The Body Image Concern Inventory: Validation in a multiethnic sample and initial development of a Spanish language version

Abstract

Dysmorphic appearance concern encompasses preoccupation with a perceived appearance defect, defect checking and camouflaging, and social avoidance. The current study sought to evaluate the internal consistency, factor structure, and convergent validity of a measure of dysmorphic appearance concern, the Body Image Concern Inventory, as well as evaluate the psychometric properties of a Spanish version of the instrument. Women recruited as part of a reproductive clinic based clinical trial completed the BICI and other self-report measures of distress. A total of 1,043 women completed the measures in English (M = 29 years, range = 18–55 years) and 573 women completed the measures in Spanish (M = 32 years, range = 18–55 years). Both the English and Spanish BICI were internally consistent and correlated moderately with measures of current psychological distress (STAI-S, CES-D). Confirmatory factor analyses replicated the measure’s proposed factor structure. Applications of the BICI for future research are discussed.

Dysmorphic appearance concern is emerging as a key construct in body image research. Broader than dissatisfaction with appearance, dysmorphic appearance concern also includes intense preoccupation with one’s appearance and perceived appearance flaws as well as a number of associated behaviors. These include checking behaviors, such as weighing oneself, measuring oneself, or examining oneself in a mirror; attempts to camouflage perceived flaws, such as through clothing, body posture, or cosmetic products; and frequent appearance-related comparisons. Dysmorphic appearance concern may also include avoidance of activities due to concerns about one’s appearance as well as avoidance of viewing oneself (e.g., covering mirrors; Littleton, Axsom, & Pury, 2005; Oosthuizen, Lambert, & Castle, 1998; Reas, Whisenhunt, Netemeyer, & Williamson, 2002). Dysmorphic appearance concern is not restricted solely to concerns about weight and body shape, and indeed, individuals may have several areas of concern (Bartsch, 2007; Cansever, Uzun, Dönmez, & Oz ahin, 2003). Frequently noted areas of concern in addition to those related to weight or body shape are: skin, teeth, face shape, nose size or shape, feet, genitals, lips, and hair (Altamura, Paluello, Mundo, Medda, & Mannu, 2001; Bartsch, 2007; Cansever et al., 2003).

Dysmorphic appearance concern is the hallmark of body dysmorphic disorder (BDD), which is characterized by a preoccupation with an imagined or exaggerated defect in appearance that leads to distress or impairment in functioning (American Psychiatric Association, 1994; Phillips, 1991). Indeed, a survey by Phillips, McElroy, Keck, Pope, and Hudson (1993) of individuals with BDD found very high rates of dysmorphic appearance concern. For example, they found that 73% of the individuals with BDD engaged in excessive mirror checking and 97% reported avoiding social or occupational activities because of concerns about their appearance. Similarly, a recent survey of psychiatric inpatients by Jorgensen, Castle, Roberts, and Groth-Marnat (2001) found that patients with a diagnosis of BDD reported significantly higher dysmorphic appearance concern than patients without a diagnosis of BDD. BDD itself appears to be much more prevalent than previously believed, particularly in young adult samples. College surveys have found prevalence rates for BDD of 2–5% (Bartsch, 2007; Bohne et al., 2002; Cansever et al., 2003), population survey prevalence rates have ranged from 0.7% to 1.1% (Otto, Wilhelm, Cohen, & Harlow, 2001; Phillips & Crino, 2001), and, among inpatient samples, prevalence rates have ranged from 7% to 13% (Dyl, Kittler, Phillips, & Hunt, 2006; Grant, Kim, & Crow, 2001).

Dysmorphic appearance concern also appears to be strongly associated with eating disorders. For example, one study found that 33% of a sample of individuals with BDD had a history of a clinical or subclinical eating disorder (Ruffolo, Phillips, Menard, Fay, & Weisberg, 2006). In addition, a study of adolescents with anorexia or bulimia found that 25% of the young women with anorexia had a history of BDD (Jolanta & Tomasz, 2000). Studies of individuals with eating disorders have found that these individuals frequently report symptoms of dysmorphic appearance concern (Grant & Phillips, 2004; Mazzeo, 1999). Indeed, a study by Rosen and Ramirez (1998) found that individuals with eating disorders reported similar levels of dysmorphic appearance concern as individuals with BDD.

Subclinical dysmorphic appearance concern also appears to be fairly common, particularly among young adults. For example, Fitts, Gibson, Redding, and Deiter (1989) found that 46% of a sample of college students reported being at least somewhat preoccupied with a perceived defect in their appearance. Similarly, Bartsch (2007) found that 30% of a sample of Australian college students was preoccupied with a perceived defect in their appearance.

The apparent prevalence of dysmorphic appearance concern at clinical and subclinical levels suggests a need for the development of reliable and valid measures of this construct. However, until recently, there were few available instruments to assess dysmorphic appearance concern, particularly self-report measures. In addition, there is little information available about the psychometric properties of extant self-report measures of dysmorphic concern. For example, there is no published data available about the psychometric properties of a self-report version of an extant interview measure of dysmorphic appearance concern, the Body Dysmorphic Disorder Examination (Rosen & Reiter, 1996). While there is data available about the convergent validity, internal consistency, and test-retest reliability of the Body Image Disturbance Questionnaire in college samples (Cash & Grasso, 2005; Cash, Phillips, Santos, & Hrabosky, 2004; Rudiger, Cash, Roehrig, & Thompson, 2007), there is no published information about the construct validity of this measure, its performance in non-college samples, its convergent validity with interviewer administered measures, or its sensitivity or specificity with regard to clinical diagnoses.

One particularly promising measure that has recently been developed to assess this construct is the Body Image Concern Inventory (BICI; Littleton et al., 2005). The BICI is a brief self-report measure designed to assess multiple aspects of dysmorphic appearance concern, as opposed to only focusing on appearance dissatisfaction or one aspect of dysmorphic appearance concern (e.g., appearance-related checking). Thus, in addition to appearance dissatisfaction, items tap checking of appearance defects, attempts to camouflage defects, reassurance seeking related to defects, and engaging in appearance-related comparisons. Items also assess interference in functioning due to appearance concerns (e.g., missing social activities because of concerns), and avoidance of viewing oneself. Research supported the internal consistency, convergent validity with regard to self-report and interviewer administered measures, and hypothesized factor structure of this measure in college samples (Littleton et al., 2005). In addition, the BICI was found to have excellent sensitivity (96%) and reasonable specificity (67%) in classifying individuals with a current diagnosis of bulimia or BDD from individuals with subclinical levels of symptomatology. Expert raters also supported the measure’s construct validity (Littleton et al., 2005).

While initial studies of the BICI suggest that it is a reliable and valid instrument for assessing dysmorphic appearance concern, research on the measure is limited to college samples. Thus, the primary aim of the current study was to evaluate the internal consistency and factor structure of this instrument in a multi-ethnic sample of primarily low socioeconomic status (SES) women. In addition, to further establish the measure’s convergent validity, its relationship with measures of general distress and unhealthy dieting behaviors was evaluated. In addition, a secondary aim was to provide initial evaluation of a Spanish version of the instrument to facilitate assessment of dysmorphic concern in diverse populations.

Method

Participants

A total of 1043 women provided complete survey data in English. English survey participants were 29 years old on average (SD = 9.3 years, range = 18–55 years). Fifty-one percent self-identified as Hispanic or Latina, 34% as European American, and 16% as African American. Eighty-seven percent reported that they were born in the United States. With regard to the highest educational level completed, 20% reported that they did not complete high school, 34% reported that they completed high school or obtained a GED, 31% reported some postsecondary education, and 15% did not complete this item or marked several responses. Seventy-five percent reported an annual household income of less than $25,000, with 24% reporting an annual household income of less than $5,000. A total of 69% of respondents reported that they had dependent children in the home (M = 2.1, range = 1–12) and six percent reported that they lived in public housing.

A total of 573 women provided complete survey data in Spanish. Spanish survey participants were 32 years old on average (SD = 8.1 years, range = 18–55 years). Ninety-nine percent self-identified as Hispanic or Latina. Two percent reported that they were born in the United States, 83% reported that they were born in Mexico, and the remaining 14% reported that they were born in other countries. With regard to the highest educational level completed, 69% reported that they did not complete high school, 13% completed high school or obtained a GED, 10% reported some postsecondary education, and 8% did not complete this item or marked several responses. Sixty-seven percent reported an annual household income of less than $25,000, with 12% reporting an annual household income of less than $5,000. A total of 87% of respondents reported that they had dependent children in the home (M = 2.4, range = 1–7) and 10% reported that they lived in public housing.

Procedures

Data for the present study were collected between October 31, 2006 and July 27, 2007 as part of an ongoing prospective clinical trial addressing cervical cancer prevention. Latina, European American, or African American women between 18 and 55 years of age who presented to one of four Regional Maternal & Child Health Program (RMCHP) clinics in southeast Texas for a well-woman/annual exam were screened for eligibility. At the time of screening, women who were pregnant, had a diagnosis of cervical cancer, were unable to understand spoken English or Spanish, or were under the age of 18 or over the age of 55 were excluded from participation. During the recruitment period captured in the present study, approximately 90% of those eligible agreed to participate and underwent informed consent for the study. Women were reimbursed $5 for their participation. All study procedures and measures were approved by the Institutional Review Board of the University of Texas Medical Branch, Galveston.

As part of the screening survey, participants completed a demographic questionnaire and completed several items assessing recent unhealthy dieting behaviors including whether they had used any of the following strategies in the past 30 days to lose weight: took diet pills, skipped meals, or used liquid replacement meals. Participants also completed measures assessing their current depressive and anxious symptomatology and completed the BICI.

Measures

Body Image Concern Inventory

The Body Image Concern Inventory (BICI) is a 19-item self-report measure designed to assess dysmorphic appearance concern (Littleton et al., 2005). For each item, individuals indicated how often they have the described feeling or perform the described behavior on a 5-point Likert scale bounded by 1 (never) and 5 (always). Cronbach’s alpha for the measure in a college sample was .93. Factor analyses supported a two factor structure of two highly correlated factors, the first which tapped dysmorphic appearance concern and the second which tapped interference in functioning due to appearance concerns. The authors recommend computing a total score, however, both subscales showed adequate internal consistency: dysmorphic concern α = .92, interference due to appearance concerns α = .76. Supporting the validity of the measure, scores have been found to correlate moderately to strongly with other self-report and clinician-administered dysmorphic appearance concern measures in college samples (Littleton et al., 2005). Finally, the measure successfully distinguished individuals with subclinical eating disordered or body dysmorphic symptoms from individuals with bulimia or body dysmorphic disorder as determined by structured clinical interview (Littleton et al., 2005).

To develop a Spanish version of the measure, it was translated and back translated by two native speakers with experiencing conducting translations for health surveys. Discrepancies in translation were resolved through conferral.

Center for Epidemiological Studies Depression Scale

The Center for Epidemiologic Studies Depression Scale (CES-D; Radloff, 1977) is a 20-item, self-report measure of depression. For each item, individuals indicated how often they have felt that way in the past week on a 4-point Likert scale bounded by 0 (rarely or none of the time/less than one day) and 3 (most or all of the time/5–7 days). A sample item is, “I felt sad.” Cronbach’s alpha for the measure in community and patient samples have ranged from .84 to .90 and the four-week test-retest reliability was found to be .67 (Radloff, 1977). Supporting the validity of the measure, scores have been found to correlate moderately to strongly with other clinician-administered and self-report measures of depression (Weissman, Sholomskas, Pottenger, Prusoff, & Locke, 1977). In addition, scores have been found to be sensitive to changes in depressive symptomatology following treatment (Weissman et al., 1977).

The Spanish language CES-D used in the present study was developed by the Center for AIDS Prevention Studies at the University of California, San Francisco. The Spanish CES-D has been shown to have good internal consistency (Cronbach’s alpha of .90) and a sensitivity and specificity of .95 and .91, respectively, to the detection of depressive symptomatology (Soler et al., 1997).

State Trait Anxiety Inventory

The Spielberger Stait-Trait Anxiety Inventory (STAI; Spielberger, 1983; English and Spanish versions) was administered to measure anxiety symptoms. In the current study, the state anxiety scale was analyzed. State anxiety is defined as transitory feelings of apprehension, tension, worry, and nervousness accompanied by arousal of the autonomic nervous system (Spielberger, 1983). The state anxiety scale consists of 20 items that reflect how the respondent feels “right now, at this moment” using a 4-point Likert scale bounded by 1 (not at all) and 4 (very much so). Scores are calculated as the sum of the responses (range 20–80). The psychometric adequacy of both versions has been demonstrated, and non-significant differences in mean scores and internal consistency have been shown between the English and Spanish language versions of the STAI (Novy, Nelson, Smith, Rogers, & Rowzee, 1995).

Analysis Plan

Analyses were conducted separately for the English and Spanish language versions of the BICI. Pearson’s correlation coefficients were calculated between participants’ total score on the BICI and their scores on the CES-D and STAI state. Unequal variance t-tests were conducted to compare the total BICI score of participants who did and did not report engaging in unhealthy dieting behaviors. Unequal variance t-tests were chosen because the Student’s t-test has not been found to be robust to violation of the assumption of equal variances when there are unequal sample sizes in the two groups, as in the current investigation (Hays, 1994; Ruxton, 2006).

A univariate ANOVA was conducted to examine differences in BICI scores among participants of different ethnic backgrounds (collapsing across survey versions). A significant ANOVA was followed up with Bonferroni-adjusted pairwise comparisons. A Student’s t-test was conducted comparing the BICI scores of Latina women who were born in the U.S. and those who immigrated to the U.S. Individuals with missing data on a particular measure were excluded from analyses of that measure. In addition, participants with greater than 50% missing data on the survey overall or greater than 50% of missing data on the BICI were considered to have incomplete data and were removed from analyses. A total of 35 individuals were excluded for incomplete data.

Confirmatory factor analyses of the English and Spanish BICI were conducted using the Mplus program (version 4.21; Muthén & Muthén 1998–2007). Consistent with the findings of Littleton and colleagues (2005), a two factor model with correlated factors was evaluated. To evaluate model fit, in addition to the model χ², the root-mean-square error of approximation (RMSEA), the comparative fit index (CFI), the Tucker-Lewis Index (TLI), and the standardized root mean square residual (SRMR) were utilized. The RMSEA assesses model misfit per degree of freedom, and previous studies have indicated that values of approximately .06 or less are indicative of good fit (Hu & Bentler, 1999). The CFI and TLI measure improvement in fit when the model tested is compared to an independence model in which the variables are uncorrelated with one another. Previous studies have indicated that CFI/TLI values close to .95 are indicative of good fit (Hu & Bentler, 1999). The SRMR is a measure of the mean absolute correlation residual, that is the difference between the observed and predicted correlations overall (Kline, 2005). SRMR values of less than .10 or are considered acceptable (Kline, 2005)

Results

English BICI

Nine percent of participants had missing data on one or more of the BICI items, 13% had missing data on one or more of the CES-D items, and 28% had missing data on one or more of the STAI state items. Missing data for specific items on the BICI ranged from less than 1% to 4%. Less than 1% of participants did not complete the unhealthy dieting behaviors items. Participants reported nearly the entire possible range of scores on the English BICI (M = 42.8, SD = 15.0, range = 19–92). Cronbach’s alpha for the full BICI was .94; alpha for the dysmorphic concern subscale was .92, and alpha for the interference due to appearance concerns subscale was .89. Participants’ score on the BICI was moderately correlated with their score on the CES-D, r (912) = .50, p < .001, and their score on the STAI state, r (697) = .45, p < .001. Unhealthy dieting behaviors were fairly frequent in the sample. Sixteen percent (n = 152) reported using diet pills, 19% (n = 180) reported skipping meals, and 11% (n = 101) reported using liquid meal replacements. In all cases, women who reported engaging in unhealthy dieting behaviors scored significantly higher on the BICI than women not reporting these behaviors: using diet pills, t (193) = 7.43, p < .001, ή² = .07, skipping meals, t (249) = 10.73, p < .001, ή² = .12, and using liquid meal replacements, t (121) = 5.16, p < .001, ή² = .03. The BICI scores of participants stratified by unhealthy dieting behaviors are summarized in Table 1.

Table 1.

Means and standard deviations of BICI scores among English and Spanish survey participants stratified by reported unhealthy dieting behaviors in the past thirty days

	English M (SD)		Spanish M (SD)
	Yes	No	Yes	No
Used diet pills	51.9 (16.8)	41.1 (14.0)**	37.4 (13.1)	31.7 (11.6)*
Skipped meals	53.8 (15.5)	40.3 (13.7)**	41.2 (14.2)	31.0 (11.0)**
Used meal replacements	50.6 (16.1)	41.9 (14.7)**	35.4 (13.3)	31.8 (11.7)

^*p < .05.

^**p < .001.

Results of the two-factor CFA are depicted in Figure 1. It should be noted that the error terms of two pairs of items that assessed similar content were allowed to correlate (items 11 with 13 and 2 with 7). The model in general provided adequate fit to the data, χ² (149) = 861.8, p < .001, CFI = .91, TLI = .90, RMSEA = .07, SRMR = .06. In addition, each item loaded significantly on the specified factor. The two factors were highly correlated, r (974) = .74, p < .001. A more parsimonious one-factor model did not provide adequate fit to the data, χ² (152) = 1628.24, p < .001, CFI = .82, TLI = .80, RMSEA = .10, SRMR = .07

Figure 1.

Results of two-factor CFA for English version of the BICI

Spanish BICI

Thirteen percent of participants had missing data on one or more of the BICI items, 20% had missing data on one or more of the CES-D items, and 31% had missing data on one or more of the STAI state items. Missing data for specific items on the BICI ranged from less than 1% to 5%. Less than 2% of participants did not complete the unhealthy dieting behaviors items. Participants reported nearly the entire possible range of scores on the Spanish BICI (M = 32.0, SD = 11.8, range = 19–88). Cronbach’s alpha for the Spanish BICI was .92; alpha for the dysmorphic concern subscale was .90, and alpha for the interference due to appearance concerns subscale was .86. Participants’ score on the BICI was moderately correlated with their score on the CES-D, r (416) = .46, p < .001, and their score on the STAI state, r (353) = .38, p < .001. Unhealthy dieting behaviors were less frequent than in the English sample. Five percent (n = 28) reported using diet pills, 9% (n = 45) reported skipping meals, and 5% (n = 25) reported using liquid meal replacements. Women who reported using diet pills, t (30) = 2.26, p = .031, ή² = .01, and skipping meals, t (50) = 4.68, p < .001, ή² = .06, scored significantly higher on the BICI than women who did not report engaging in these unhealthy dieting behaviors. However, there were no significant differences in the BICI scores of women who reported using and those who did not report using liquid meal replacements, t (26) = 1.30, p = .204, ή² = .00. The BICI scores of participants stratified by unhealthy dieting behaviors are summarized in Table 1. It should be noted that women who completed the BICI in Spanish scored significantly lower than women who completed the BICI in English overall, t (1444) = 13.97, p < .001, d = .77, and significantly lower than Latina women in particular who completed the BICI in English, t (971) = 11.06, p < .001, d = .71.

The initial two-factor model did not provide an adequate fit to the data, χ² (151) = 607.88, p < .001, CFI = .85, TLI = .83, RMSEA = .08, SRMR = .09. Allowing two items from the interference due to appearance concerns factor to load instead on the dysmorphic concern factor (items 3 and 4) improved the model fit. It should also be noted that the error terms for one pair of items that assessed similar content were allowed to correlate (items 11 with 12). This final model is depicted in Figure 2. This model in general provided adequate fit to the data, χ² (150) = 466.24, p < .001, CFI = .90, TLI = .88, RMSEA = .07, SRMR = .07. In addition, each item loaded significantly on the specified factor. The two factors were highly correlated, r (497) = .62, p < .001. A more parsimonious one-factor model did not provide adequate fit to the data, χ² (151) = 970.64, p < .001, CFI = .81, TLI = .79, RMSEA = .09, SRMR = .07.

Figure 2.

Results of modified two-factor CFA for Spanish version of the BICI

BICI Scores and Participant Ethnicity

BICI scores differed significantly among women of the three ethnic groups, F (2, 1443) = 59.17, p < .001. Bonferroni adjusted post-hoc comparisons supported that European American women scored significantly higher on the BICI than Latina or African American women. A Student’s t test comparing Latina women who were born in the U.S. to women who immigrated to the U.S. supported that Latinas who were born in the U.S. scored significantly higher on the BICI than immigrants, t (962) = 10.53, p < .001, d = .69. Participants’ scores on the BICI stratified by ethnicity are summarized in Table 2.

Table 2.

Means and standard deviations of BICI scores among participants of the three ethnic groups and among Latina women born in the U.S. and U.S. immigrants

	M	SD	n
European American	46.75	14.48	321
African American	37.58	12.89	152
Latina	36.79	14.51	973
Born in the U.S.	42.49	15.36	379
U.S. immigrant	33.01	12.42	585

Discussion

This study provides support for the internal consistency and validity of the English language BICI in an ethnically diverse, primarily low SES sample. Overall internal consistency values were acceptable as were internal consistency values on the two subscales. Women’s scores on the BICI were moderately-strongly correlated with their scores on general distress measures. In addition, women who reported recently engaging in unhealthy dieting behaviors scored significantly higher on the BICI than women who did not report engaging in these behaviors. The CFA also replicated the factor structure found by Littleton and colleagues (2005). Thus, overall the results support the validity of the English BICI in a non-college sample.

This study also provides initial support for the Spanish language version of the BICI. The Spanish BICI had good internal consistency (both overall and subscale internal consistency values) and was moderately-highly correlated with the general distress measures. Also, women reporting using diet pills and skipping meals scored significantly higher on the BICI than women not reporting these behaviors. While there were no significant differences in BICI scores between women who did and did not report using liquid meal replacements, this may to some extent reflect low power to detect differences between the two groups as only 5% of the women in the sample reported engaging in this behavior. The CFA in general also replicated the factor structure of the English version. While two items that loaded on the interference due to appearance concerns factor of the English version instead loaded on the dysmorphic concern factor in the final model, the loadings of these two items on the interference factor were statistically significant.

Participants in the present sample, on average, scored somewhat lower on the BICI than previous college samples. While the mean score of the English language BICI was 42.8 in the current sample, the mean in college samples was 50.4. This may reflect the fact that college samples tend to have elevated prevalence rates of eating disordered and dysmorphic symptomatology (e.g., Bartsch, 2007; Bohne et al., 2002; Cansever et al., 2003; Franko, Becker, Thomas, & Herzog, 2007). This pattern of results might also reflect the fact that European American women and higher SES women tend to have higher levels of eating disordered and dysmorphic symptomatology than some ethnic minority women and women of lower SES (Cachelin, Rebeck, Chung, & Pelayo, 2002).

Of note, European American women scored higher on the BICI than Latina or African American women in the present sample. Results also supported that time lived in the U.S. and exposure to mainstream U.S. media may also affect dysmorphic symptomatology among Latina women, as women completing the BICI in Spanish and Latinas who were U.S. immigrants also scored significantly lower than women completing the BICI in English and Latina women born in the U.S. Interestingly, the exact pattern of overall ethnic differences and differences among subgroups of Latina women with regard to body surveillance was found in a recent study of body image among low SES women (Radecki Breitkopf, Littleton, & Berenson, 2007). Body surveillance encompasses concerns about what others think about one’s appearance and thus appears to tap some aspects of dysmorphic concern. In this same study, no ethnic or subgroup differences were found in other aspects of body image less clearly associated with dysmorphic concern, including shame about one’s body and belief that one can control one’s weight/shape.

Future research on the BICI should focus on further establishing its reliability and validity, particularly with male samples. There also is a need to establish cutoffs for clinically significant dysmorphic symptomatology on the BICI in non-college samples (given that a lower cut-off may be needed in a non-college sample due to lower levels of symptomatology overall) as well as validation of the measure in clinical and other samples (e.g., individuals seeking cosmetic or corrective surgery, individuals who have experienced disfigurement). In addition, future work should further establish the validity of the Spanish version of the BICI. In particular, there is a need to evaluate the convergent validity of the Spanish BICI and the ability of the Spanish BICI to differentiate individuals with clinical levels of dysmorphic concern from individuals with non-clinical levels of dysmorphic concern.

Bearing these limitations in mind, extant research supports that the BICI is an internally consistent, valid measure of dysmorphic appearance concern that taps multiple facets of this construct (e.g., appearance related checking, camouflaging, avoidance) and does not focus exclusively on weight and shape concerns. Use of the BICI can potentially advance research on this construct in diverse populations, including Latino populations and low SES samples. With further research establishing the internal consistency and validity of the measure in clinical samples, use of the BICI can also potentially advance understanding of the role of dysmorphic concern in BDD, eating disorders, and other disorders. Similarly, with further research, the BICI could also potentially prove useful as a screening tool in mental health settings as well as medical settings, such as cosmetic surgery clinics or burn treatment centers. Overall, it appears that the BICI is a promising instrument for use in research into dysmorphic concern.

Appendix

1. I am dissatisfied with some aspect of my appearance (Estoy insatisfecho con algunos aspectos de mi apariencia)

2. I spend a significant amount of time checking my appearance in the mirror (Gasto una significativa parte de tiempo evaluando mi apariencia frente al espejo)

3. I feel others are speaking negatively of my appearance (Siento que otros hablan significativamente de mi apariencia)

4. I am reluctant to engage in social activities when my appearance does not meet my satisfaction (Trato de no participar en actividades sociales cuando mi apariencia no me satisface)

5. I feel there are certain aspects of my appearance that are extremely unattractive (Creo que algunos rasgos de mi apariencia son extremadamente poco atractivos)

6. I buy cosmetic products to try to improve my appearance (Compro productos cosméticos para mejorar mi apariencia)

7. I seek reassurance from others about my appearance (Busco la opinión de otros sobre mi apariencia)

8. I feel there are certain aspects of my appearance I would like to change (Siento que podría cambiar ciertos aspectos de mi apariencia)

9. I am ashamed of some part of my body(Me siento avergonzado de alguna parte de mi cuerpo)

10. I compare my appearance to fashion models or others (Comparo mi apariencia con la de modelos y otros)

11. I try to camouflage certain flaws in my appearance (Trato de camuflar ciertos imperfecciones en mi apariencia)

12. I examine flaws in my appearance (Examino las fallas de mi apariencia)

13. I have bought clothing to hide a certain aspect of my appearance (He comprado ropa para esconder algunas partes de mi cuerpo)

14. I feel others are more physically attractive than me (Siento que otros son mas atractivos físicamente que yo)

15. I have considered consulting/consulted some sort of medical expert regarding flaws in my appearance (He considerado consultar a un medico(o he consultado) buscando ayuda relacionada a mis defectos fiscos)

16. I have missed social activities because of my appearance (He perdido actividades sociales por mi apariencia)

17. I have been embarrassed to leave the house because of my appearance (Ha estado avergonzada de salir de mi casa por causa de mi apariencia)

18. I fear that others will discover my flaws in appearance (Tengo miedo a que otros descubran las fallas en mi aspecto físico)

19. I have avoided looking at my appearance in the mirror (He evitado mirarme en el espejo)