Development and validation of the brief-satisfaction with appearance scale for systemic sclerosis

Authors

  • Lisa R. Jewett,

    1. McGill University and Lady Davis Institute of the Jewish General Hospital, Montréal, Québec, Canada
    Search for more papers by this author
  • Marie Hudson,

    1. McGill University and Lady Davis Institute of the Jewish General Hospital, Montréal, Québec, Canada
    Search for more papers by this author
  • Jennifer A. Haythornthwaite,

    1. Johns Hopkins University School of Medicine, Baltimore, Maryland
    Search for more papers by this author
  • Leslie Heinberg,

    1. Cleveland Clinic Lerner College of Medicine of Case Western Reserve University, Cleveland, Ohio
    Search for more papers by this author
  • Fredrick M. Wigley,

    1. Johns Hopkins University School of Medicine, Baltimore, Maryland
    Search for more papers by this author
    • Dr. Wigley has received consultancy fees, speaking fees, and/or honoraria (less than $10,000 each) from Actelion Pharmaceuticals and Orion Pharmaceuticals, and (more than $10,000 each) from Novartis Pharmaceuticals and United Therapeutics.

  • Murray Baron,

    1. McGill University and Lady Davis Institute of the Jewish General Hospital, Montréal, Québec, Canada
    Search for more papers by this author
    • Dr. Baron is the director of the Canadian Scleroderma Research Group

  • Brett D. Thombs,

    Corresponding author
    1. McGill University and Lady Davis Institute of the Jewish General Hospital, Montréal, Québec, Canada
    • Jewish General Hospital, 4333 Cote Saint Catherine Road, Montreal, Québec, Canada H3T 1E4
    Search for more papers by this author
  • The Canadian Scleroderma Research Group

    Search for more papers by this author
    • Investigators of the Canadian Scleroderma Research Group are listed in Appendix A.


Abstract

Objective

Body image concerns are understudied in systemic sclerosis (SSc; scleroderma). The objective was to develop and cross-validate a brief version of the Satisfaction with Appearance Scale (SWAP) in order to reduce item redundancy, increase SSc relevancy, and improve the feasibility of body image assessment in SSc.

Methods

Female patients with SSc in a development sample (Johns Hopkins Scleroderma Center) and a validation sample (Canadian Scleroderma Research Group Registry) completed the 14-item SWAP. Items for the 6-item Brief-SWAP were selected based on theoretical considerations and psychometric data from the development sample. In both samples, internal consistency reliability, convergent validity, and the hypothesized 2-factor structure (perceived social impact and subjective dissatisfaction) were compared between the Brief-SWAP and SWAP.

Results

Two hundred seventeen women from the development sample and 654 women from the validation sample completed the SWAP. Cronbach's alpha for the Brief-SWAP was 0.82 in both samples, compared with 0.90 and 0.91 for the full SWAP. Correlations between the Brief-SWAP and SWAP were 0.94 and 0.95 in the development and validation samples, respectively. All correlations of the Brief-SWAP and SWAP with measures of convergent validity were substantively equal with no statistically significant differences in either sample. Based on confirmatory factor analysis, model fit for the Brief-SWAP was good in the development (χ2[4] = 9.0, comparative fit index [CFI] = 0.99, Tucker-Lewis Index [TLI] = 0.99, root mean square error of approximation [RMSEA] = 0.07) and validation samples (χ2[4] = 19.5, CFI = 0.99, TLI = 0.99, RMSEA = 0.08) and better than for the SWAP.

Conclusion

The Brief-SWAP is a reliable and valid measure of body image dissatisfaction and social discomfort in SSc that is shorter and more feasibly implemented than the SWAP.

INTRODUCTION

Systemic sclerosis (SSc; scleroderma) is a chronic autoimmune connective tissue disease characterized by abnormal fibrotic processes and excessive production of collagen, which manifests itself in thickening of the skin and fibrosis of internal organs, including the lungs, heart, and gastrointestinal tract (1). The rate of disease onset is highest between 30 and 50 years of age, with risk for women being 4–5 times that for men (2, 3). Median survival time from diagnosis is ∼11 years, and patients are 3.7 times more likely to die within 10 years of diagnosis (44.9% mortality) than age-, sex-, and race-matched individuals without SSc (12.0% mortality) (3).

SSc often results in disfiguring physical changes that commonly occur in visible and socially relevant areas of the body, including the face, mouth, and hands (1, 4–7). Patients with acquired disfigurement from disease or injury often struggle with concerns about appearance and body image, with many experiencing difficulty in maintaining healthy social interactions (8). Body image is a multifaceted construct that comprises body image satisfaction, social comfort related to appearance, investment in appearance, and behavioral tendencies associated with body image concerns, among other dimensions (9). Most measurement tools of body image constructs focus on eating disorders and weight- and size-related distress, including measures of body image avoidance (the Body Image Avoidance Questionnaire [10]) and importance of appearance (e.g., the Appearance Schemas Inventory [11]). The only validated measure of body image designed for individuals with disfigurements from injury or illness is the Satisfaction with Appearance Scale (SWAP) (12). This tool measures social discomfort and dissatisfaction with body parts that are not necessarily related to weight and size issues.

Studies that have investigated body image in SSc have found that patients with more severe SSc (e.g., more significant skin changes in the hands) reported higher levels of body image dissatisfaction and low appearance self-esteem, which were in turn associated with depressive symptoms and reduced psychosocial functioning (13). Two studies on body image in patients with SSc (5, 6) used the appearance subscale of the State Self-Esteem Scale (14), which provides an overall estimate of self-esteem related to appearance, but is a general measure that is not validated in acquired disfigurement. A third study (13) employed the 14-item SWAP (12), which was developed to assess body image concerns related to disfigurement from burn injuries. A 15-item version of the SWAP, the Adapted SWAP (ASWAP) (15), has 1 additional item, “My appearance makes others feel uncomfortable,” and was validated for SSc. Both the SWAP and the ASWAP items load on 2 factors, perceived social impact and subjective dissatisfaction.

A concern with the SWAP and ASWAP is that many of the items are highly redundant. Eight items make the same inquiry about satisfaction with different body parts (e.g., “I am satisfied with the appearance of my arms,” “I am satisfied with the appearance of my chest”), only some of which are focal for patients with SSc. Several other items repeat the same question about social comfort in different social settings (e.g., among family, friends, or strangers). A shorter version of the SWAP that performs as well as the original SWAP, but with less redundancy and with a focus on body parts of particular relevance to SSc (e.g., face, hands), would increase the feasibility of body image assessment in SSc and reduce unnecessary patient burden. The objective of the current study was to develop a 6-item Brief-SWAP, with 3 items on each of 2 factors, perceived social impact and subjective dissatisfaction. To do this, we selected items for the Brief-SWAP from the original SWAP using a development sample of 217 women with SSc from the US. We then tested the reliability and validity of the Brief-SWAP compared with the original SWAP in a validation sample of 654 female patients with SSc from Canada.

PATIENTS AND METHODS

Development and validation samples.

The development sample consisted of patients treated for SSc at the Johns Hopkins Scleroderma Center who enrolled in a longitudinal study examining psychosocial adjustment to SSc that took place between 1997 and 2002 (13). Patients had a diagnosis of SSc based on American College of Rheumatology (ACR; formerly the American Rheumatism Association) criteria (16). The study was approved by the Johns Hopkins University School of Medicine Internal Review Board, and informed consent was obtained from all participants.

The validation sample consisted of patients enrolled in the 15-center Canadian Scleroderma Research Group (CSRG) Registry between 2004 and 2009. To be eligible for the CSRG Registry, patients must have a diagnosis of SSc made by a CSRG Registry rheumatologist, be ≥18 years of age, and be fluent in English or French. Specific diagnostic criteria were not required for enrollment. ACR criteria (16), published in 1980, have been shown to be outdated. Subsequent classification systems have been proposed, but none has gained universal approval (17). As such, an objective of the CSRG Registry is to improve upon existing diagnostic systems. All CSRG Registry patients provided informed consent, and the research ethics board of each study site approved the data collection protocol.

Only female patients were included in this study due to the relatively small number of male patients. In some instances, there were patients in both samples who had completed study measures more than once, but for the purposes of the present study, only data from the first administration were analyzed.

Self-report questionnaires.

The SWAP (12) and Short Form McGill Pain Questionnaire (SF-MPQ) (18) were administered in both the development and validation samples; the Beck Depression Inventory (BDI) (19) in only the development sample; and the Patient Health Questionnaire-9 (PHQ-9) (20), Center for Epidemiologic Studies Depression Scale (CES-D) (21), and the Short Form 36 (SF-36) Health Survey (22) only in the validation sample. Self-report questionnaires were administered in English in the development sample and in English or French in the validation sample.

The SWAP.

The 14-item SWAP (12) was developed to measure non–weight-related body image dissatisfaction among burn survivors. The 15-item ASWAP, which was validated for SSc, contains 14 items from the original SWAP and 1 additional item. The SWAP was used in the current study because the ASWAP had not been published when data collection began in the validation sample. Respondents to the SWAP rate the degree to which they feel each item reflects their thoughts and feelings about their appearance on a 7-point scale ranging from 1 to 7, where 1 = strongly disagree and 7 = strongly agree. The SWAP has a 2-factor structure, with perceived social impact reflecting social discomfort and subjective dissatisfaction reflecting dissatisfaction with various body parts. Higher scores indicate greater body image dissatisfaction. The total SWAP score is calculated by subtracting 1 from each item in order to anchor items at 0 and then totaling the item scores, which is equivalent to scoring items 0–6. A previous study with the development sample reported an internal consistency reliability of the SWAP total score of α = 0.90 (13).

The BDI.

The 21-item BDI (19) has been widely used in studies of patients with chronic illness. Each item has 4 possible answers, scored 0–3, indicating increasing symptom severity. Respondents are instructed to describe the way they have been feeling during the past week. The authors of the BDI recommend a cutoff score of ≥10 for at least mild symptoms of depression (23). The BDI had good internal consistency reliability (α = 0.90) in the development sample (13).

The CES-D.

The 20-item CES-D assesses the frequency of depressive symptoms during the past week on a 0–3 Likert-type scale (ranging from rarely or none of the time to most or all of the time). Total scores range from 0 to 60. Standard cutoffs are ≥16 for possible depression and ≥23 for probable depression. The CES-D demonstrated good internal consistency reliability (α = 0.87) and convergent validity with related self-report measures in a sample of 470 patients with SSc from the CSRG Registry (21).

The PHQ-9.

The 9-item PHQ-9 rates the frequency of depressive symptoms over the past 2 weeks on a 0–3 Likert-type scale (ranging from not at all to nearly every day). The total score ranges from 0 to 27 and the standard cutoff threshold for moderate depression severity is a score of ≥10 (24–27). A previous study of 566 patients from the CSRG Registry found that the PHQ-9 had good internal consistency reliability (α = 0.87) and similar convergent validity indices compared with the CES-D (28).

The SF-36.

The SF-36 (22, 29) is a widely used and evaluated health outcomes measure, with well-established validity and reliability in multiple populations (21). It is made up of 8 self-report domains and can be summarized into mental component summary (MCS) and physical component summary (PCS) scores. These summaries are scored using norm-based scoring from a general population sample to produce T scores for each patient (mean ± SD of 50 ± 10). Higher scores indicate better functioning.

The SF-MPQ.

The SF-MPQ (18) is a 15-item measure of sensory (11 items) and affective (4 items) dimensions of pain, where higher scores indicate greater pain. Each descriptor is ranked on a 4-point intensity scale (ranging 0–3, where 0 = none and 3 = severe), and total scores range from 0 to 45. The SF-MPQ has been used extensively and has excellent psychometric properties (30). Cronbach's alpha was 0.91 in the development sample and 0.89 in the validation sample.

Statistical analysis.

Development of the Brief-SWAP.

Items were selected for the Brief-SWAP based on theoretical and psychometric considerations. Three items were selected for each subscale because at least 3 items are needed to assess the factor validity of the subscales. Items on the 3-item subjective dissatisfaction subscale were chosen to reflect the body parts most salient to disfigurement from SSc (face, hands, and arms). Items related to the scalp, neck, legs, chest, and overall appearance were not included. For the social discomfort subscale, items were selected based on psychometric considerations, including variance of item responses, item-total correlations, and the effect of removing an item on internal consistency reliability, as well as theoretical considerations. For instance, we chose only 1 of 3 items that assessed social comfort with the same question and 3 different social contexts (among family, friends, and strangers).

Assessment of reliability and validity of the 6-item Brief-SWAP compared with the 14-item SWAP in development and validation samples.

In the development and validation samples, Cronbach's alpha was computed for both the Brief-SWAP and SWAP to measure internal consistency reliability. Pearson's correlations were computed to assess convergent validity with other outcome measures. It was predicted that the Brief-SWAP and SWAP would both correlate moderately with the measures of depressive symptoms (BDI, CES-D, and PHQ-9 scores) and quality of life (SF-36 MCS and PCS scores), but less strongly with the measure of pain (SF-MPQ). To compare the psychometric characteristics of the Brief-SWAP and the full 14-item SWAP, 95% confidence intervals (95% CIs) were computed for the correlations of each with measures of convergent validity, as well as for the differences between Brief-SWAP and SWAP correlations with each of the other measures.

Confirmatory factor analysis was conducted with Mplus (31) to assess the validity of the previously reported (15) 2-factor structure (subjective dissatisfaction and perceived social impact) for both the SWAP and Brief-SWAP in the development and validation samples. Item responses for the SWAP and Brief-SWAP were ordinal Likert data, so the weighted least squares estimator with a diagonal weight matrix, robust SEs, and a mean- and variance-adjusted chi-square statistic was used with delta parameterization (31). Modification indices were used to identify pairs of items within scales for which model fit would improve if error estimates were freed to covary and for which there appeared to be theoretically justifiable shared method effects (32). A chi-square goodness-of-fit test and 3 fit indices were used to assess model fit, including the Tucker-Lewis Index (TLI) (33), the comparative fit index (CFI) (34), and the root mean square error of approximation (RMSEA) (35). Because the chi-square test is highly sensitive to sample size and can lead to the rejection of well-fitting models, practical fit indices were emphasized (36). Guidelines proposed by Hu and Bentler (37) suggest that models with TLI and CFI close to 0.95 or higher and the RMSEA close to 0.06 or lower are representative of good-fitting models. A CFI of ≥0.90 (38) and a RMSEA of ≤0.08 (39) may also be considered to represent a reasonably acceptable model fit.

RESULTS

Development and validation sample characteristics.

There were 217 women in the Johns Hopkins Scleroderma Center development sample and 654 women in the CSRG validation sample. Sociodemographic variables, medical variables, and self-report questionnaire scores for both samples are displayed in Table 1. The mean ± SD age of the development sample was 52.6 ± 13.0 years (range 27–78 years) and the mean ± SD age of the validation sample was 57.1 ± 11.3 years (range 18–84 years). Sociodemographic variables in the development and validation samples were similar with the exception of employment status. Only 22% of the validation sample was employed, whereas 41% of the development sample was employed. Approximately 70% of patients in both samples had limited SSc. The mean ± SD time since diagnosis of SSc in the development sample was 7.5 ± 7.5 years compared with 8.4 ± 9.2 years in the validation sample.

Table 1. Sociodemographic and disease variables for female patients with scleroderma*
 Johns Hopkins Scleroderma Center (n = 217)Canadian Scleroderma Research Group (n = 654)
  • *

    Values are the mean ± SD unless otherwise indicated. SSc = systemic sclerosis (scleroderma); MRSS = modified Rodnan skin thickness score; Brief-SWAP = Brief-Satisfaction with Appearance Scale; SWAP = Satisfaction with Appearance Scale; BDI = Beck Depression Inventory; PHQ-9 = Patient Health Questionnaire-9; CES-D = Center for Epidemiologic Studies Depression Scale; SF-36 = Short Form 36 Health Survey; MCS = mental component summary; PCS = physical component summary; SF-MPQ = Short Form McGill Pain Questionnaire.

Demographic variables  
 Age, years52.6 ± 13.057.1 ± 11.3
 White, no. (%)165 (76.4)190 (89.6)
 College or higher, no. (%)136 (64.2)307 (47.7)
 Employed, no. (%)86 (40.6)142 (22.0)
 Married, no. (%)130 (60.2)450 (69.6)
Medical variables  
 Time since diagnosis of SSc, years7.5 ± 7.58.4 ± 9.2
 Limited scleroderma, no. (%)152 (70.0)472 (72.2)
 MRSS score (range 0–51)9.5 ± 8.6
Self-report questionnaire scores  
 Brief-SWAP13.6 ± 8.413.5 ± 8.8
 SWAP28.9 ± 17.428.3 ± 18.5
 BDI11.3 ± 8.4
 PHQ-96.1 ± 5.5
 CES-D13.9 ± 10.8
 SF-36 MCS48.5 ± 11.7
 SF-36 PCS37.7 ± 10.9
 SF-MPQ9.2 ± 8.96.0 ± 5.5

Brief-SWAP development.

The 3 items that were selected for the subjective dissatisfaction subscale of the Brief-SWAP, based on relevance for SSc, included items related to dissatisfaction with the face, hands, and arms. For the perceived social impact subscale, the 3 items that assessed discomfort in the presence of family, friends, and strangers correlated robustly with each other. The item related to social discomfort in the presence of strangers was retained, and the other 2 were removed because the items for friends and family had low endorsement rates and limited item variance. The item “Changes in my appearance have interfered with my relationships” (r = 0.49) was removed because it had the lowest item-total correlation among the other 3 items on the factor in the original SWAP. Thus, the other 2 items that were retained for the perceived social impact factor of the Brief-SWAP related to feeling unattractive and to believing that other people would “not want to touch me.” In the development sample, all 6 items on the final version of the Brief-SWAP had item-total correlations from 0.53 to 0.62 (they were from 0.54 to 0.65 in validation sample). The correlation between the 3-item subjective dissatisfaction and perceived social impact subscales for the Brief-SWAP was r = 0.48 (95% CI 0.37, 0.58) in the development sample and r = 0.46 (95% CI 0.40, 0.52) in the validation sample. The correlation between the 8-item subjective dissatisfaction subscale and the 6-item perceived social impact subscale for the SWAP was r = 0.46 (95% CI 0.35, 56) in the development sample and r = 0.45 (95% CI 0.39, 0.51) in the validation sample.

Assessment of reliability and validity of the 6-item Brief-SWAP compared with the 14-item SWAP.

The correlation between the Brief-SWAP and SWAP was 0.94 in the development sample and 0.95 in the validation sample. Internal consistency reliability as measured by Cronbach's alpha was 0.82 for the Brief-SWAP in both the development and validation samples. Cronbach's alpha for the full SWAP was 0.90 in the development sample and 0.91 in the validation sample.

Table 2 shows Pearson's correlations calculated for both the Brief-SWAP and the SWAP with the BDI and SF-MPQ in the development sample, and the CES-D, PHQ-9, SF-36 MCS, SF-36 PCS, and SF-MPQ in the validation sample, as well as differences in Pearson's correlations for the Brief-SWAP and SWAP compared with each measure. As hypothesized, the Brief-SWAP and SWAP correlated moderately with all measures of depressive symptoms and quality of life and, to a slightly lesser degree, with a measure of pain. There were no significant or substantive differences between the correlations of the Brief-SWAP and SWAP with any of the measures in either the development or validation samples.

Table 2. Convergent validity for the Brief-SWAP and SWAP*
 Johns Hopkins Scleroderma Center (n = 217)Canadian Scleroderma Research Group (n = 654)
Brief-SWAPSWAPDifferenceBrief-SWAPSWAPDifference
  • *

    Values are the r (95% confidence interval). All correlations are significant (P < 0.01). None of the differences in correlations between the Brief-SWAP and SWAP were statistically significant (P < 0.05). See Table 1 for definitions.

BDI0.52 (0.41, 0.61)0.53 (0.43, 0.62)−0.01 (−0.21, 0.18)
PHQ-90.39 (0.32, 0.45)0.44 (0.38, 0.50)−0.06 (−0.16, 0.04)
CES-D0.41 (0.34, 0.47)0.47 (0.41, 0.53)−0.07 (−0.18, 0.03)
SF-36 MCS−0.33 (−0.26, −0.40)−0.39 (−0.32, −0.45)0.07 (−0.04, 0.17)
SF-36 PCS−0.32 (−0.25, −0.39)−0.35 (−0.28, −0.41)0.03 (−0.07, 0.14)
SF-MPQ0.34 (0.21, 0.46)0.32 (0.19, 0.44)0.02 (−0.18, 0.21)0.27 (0.20, 0.34)0.27 (0.20, 0.34)0.00 (−0.10, 0.10)

Confirmatory factor analysis was used to test the hypothesized 2-factor (subjective dissatisfaction and perceived social impact) structure for the Brief-SWAP and the SWAP in both the development and validation sample. In all analyses, 2 pairs of item error covariances were freed based on modification indices. In each case, both members of the pair demonstrated shared method or format features. Error variances were freed to covary for 1) a pair of items referring to satisfaction with particular body parts (“I am satisfied with the appearance of my face” and “I am satisfied with the appearance of my hands”), and 2) a pair of items referring to social discomfort in relation to other people (“I feel that my scleroderma is unattractive to others” and “I don't think that people would want to touch me”). In the development sample, model fit for the Brief-SWAP was good (χ2[4] = 9.0, CFI = 0.99, TLI = 0.99, RMSEA = 0.07) and somewhat better than for the SWAP (χ2[25] = 115.8, CFI = 0.95, TLI = 0.98, RMSEA = 0.13). In the validation sample, model fit for the Brief-SWAP was again good (χ2[4] = 19.5, CFI = 0.99, TLI = 0.99, RMSEA = 0.08) and better than for the SWAP (χ2[25] = 391.2 CFI = 0.95, TLI = 0.98, RMSEA = 0.15). Factor loadings for the Brief-SWAP and SWAP items in both the development and validation samples were similar, as were correlations between the subjective dissatisfaction and perceived social impact factors (Table 3).

Table 3. Factor loadings of the Brief-SWAP and SWAP*
ItemsJohns Hopkins Scleroderma Center (n = 217)Canadian Scleroderma Research Group (n = 654)
Brief-SWAPSWAPBrief-SWAPSWAP
  • *

    Values are the factor loadings. See Table 1 for definitions.

Perceived social impact factor    
 Because of changes in my appearance caused by my scleroderma, I am uncomfortable in the presence of my family.0.860.92
 Because of changes in my appearance caused by my scleroderma, I am uncomfortable in the presence of my friends.0.960.99
 Because of changes in my appearance caused by my scleroderma, I am uncomfortable in the presence of strangers.0.800.870.900.91
 Changes in my appearance have interfered with my relationships.0.760.74
 I feel that my scleroderma is unattractive to others.0.900.790.750.73
 I don't think people would want to touch me.0.860.720.670.68
Subjective dissatisfaction factor    
 I am satisfied with my overall appearance.0.820.86
 I am satisfied with the appearance of my scalp.0.580.68
 I am satisfied with the appearance of my face.0.990.890.950.90
 I am satisfied with the appearance of my neck.0.830.87
 I am satisfied with the appearance of my hands.0.850.610.860.69
 I am satisfied with the appearance of my arms.0.670.870.750.86
 I am satisfied with the appearance of my legs.0.610.79
 I am satisfied with the appearance of my chest.0.820.86
Correlation of perceived social impact and subjective dissatisfaction factors0.540.560.620.55

DISCUSSION

The 6-item Brief-SWAP, which was developed in a sample of 217 women with SSc from the US and cross-validated in a sample of 654 women with SSc from Canada, performed similarly to the full 14-item SWAP in both samples. The correlations of the Brief-SWAP and full SWAP with convergent validity measures were substantively identical. Analysis with confirmatory factor analysis found that the Brief-SWAP replicated the original 2 factors of the full SWAP, subjective dissatisfaction and perceived social impact, and that the model fit the data somewhat better than for the full SWAP. The better fit of the 2-factor confirmatory factor analysis model for the Brief-SWAP compared with the full 14-item SWAP may have occurred because the Brief-SWAP focused on body parts of particular relevance to SSc and because it eliminated items from the full 14-item SWAP that were endorsed by small numbers of patients.

Cronbach's alpha was 0.82 for the Brief-SWAP in both the development and validation samples, compared with α = 0.90–0.91 for the full SWAP. Ideally, Cronbach's alpha will be between 0.70 and 0.90; Cronbach's alpha results of ∼0.90 or greater are considered to reflect a high level of item redundancy and indicate that items may be removed from a scale (40). This is consistent with the findings from the current study, which showed that the SWAP could be cut from 14 to 6 items without weakening convergent validity substantively. The development of the Brief-SWAP is an example of how an existing measure can be adapted and made more efficient and feasibly administered, reducing the burden to researchers and patient respondents.

Research examining body image concerns in acquired disfigurement, specifically SSc, is limited. Existing studies of body image in SSc have largely relied on general measures developed for other populations, which assess general constructs such as self-esteem related to appearance (5, 6). Experts in research on body image, however, have emphasized the need for measures that assess specific dimensions of body image (e.g., dissatisfaction with appearance, social discomfort, investment in appearance, and behavioral tendencies associated with body image disturbance) that are appropriately adapted for specific patient groups, and are subsequently validated in these groups (9). The current study advances previous work validating the longer ASWAP (15) by validating the Brief-SWAP, which similarly assesses dissatisfaction with appearance and social discomfort. Another important area where research is needed in SSc relates to behavioral tendencies associated with body image distress, or body image avoidance (10). Body image avoidance refers to the avoidance of objects and/or situations because they elicit body image distress or concerns (41). Among individuals with visible disfigurement from injury or illness, body image avoidance can often lead to an acute fear of negative evaluation and social anxiety (42, 43); therefore, this is an area that merits further investigation in the future.

There are limitations that should be considered when interpreting the results from the current study. Both the development and validation samples were convenience samples of patients receiving treatment at the Johns Hopkins Scleroderma Center or from the CSRG centers. Therefore, limitations associated with this sampling strategy should be considered. The results reported here may depart to some degree from what might be found in other settings and for different patient groups. For instance, patients not being cared for by a rheumatologist and those too sick to participate were not included in the present study. It is therefore possible that the samples used in the current study may consist of an overrepresentation of healthier patients, and the results may not be generalizable to the full spectrum of patients with SSc. The current study only assessed female patients with SSc due to the relatively small number of men who completed the study measures; therefore, results may not be generalizable to men with SSc. Similarly, the majority of the patients sampled had limited SSc, meaning that most individuals had less severe disfigurement.

In terms of the Brief-SWAP, it is possible that some of the excluded items, for instance, items related to dissatisfaction with the neck and chest, may be relevant to the disfigurement experienced by some patients with SSc. Nonetheless, the psychometric properties of the Brief-SWAP with the 3 selected items related to subjective dissatisfaction were quite robust. It was not possible to examine test–retest reliability for the Brief-SWAP because annual patient visits occurred a year apart, which is too long to determine whether changes in scores might have been due to body image changes versus test–retest variation. Furthermore, Brief-SWAP items were administered as part of the full 14-item SWAP, and it is possible that order effects could have influenced responses (44). Demographic correlates of the Brief-SWAP and SWAP were not examined. A study using structural equation modeling to examine relationships between Brief-SWAP scores and demographic and disease correlates is currently being conducted.

In summary, the Brief-SWAP is a reliable and valid measure of dissatisfaction with appearance and social discomfort related to disfigurement among patients with SSc. Compared with the full 14-item SWAP, the 6-item Brief-SWAP reduced item redundancy, increased relevance to the experience of patients with SSc, and demonstrated good psychometric properties, including reliability and validity, thus providing a less burdensome and more feasibly administered scale. The results from the current study constitute a significant step toward the improvement of measurement of important body image constructs for SSc. In the future, more research to identify other relevant dimensions of body image that are significant to patients is needed, in addition to a move toward developing and testing interventions that target these body image concerns.

AUTHOR CONTRIBUTIONS

All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be submitted for publication. Dr. Thombs had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study conception and design. Jewett, Haythornthwaite, Thombs.

Acquisition of data. Hudson, Haythornthwaite, Wigley, Baron, Thombs.

Analysis and interpretation of data. Jewett, Hudson, Heinberg, Baron, Thombs.

ROLE OF THE STUDY SPONSOR

None of the sponsors of the Canadian Scleroderma Research Group (CSRG) had any role in the study design, data collection, data analysis, or writing of the manuscript. No CSRG sponsor was consulted in relation to this study, none reviewed the manuscript submitted for publication, and publication was not contingent upon sponsor approval.

APPENDIX A:

CANADIAN SCLERODERMA RESEARCH GROUP INVESTIGATORS

M. Baron, Montréal, Québec; J. Pope, London, Ontario; J. Markland, Saskatoon, Saskatchewan; D. A. Masetto, Sherbrooke, Québec; E. Sutton, Halifax, Nova Scotia; N. A. Khalidi, Hamilton, Ontario; D. Robinson, Winnipeg, Manitoba; N. Jones, Edmonton, Alberta; E. Kaminska, Hamilton, Ontario; P. Docherty, Moncton, New Brunswick; C. D. Smith, Ottawa, Ontario; J.-P. Mathieu, Montréal, Québec; M. Hudson, Montréal, Québec; S. Ligier, Montréal, Québec; T. Grodzicky, Montréal, Québec; S. Mittoo, Winnipeg, Manitoba; and M. Fritzler, Advanced Diagnostics Laboratory, Calgary, Alberta, Canada.

Ancillary