The Body Dysmorphic Disorder Symptom Scale: Development and preliminary validation of a self-report scale of symptom specific dysfunction (2024)

Abstract

The Body Dysmorphic Disorder Symptom Scale (BDD-SS) is a new self-report measure used to examine the severity of a wide variety of symptoms associated with body dysmorphic disorder (BDD). The BDD-SS was designed to differentiate, for each group of symptoms, the number of symptoms endorsed and their severity. This report evaluates and compares the psychometric characteristics of the BDD-SS in relation to other measures of BDD, body image, and depression in 99 adult participants diagnosed with BDD. Total scores of the BDD-SS showed good reliability and convergent validity and moderate discriminant validity. Analyses of the individual BDD-SS symptom groups confirmed the reliability of the checking, grooming, weight/shape, and cognition groups. The current findings indicate that the BDD-SS can be quickly administered and used to examine the severity of heterogeneous BDD symptoms for research and clinical purposes.

Keywords: Body dysmorphic disorder, Self-report, Symptom subtypes, Symptom severity, Psychometric properties

Introduction

Measures of body dysmorphic disorder (BDD) and associated symptoms typically fall into one of three categories: measures of overall severity, diagnostic and screening measures, and measures of body image beliefs and satisfaction. The current standard for assessing body dysmorphic disorder, the Yale-Brown Obsessive Compulsive Scale, Modified for BDD (BDD-YBOCS; Phillips et al., 1997), measures the severity of BDD-related obsessions, compulsions, and avoidance. The BDD-YBOCS, a modified version of the original Y-BOCS (Goodman, Price, Rasmussen, Mazure, Delgado, et al., 1989; Goodman, Price, Rasmussen, Mazure, Fleischmann, et al., 1989), has strong psychometric properties and clinical and research utility. Although its total score is practical for treatment outcome research, the BDD-YBOCS does not capture comprehensive information with regard to specific BDD symptoms (e.g., specific cognitions).

The most commonly used measure for screening and diagnosis is the Structured Clinical Interview for DSM (; ), which includes a diagnostic module specific to body dysmorphic disorder. The Body Dysmorphic Disorder Examination (), designed to measure dysmorphic concern in eating disorders, also provides information with regard total severity and diagnostic status yielding a total score and a suggested cutoff for BDD diagnosis. However, its use as a measure for BDD has waned in recent years perhaps due to its particular relevance to eating disorders rather than to BDD specifically. Other screening measures include the Body Dysmorphic Disorder Questionnaire (BDDQ) which consists of yes or no questions reflective of the DSM-IV diagnostic criteria for BDD (). Recognizing the gaps in the field with regard to screening for BDD, Mancuso and colleagues sought to validate the 7-item self-report Dysmorphic Concern Questions (DCQ) as a screening measure for BDD. The DCQ yields information with regard to potential diagnostic status, but similar to the aforementioned measures, does not assess the severity or range of symptoms specific to BDD ().

The extant literature additionally includes several self-report measures concerning body image beliefs and behaviors. Specifically, the Appearance Schemas Inventory-Revised (ASI-R) assesses psychological investment in physical appearance, as well as the importance, meaning, and influence of appearance in one’s life (Cash, 2008; ). The Multidimensional Body-Self Relations Questionnaire (MBSRQ) measures several facets of body image, including evaluation of and orientation toward appearance, fitness, and health (; Cash, 2000). The scale includes subscales related to weight and more general body satisfaction. The more recently developed Body Image Disturbance Questionnaire (BIDQ) measures concern and preoccupation with physical appearance, as well as associated distress, impairment, and avoidance (Cash, 2008; ). Although widely used in body image research, these questionnaires are not specific to the multitude of symptoms that characterize BDD. Rather, they tend to assess more global or eating disorder-specific body image beliefs and behaviors. Body image beliefs and behaviors in individuals with BDD often differ markedly from those of individuals with eating disorders, thus suggesting the need for separate assessments (Hrabosky et al., 2009).

The heterogeneous nature of BDD has received growing attention. Most patients with BDD engage in a range of compulsive behaviors to check, hide, or improve the perceived defect (). Indeed, an additional criterion (B) was added to the newly revised diagnostic criteria to capture such repetitive behaviors (e.g., mirror checking, excessive grooming, skin picking, or surgery seeking) or mental acts (e.g., comparing their appearance with that of others) in BDD (APA, 2013). Similarly, a specifier for muscle dysmorphia was added to further characterize individuals (mostly male) who are preoccupied with body build, considering themselves insufficiently big or muscular.

Despite the usefulness of current assessments for diagnostic, clinical, and research purposes, they do not fully capture the broad range of symptoms that mark BDD. Thus, we designed the Body Dysmorphic Disorder Symptom Scale (BDD-SS, Wilhelm, 2006; ; available upon request) to create a comprehensive instrument that provides a profile of the severity of a wide range of BDD symptoms and one that could be easily administered and interpreted in clinical and research settings. The BDD-SS could prove useful in elucidating the prevalence of specific symptoms and behaviors in BDD. Importantly, the BDD-SS captures several key aspects of BDD (compulsive behaviors, negative appearance-related cognitions, and avoidance) that need assessment, monitoring, and targeting in treatment. This new scale would allow clinicians to monitor progress in treatment over time by specific symptom dimension, rather than through more global severity measures. Such a measure could provide vital information as to how various BDD symptoms respond to interventions and help clinicians to adapt treatments accordingly. Thus, the current study was an initial step in evaluating a BDD self-report measure that could achieve both screening and severity scaling with good psychometrics. Specifically, we assessed the internal consistency of the BDD-SS. We also evaluated the convergent and discriminant validity of the BDD-SS with measures of BDD (BDD-YBOCS and BABS), body image (MBSRQ-AE and MBSRQ-AO), and depressive symptoms (the BDI-II).

Method

Results

The means and standard deviations of study variables were: BDD-YBOCS (M = 30.6, SD = 4.2), BABS (M = 15.2, SD = 4.4), BDI-II (M = 18.4, SD = 11.6), MBSRQ-AE (M = 2.3, SD = 0.78), MBSRQ-AO (M = 4.3, SD = 0.50), BDD-SS severity (M = 36.2, SD = 10.9), and BDD-SS symptoms (M = 26.8, SD = 6.8). The internal consistency for the BDD-SS Symptom scale was KR-20 = .81. The internal consistency for the BDD-SS Severity scale was Cronbach’s α = .75.

Discussion

A growing body of research points to the heterogeneous nature of BDD (APA, 2013; Wilhelm et al., 2014). Measures are available to assess global BDD severity and more general body dissatisfaction. A measure that adequately captures both core symptoms of BDD and unique symptoms that may affect only a subset of individuals (e.g., skin picking, muscle dysmorphia) is needed. In the present article, we introduced the BDD-SS and conducted an initial examination of its psychometric properties. The BDD-SS is a brief, self-report scale that captures a profile of BDD symptoms and their severity levels over a specific time period. Initial results from the development and validation of the BDD-SS suggest that the BDD-SS is a reliable and valid instrument when considering all items. Thus, the BDD-SS could be a useful tool for clinical and research purposes in order to characterize individuals and samples.

The BDD-SS yields two summary scores, a total severity score (the sum of the 7 severity scores) and a total symptom score (the number of specific symptoms endorsed). Supporting their convergent validity, both total scores correlated strongly with the BDD-YBOCS (Phillips et al., 1997). The total BDD-SS symptom score correlated more strongly than the total BDD-SS severity score with the BDD-YBOCS. Thus, despite the moderate correlation between the total BDD-SS symptom and severity scores (r = .55), the former was more closely associated with global BDD severity (BDD-YBOCS), which is independent from the number of symptoms endorsed. It may be that number of symptoms endorsed can be used as a proxy of symptom severity and to gauge treatment effects over time. However, both of the BDD-SS summary scores are based in part on the number of symptoms endorsed, and thus neither should be regarded as a pure measure of severity.

There are limitations to the BDD-SS that should be highlighted. Firstly, there are some disadvantages to the scoring system we used. The BDD-SS summary scores are confounded with the number of symptoms; therefore it is possible that an individual with few types of symptoms that are very severe may, in error, receive a lower total severity and symptom score than an individual with many symptoms, but who is less impaired. This limitation is not unique to the BDD-SS and affects other measures frequently administered by clinicians and researchers (e.g., the Obsessive Compulsive Symptoms Rating Scale; Yovel et al., 2012). Thus, there is precedence for the use of this scoring system in the field. One major advantage of our simple scoring system is that the scores can be quickly and easily calculated without a calculator within seconds, which might enhance the clinical utility of our measure. Rather than calculate a total score, one could also look at the BDD-SS as a profile of item scores used to guide treatment planning. Clinically, we use this approach routinely. In a recently published treatment manual on CBT for BDD, we recommend any symptom group with a severity level of 5 (moderate) should warrant clinical attention (Wilhelm et al., 2013). Items with a severity of 0 provide no evidence of need for clinical attention. This can inform clinical decision making and help to evaluate the effectiveness of treatment. For example, high severity on cosmetic symptoms might guide a clinician to using a modular approach early in treatment incorporating motivational interviewing and psychoeducation about the ineffectiveness of cosmetic treatment, whereas multiple obsessions and compulsions might lead to an integrative CBT approach targeting first the most severe cluster followed by less severe ones.

A second limitation of our measure pertains to some of its psychometrics. Whereas the BDD-YBOCS provides a single global measure of symptom severity, an important aim of the BDD-SS is to capture the severity of a wide range of symptoms associated with BDD. Our results suggest that the BDD-SS is a reliable and valid instrument when considering all items. However, internal consistency decreased when considering some of the specific clusters, especially for those clusters with few items within a cluster internal consistency where outcomes were bound to be lower. Symptom groups of the BDD-SS may require further refinement in order to be adequately captured by a single score. Item selection and symptom clusters for the BDD-SS were theoretically informed. However, the symptom groups of the BDD-SS were not designed to measure distinct symptom constructs and it is possible that some items might fit better within a different symptom group (e.g., “becoming upset by compliments” might fit better under Cognitions than Avoidance). Given the subject to item ratio, factorial analysis is beyond the scope of the current paper; however, future research should empirically assess its structure via factor analysis to determine support for the validity of the specific BDD-SS groups. Some of the internal consistency estimates of measures from our sample are somewhat low, and thus our results need to be interpreted with caution.

Despite its current limitations, the fine-grained assessment structure of the BDD-SS might ultimately help us gain a better understanding of the nature of BDD. In addition to negative cognitions and avoidance, all participants endorsed at least one compulsive behavior, most commonly checking and grooming rituals. This is consistent with prior research (Phillips et al., 2005, 2010) and further supports the recent inclusion of the ritual criterion (Criterion B) in DSM 5. There are likely non-specific symptom types that cut across symptom groups, for example, the severity scores of certain symptoms such as checking compulsions, avoidance, or negative appearance-related cognitions, that may be linked to several broad symptom subtypes. Assessment of common symptom types may benefit research into the etiology of BDD as well an understanding of the mechanisms associated with successful treatment outcome. At the same time, results highlight possible subsets of the disorder. For example, some, but not all participants, endorsed behaviors related to weight/shape, picking/plucking, and surgical/dermatological procedures. The variable endorsem*nt rate of symptoms in this study provides further support for the heterogeneity of BDD and suggests the need for specific assessment and treatment of various symptoms. Indeed, modular cognitive-behavioral therapy for BDD (CBT-BDD; Wilhelm et al., 2013), which addresses both core symptoms of BDD and provides specific strategies to flexibly address symptoms that may affect some but not all individuals with BDD (e.g., cosmetic surgery seeking, hair pulling/skin picking, mood management, and shape/weight/muscularity concerns), has been shown to be efficacious in reducing BDD severity (Wilhelm et al., 2014).

Another limitation of the current study is that it neither includes test–retest reliability data nor data on the change of this measure with treatment. Future studies should focus on the extent to which changes in BDD-SS scores reflect clinically significant improvement that occurs with treatment. In addition, it will be important to establish clinical cut-scores for the distinction between body image dissatisfaction and BDD and to provide a normative sample. Future research on the BDD-SS will also need to examine multiple samples, with adequate composition and sample size, to further establish screening value and severity scaling. These studies should also examine its test–retest reliability and its factorial structure. In addition, the BDD-SS should be examined in the context of known group validity for specific subtypes of BDD, such as muscle dysmorphia.

In sum, our initial results suggest that the BDD-SS appears to be a generally reliable and valid comprehensive measure of BDD. It (1) covers a wide range of BDD and BDD-related symptoms, (2) provides a detailed profile of BDD severity, (3) can be administered within 10 min as a self report measure, (4) can be easily scored and interpreted, and it (5) can be used to target specific symptoms intreatment. Future research is needed to further develop and test the BDD-SS. It is our hope that clinicians and researchers alike will use the information it provides to enhance our understanding of the nature and treatment of BDD.

Acknowledgments

References