تغییرات در تصویر بدن در طول درمان شناختی-رفتاری در زنان مبتلا به بولیمیا

The purposes of this study were: (1) to examine multidimensional aspects of body image of individuals with bulimia nervosa (BN) at pre-treatment, post-treatment, and at follow-up, compared to a group of participants without BN; and (2) to investigate whether measures of body image predicted outcome at post-treatment and follow-up. The clinical sample consisted of 109 females with BN who were enrolled in a 12-week cognitive-behavioral group treatment program. Participants were assessed at baseline, at the completion of treatment, and at 1- and 6-month follow-up visits. The 82 females who comprised the non-bulimic sample were assessed at comparable time intervals. At baseline, the participants with BN reported greater body dissatisfaction and overestimated body size to a significantly greater degree than the comparison group, and reported a significantly smaller ideal size relative to perceived size. Results at the end of treatment indicated significant improvement in self-reported attitudinal disturbance and size overestimation, with continued reductions at follow-up. Logistic regression analyses did not demonstrate a predictive relationship between body image measures at baseline and outcome at post-treatment or follow-up, or between post-treatment and follow-up. Implications for treatment include specifying the source of body image-related distress and enhancing treatment efforts for perceptual and attitudinal aspects of body image.

The importance of body image in bulimia nervosa (BN) is widely recognized, and it has been described as one of the central features of the disorder (Fairburn & Garner, 1988). Dissatisfaction with and concerns about weight and shape are considered significant risk factors for the development of BN (Striegel-Moore, Silberstein, & Rodin, 1986), and have been found to be predictive of the onset of bulimic symptoms in longitudinal studies of eating disorders (Killen et al., 1996; Leon, Fulkerson, Perry, & Cudeck, 1993; Stice & Agras, 1998). In addition, the DSM-IV diagnostic criteria for BN include the overvaluation of shape and weight in self-evaluation (American Psychiatric Association, 1994). Although widely studied, the exact nature of body image in eating disorders remains unclear, in part because of problems in defining and measuring the body image construct. Definitions of body image have varied, but most definitions have generally included a perceptual aspect, referring to the accuracy with which one perceives size and shape, and a cognitive-affective component, involving the evaluative experience of one’s body (Cash & Deagle, 1997; Garner & Garfinkel, 1981–1982; Rosen, 1996). Although these two components are considered to be independent, several investigations have indicated that the perceptual component may be strongly influenced by cognitions and attitudes (Smeets, 1997; Szymanski & Seime, 1997; Williamson, 1996) rather than reflecting a purely visual sensory process. Recent attention has also been paid to understanding body image as a self-schema (Altabe & Thompson, 1996; Kearney-Cooke & Striegel-Moore, 1997). The complexity of the body image construct has made its measurement problematic (see reviews by Ben-Tovim & Walker, 1991; Cash & Brown, 1987; Hsu & Sobkiewicz, 1991; Smeets, Smit, Panhuysen, & Ingleby, 1997; Thompson, 1990;Thompson, Penner, & Altabe, 1990). As a result, greater emphasis has been placed on recognizing the multidimensional nature of body image, and attempting to measure its various components systematically (Cash & Brown, 1987; Thompson et al., 1990). Cash (1990) has even suggested referring to the phenomenon as body “images” in plural form to reflect its complexity and multidimensional nature. Investigations of body image in individuals with bulimia nervosa have focused primarily on three different types of measures: (1) perceived body size, including the accuracy of overall body size estimates and specific body parts; (2) attitudinal evaluation of size, shape, and body parts; and (3) ideal body size, and the discrepancy between perceived and ideal size. Studies of size perception have included variations of the moving caliper technique (MCT) (Slade & Russell, 1973) to assess perceptual accuracy in estimating body parts, as well as video and photographic image distortion methods to evaluate whole-body estimates. The majority of these studies have found that individuals with BN overestimate their body size and parts to a greater extent than non-eating disorder participants when using image distortion methods (Collins et al., 1987; Freeman, Thomas, Solyom, & Koopman, 1985a; Freeman, Thomas, Solyom, & Miles, 1983; Gleghorn, Penner, Powers, & Schulman, 1987; McKenzie, Williamson, & Cubic, 1993; Powers, Schulman, Gleghorn, & Prange, 1987; Whitehouse, Freeman, & Annandale, 1986) and various moving caliper procedures (Bowden, Touyz, Rodriguez, Hensley, & Beumont, 1989; Gleghorn et al., 1987; Norris, 1984; Nudelman, Rosen, & Leitenberg, 1988; Powers et al., 1987; Sunday, Halmi, Werdann, & Levey, 1992; Thompson, Berland, Linton, & Weinsier, 1986; Whitehouse et al., 1986; Willmuth, Leitenberg, Rosen, Fondacaro, & Gross, 1985). However, some studies did not find significant group differences in size estimates (Birtchnell, Lacey, & Harte, 1985; Fernandez, Probst, Meermann, & Vandereycken, 1994; Huon & Brown, 1986; Lindholm & Wilson, 1988; Probst, Vandereycken, van Coppenolle, & Pieters, 1995). In studies that have modified verbal directions to differentiate between affective versus cognitive perception, participants with BN have reported “feeling” larger than they “thought” they looked, and to a greater extent than participants without BN (Bowden et al., 1989; Huon & Brown, 1986; Probst et al., 1995). Inconsistent results in size perception studies may be related to measurement error and method variance in the assortment of instruments that have been utilized (Bowden et al., 1989), as well as the potential influence of other factors on size perception including body weight, mood, menstrual cycle, and recent food consumption (Altabe & Thompson, 1990; Birtchnell, Dolan, & Lacey, 1987; Bowden et al., 1989; Cash & Deagle, 1997; McKenzie et al., 1993 and Thompson, 1987). Attitudinal measures of body image have relied on participant responses to questionnaires or interviews. Studies using attitudinal measures have found that individuals with BN consistently report greater body dissatisfaction and concerns about appearance compared to non-eating disorder participants (Cooper, Cooper & Fairburn, 1989; Cooper & Fairburn, 1992; Cooper, Taylor, Cooper, & Fairburn, 1987; Garfinkel, 1992; Garfinkel, Goldbloom, Davis, Olmsted, Garner, & Halmi, 1992; Gross, Rosen, Leitenberg, & Willmuth, 1986; Hadigan & Walsh, 1991; Huon & Brown, 1989; Laessle, Kittle, Fichter, & Pirke, 1988; Laessle, Tuschl, Waadt, & Pirke, 1989; McKenzie et al., 1993, Nudelman et al., 1988, Powers et al., 1987, Sunday et al., 1992 and Whitehouse et al., 1986; Wilson & Smith, 1989). In a meta-analysis, Cash and Deagle (1997) found that attitudinal measures yielded a larger effect size than perceptual measures in the degree of difference between individuals with eating disorders and comparison samples; in addition, individuals with BN reported greater body image disparagement than participants with anorexia nervosa. Although relatively less attention has been paid to ideal body size, a number of studies using silhouette selection techniques have found that individuals with BN selected a smaller ideal than that selected by participants without BN (Counts & Adams, 1985; Horne, Van Vactor, & Emerson, 1991; McKenzie et al., 1993; Nelson & Gidycz, 1993; Williamson, Cubic, & Gleaves, 1993; Williamson, Davis, Goreczny, & Blouin, 1989; Williamson, Kelley, Davis, Ruggiero, & Blouin, 1985; Williamson, Prather, McKenzie, & Blouin, 1990). Consistent findings have been obtained in studies using visual distortion techniques, in which participants with BN have selected smaller ideal sizes relative to perceived sizes compared to samples of participants without BN (Freeman et al., 1983 and Probst et al., 1995; Szymanski & Seime, 1997). Although numerous studies have been conducted to investigate body image in BN, the role of body image in the treatment outcome of BN is not well understood. As highlighted in a review by Rosen (1996), many treatment outcome studies of BN have used attitudinal assessments of body dissatisfaction, including the eating disorders inventory (Garner & Olmsted, 1986) as secondary measures of treatment outcome. In the majority of these studies, attitudinal measures of body image have shown improvement after treatment using manual-based interventions, especially cognitive-behavioral therapy (Rosen, 1996). Although body image has been assessed less often in pharmacotherapy than in psychotherapy studies, some investigations have found that medication can lead to improved scores on body image measures (Walsh, Hadigan, Devlin, Gladis, & Roose, 1991; FBNC Study Group, 1992). In contrast to the measurement of body dissatisfaction, few studies have evaluated the effects of treatment on the perception of body size and ideal size. Birtchnell et al. (1985) evaluated size estimation in a sample of participants with BN using a moving caliper device and found a significant reduction in size overestimation after a 10-week outpatient treatment program. Sunday et al. (1992) found significant decreases in size overestimation in patients with BN following inpatient treatment. Relatively few studies have investigated whether aspects of body image are associated with outcome in BN. Maddocks and Kaplan (1991) found that the body dissatisfaction subscale of the Eating Disorders Invent (EDI) was one of five significant prognostic indicators. Freeman, Beach, Davis, and Solyom (1985b) observed that the size of the discrepancy between ideal and perceived body size using an image distortion technique was one of the best predictors of relapse in a sample of participants with BN. Fahy and Russell (1993) found that the shape concern subscale of the Eating Disorder Examination (Cooper & Fairburn, 1987) predicted the frequency of self-induced vomiting after 8 weeks of cognitive-behavioral treatment, but was not associated with longer-term outcome. Interestingly, Fairburn, Peveler, Jones, Hope, and Doll (1993) found that participants with more severe attitudinal disturbance about shape and weight did better in short-term psychotherapy, a finding consistent with Walsh et al. (1991) in a study of desipramine. In summary, although many studies have investigated the role of body image in BN, very little is known about the effects of treatment on various aspects of body image, or about aspects of body image as prognostic factors. The purpose of this study was to assess body perception, body dissatisfaction, and ideal body size in women with BN compared to those without BN, to evaluate changes in body image over the course of treatment and follow-up, and to determine whether body size overestimation, body dissatisfaction, and ideal body size were predictive of treatment outcome. Although previous studies have evaluated changes in measures of body image from pre- to post-treatment, follow-up assessments of size estimation and ideal size have not been conducted. Because of measurement problems and the considerable variability of size estimates observed in several studies (Collins et al., 1987 and Fernandez et al., 1994), the use of a comparison sample without eating disorder symptoms is essential in studies of body image measures over time; however, most studies have not utilized such samples. Very few studies of individuals with BN have evaluated the role of body size estimates as predictors of treatment outcome. Ideal size has been evaluated in predicting relapse (Freeman et al., 1985b), but has never been investigated as a predictor of short-term treatment outcome. The main hypotheses of this study were: (1) participants with BN would exhibit greater size overestimation and attitudinal disturbance, as well as a smaller average ideal body size, compared to participants without BN at baseline; (2) participants with BN would show reductions in size overestimation and attitudinal disturbance at post-treatment and follow-up, while measures for participants without BN would remain consistent over time; (3) participants with BN would demonstrate an increase in ideal size at post-treatment and follow-up, while measures for participants without BN would not change; and (4) body image measures at baseline for the BN participants would predict treatment status at outcome and follow-up, as well as outcome at follow-up based on scores at post-treatment.