Social Anxiety (MASA). J Clin Psychol 2012;68:432-47. [26] Leary MR. A brief version of the Fear of Negative Evaluation Scale. Pers Soc Psychol Bull 1983;9:371-5. [27] Carleton RN, McCreary DR, Norton PJ, Asmundson GJ. Brief Fear of Negative Evaluation Scale—revised. Depress Anxiety 2006;23:297-303. [28] Lovibond PF, Lovibond SH. The structure of negative emotional states: comparison of the Depression Anxiety Stress Scales (DASS) with the Beck Depression and Anxiety Inventories. Behav Res Ther 1995;33:335-43. [29] Brown TA, Chorpita BF, Korotitsch W, Barlow DH. Psychometric properties of the Depression Anxiety Stress Scales (DASS) in clinical samples. Behav Res Ther 1997;35:79-89. [30] Clara IP, Cox BJ, Enns MW. Confirmatory factor analysis of the Depression–Anxiety–Stress Scales in depressed and anxious patients. J Psychopathol Behav Assess 2001;23:61-7.
A.S. Farmer et al. / Comprehensive Psychiatry 55 (2014) 1906–1913 [31] Antony MM, Bieling PJ, Cox BJ, Enns MW, Swinson RP. Psychometric properties of the 42-item and 21-item versions of the Depression Anxiety Stress Scales in clinical groups and a community sample. Psychol Assess 1998;10:176-81. [32] Peterson RA, Heilbronner RL. The anxiety sensitivity index: construct validity and factor analytic structure. J Anxiety Disord 1987;1:117-21. [33] Zinbarg RE, Brown TA, Barlow DH, Rapee RM. Anxiety sensitivity, panic, and depressed mood: a reanalysis teasing apart the contributions of the two levels in the hierarchical structure of the Anxiety Sensitivity Index. J Abnorm Psychol 2001;110:372-7. [34] Carleton RN, Norton MAPJ, Asmundson GJG. Fearing the unknown: a short version of the Intolerance of Uncertainty Scale. J Anxiety Disord 2007;21:105-17. [35] Devins GM, Binik YM, Hutchinson TA, Hollomby DJ, Barré PE, Guttmann RD. The emotional impact of end-stage renal disease: importance of patients' perception of intrusiveness and control. Int J Psychiatry Med 1983;13:327-43. [36] Devins GM. Using the Illness Intrusiveness Ratings Scale to understand health-related quality of life in chronic disease. J Psychosom Res 2010;68:591-602. [37] Menard S. Applied logistic regression analysis. Sage; 2002.
1913
[38] Zimmerman M, Chelminski I. Clinician recognition of anxiety disorders in depressed outpatients. J Psychiatr Res 2003;37: 325-33. [39] Erwin BA, Heimberg RG, Juster H, Mindlin M. Comorbid anxiety and mood disorders among persons with social anxiety disorder. Behav Res Ther 2002;40:19-35. [40] Norton PJ. A randomized clinical trial of transdiagnostic cognitivebehavioral treatments for anxiety disorder by comparison to relaxation training. Behav Ther 2012;43:506-17. [41] Barlow DH, Allen LB, Choate ML. Toward a unified treatment for emotional disorders. Behav Ther 2004;35:205-30. [42] Chamberlain LD, Norton PJ. An evaluation of the effects of diagnostic composition on individual treatment outcome within transdiagnostic cognitive–behavioral group therapy for anxiety. Cogn Behav Ther 2013;42:56-63. [43] Norton PJ, Barrera TL. Transdiagnostic versus diagnosis-specific CBT for anxiety disorders: a preliminary randomized controlled noninferiority trial. Depress Anxiety 2012;29:874-82. [44] Chartier MJ, Hazen AL, Stein MB. Lifetime patterns of social phobia: a retrospective study of the course of social phobia in a nonclinical population. Depress Anxiety 1998;7:113-21.