خوددلسوزی رابطه بین شاخص توده بدنی و هر دو اختلال آسیب شناسی تغذیه و انعطاف پذیری تصویر ذهنی از بدن را تعدیل می خود

Abstract The current study examined whether self-compassion, the tendency to treat oneself kindly during distress and disappointments, would attenuate the positive relationship between body mass index (BMI) and eating disorder pathology, and the negative relationship between BMI and body image flexibility. One-hundred and fifty-three female undergraduate students completed measures of self-compassion, self-esteem, eating disorder pathology, and body image flexibility, which refers to one's acceptance of negative body image experiences. Controlling for self-esteem, hierarchical regressions revealed that self-compassion moderated the relationships between BMI and the criteria. Specifically, the positive relationship between BMI and eating disorder pathology and the negative relationship between BMI and body image flexibility were weaker the higher women's levels of self-compassion. Among young women, self-compassion may help to protect against the greater eating disturbances that coincide with a higher BMI, and may facilitate the positive body image experiences that tend to be lower the higher one's BMI.

Introduction Self-compassion has been defined as the tendency to respond to personal distress and inadequacies with self-kindness rather than self-judgment, an attitude of mindfulness rather than over-identification, and the perspective that suffering is common to humanity rather than isolating (Neff, 2003a). Just like self-esteem, self-compassion is a positive attitude toward self and the two variables correlate moderately with one another (Neff, 2003a). However, whereas self-esteem is a positive view of oneself that stems from appraisals of one's worth, attributes, and performance (Rosenberg, 1965), self-compassion derives from the human capacity for caregiving (Gilbert, 2005). It involves showing oneself support and warmth in the face of setbacks and disappointments, and unlike self-esteem, does not require that one's attributes or abilities be superior to those of others. A growing body of research now shows that self-compassion and self-esteem contribute independently to well-being and psychopathology (Neff, 2003a and Neff et al., 2007). Although both self-compassion and self-esteem appear to play a positive role in psychosocial functioning, self-compassion seems to offer important benefits that self-esteem does not. These benefits are especially salient when examining how individuals respond to personal failures, disappointments, or setbacks. Self-esteem has been associated with greater denial and minimizing of personal failures (Crocker & Park, 2004), and more defensive reactions after negative feedback (Leary, Tate, Adams, Allen, & Hancock, 2007). Self-compassion, by contrast, is associated with acknowledging and taking responsibility for one's role in failures, less overwhelmed emotional reactions, and the motivation to learn from one's mistakes and self-improve (Breines and Chen, 2012 and Leary et al., 2007). Self-compassion therefore appears to help weather distress and disappointment in a less personalized, more self-accepting, and growth-promoting way. There is some evidence to suggest that self-compassion may play a protective role in the area of body image and eating behavior. Wasylkiw, MacKinnon, and MacLellan (2012) found that controlling for self-esteem, female undergraduate students who were higher in self-compassion had fewer body image concerns, greater body appreciation, and less eating-related guilt. Ferreira, Pinto-Gouveia, and Duarte (2013) similarly found that higher self-compassion was associated with a lower drive for thinness in female eating disorder patients and community adults. Controlling for body mass index (BMI) and self-esteem, self-compassion has also been inversely associated with binge eating struggles and positively linked to intuitive eating, which refers to eating according to physiological hunger and satiety cues (Schoenefeld and Webb, 2013 and Webb and Forman, 2013). Finally, studies in eating disorder patients show that those who experienced greater improvements in self-compassion early in treatment fared better (Kelly, Carter, & Borairi, 2014), and those whose pre-treatment capacity for self-compassion was low fared poorly (Kelly, Carter, Zuroff, & Borairi, 2013). Although studies to date support a relationship between self-compassion, body image, and eating behavior, there is little research on whether self-compassion can attenuate disappointments or perceived inadequacies in the eating and body image realm. Because self-compassion appears to be an especially valuable protective factor when individuals encounter shortcomings and struggles, it is important to determine whether this holds true across life domains. One study by Adams and Leary (2007) investigated the effects of a self-compassion induction on restrained and guilty eaters. After eating an unhealthy preload, those who were prompted by the experimenter to think more self-compassionately about occasional indulgences had less subsequent disinhibited eating. This study supports the modulating influence that self-compassion may have when one experiences a “setback” or stressor in the realm of eating and body image, and supports the importance of further investigations on this topic. In today's Western culture, many women may experience an elevated BMI as a setback, stressor, or failure. First, the current “thin ideal” is significantly lower than the average woman's weight, and is physiologically impossible for most women to attain (Hawkins, Richards, Granley, & Stein, 2004). Second, it is difficult to escape media images of women who represent the “thin ideal,” and such exposure increases body dissatisfaction and eating disorder symptoms (Buote et al., 2011, Groesz et al., 2002 and Hawkins et al., 2004). Third, women with higher BMIs are more frequently the targets of weight stigma, including negative comments, social exclusion, and discrimination (Vartanian & Shaprow, 2008). Fourth, weight stigma and BMI each has been positively associated with disordered eating and body dissatisfaction (Myers and Rosen, 1999, Neumark-Sztainer et al., 2002, Rø et al., 2012 and Stice, 2002). BMI has also been negatively linked to body image flexibility (Wendell, Masuda, & Le, 2012), which refers to the ability to accept negative thoughts and feelings about one's body while remaining committed to desired and valued behaviors (Sandoz, Wilson, Merwin, & Kellum, 2013). Taken together, findings suggest that within the current sociocultural climate, an elevated BMI may be a source of stress for many women; it may confer vulnerability to eating pathology and undermine women's potential to experience positive body image. The Present Study The overarching aim of this study was to examine whether controlling for self-esteem, self-compassion would moderate the relationship between BMI and both eating disorder pathology and body image flexibility among undergraduate females. Our first objective was to determine whether self-compassion would moderate the relationship between BMI and eating disorder pathology – namely, global eating pathology, weight concerns, shape concerns, eating concerns, and dietary restraint. As in past studies, it was expected that higher self-compassion would be associated with less eating disorder pathology (Ferreira et al., 2013) and a higher BMI would be associated with more pathology (Rø et al., 2012 and Stice, 2002). Our central hypothesis was that the relationship between BMI and eating disorder pathology would be weaker the higher a woman's level of self-compassion. Because self-compassion has been positively associated with unconditional self-acceptance (Webb & Forman, 2013) and healthier forms of coping and self-regulation in the face of challenges (Kelly et al., 2010 and Terry and Leary, 2011), it was thought that women who were higher in self-compassion would be less prone to cope with their higher BMI by engaging in unhealthy eating and weight-control behaviors. Our second objective was to examine whether self-compassion would moderate the relationship between BMI and body image flexibility. We expected that as in previous studies, self-compassion would be positively related to this acceptance-based form of body image (Ferreira et al., 2011 and Schoenefeld and Webb, 2013), and that BMI would be negatively associated with body image flexibility. Our central hypothesis was that level of self-compassion would attenuate the strength of the negative relationship between body image flexibility and BMI. Highly self-compassionate individuals are better able to tolerate and persevere through challenges without becoming overwhelmed (Leary et al., 2007). We therefore expected that women with higher levels of self-compassion would cope more adaptively with the negative body image experiences that tend to coincide with a higher BMI, and that they would be less likely to sacrifice participation in desired and valued behaviors. These two positive coping behaviors of acceptance and commitment to action are inherent to body image flexibility.

Results Analytic Strategy Of the 153 participants in our final sample, BMI data were missing for three, EDE-Q data were missing for two, and self-esteem data were missing for one. Given that the analytic approaches used are unable to retain data from participants with missing observations, only participants with complete data on the variables examined in a given analysis were retained. Distributions for all variables were found to resemble a normal curve, except for BMI whose distribution was negatively skewed. When BMI was log-transformed, it approximated a normal distribution; therefore, the log-transformed variable was used in all analyses. First, means, SDs, and Pearson zero-order correlations were computed between all study variables (see Table 1). Next, all predictor variables were centered so as to facilitate interpretation of the main results. Six hierarchical regressions were used to test our central hypotheses. Criterion variables were five indicators of eating disorder pathology – global eating pathology, weight concerns, shape concerns, eating concern, and dietary restraint – and body image flexibility. Age was initially controlled in all analyses but was never a significant predictor so was removed from the final analyses presented below. Table 1. Means, standard deviations (SD) and zero-order correlations between study variables. 1 2 3 4 5 6 7 8 9 Mean SD 1. BMI – 0.04 −0.17* −0.26** 0.28*** 0.37*** 0.29** 0.24** 0.11 23.10 5.04 2. Self-compassion – 0.62*** 0.41*** −0.41*** −0.40*** −0.44*** −0.32*** −0.29*** 2.88 0.65 3. Self-esteem – 0.39*** −0.33*** −0.31*** −0.35*** −0.30*** −0.20* 33.60 6.41 4. Body image flexibility – −0.78*** −0.77*** −0.76*** −0.69*** −0.58*** 5.07 1.17 5. Global eating pathology – 0.94*** 0.93*** 0.84*** 0.85*** 1.87 1.24 6. Weight concerns – 0.90*** 0.73*** 0.69*** 2.20 1.50 7. Shape concerns – 0.70*** 0.69*** 2.59 1.53 8. Eating concerns – 0.62*** 0.97 1.09 9. Dietary restraint – 1.68 1.44 * p < 0.05. ** p < 0.01. *** p < 0.001. Table options In each hierarchical regression model, self-esteem was entered into Step 1 as a control variable. In Step 2, BMI and self-compassion were added to the model as main effects. Finally, self-compassion × BMI was added in Step 3. Effect size correlations for the complete models (Step 3) were computed and presented in Table 2 using Rosnow and Rosenthal, 1996 formula of r = [F/(F + df)]1/2. According to Cohen (1988), r = .10 is a small effect and r = .30 is a medium effect. In the case of a significant self-compassion × BMI effect, simple slopes were estimated ( Aiken & West, 1991) and plotted to depict the relationship between BMI and the criterion variable at low (−1 SD), average (mean), and high (+1 SD) levels of self-compassion. Table 2. Hierarchical multiple regressions predicting global eating disorder pathology, weight concerns, shape concerns, eating concerns, dietary restraint, and body image flexibility. R2 ΔR2 β (SE) t p Squared semi-partial correlation Global eating disorder pathology,F(4, 142) = 13.94,p < .001, effect sizer = .30 Step 1 0.103 Self-esteem −0.403 (0.098) −4.12 <0.001 0.105 Step 2 0.262 0.159 Self-esteem 0.005 (0.120) 0.05 0.964 000 BMI 0.373 (0.094) 3.98 <0.001 0.082 Self-compassion −0.537 (0.118) −4.55 <0.001 0.107 Step 3 0.282 0.020 Self-esteem 0.054 (0.121) 0.440 0.659 0.001 Self-compassion × BMI −0.199 (0.098) −2.03 0.045 0.021 Weight concern,F(4, 142) = 17.40,p < .001, effect sizer = .33 Step 1 0.096 Self-esteem −0.459(0.119) −3.87 <0.001 0.092 Step 2 0.308 0.212 Self-esteem 0.070 (0.141) 0.050 0.621 0.001 BMI 0.592 (0.110) 5.38 <0.001 0.140 Self-compassion −0.676 (0.138) −4.89 <0.001 0.116 Step 3 0.329 0.021 Self-esteem 0.129 (0.142) 0.91 0.366 0.004 Self-compassion × BMI −0.242 (0.115) −2.11 0.037 0.021 Shape concern,F(4, 142) = 14.95,p < .001, effect sizer = .31 Step 1 0.130 Self-esteem −0.549 (0.118) −4.65 <0.001 0.130 Step 2 0.285 0.155 Self-esteem −0.047 (0.145) −0.33 0.822 0.001 BMI 0.451 (0.113) 3.99 <0.001 0.079 Self-compassion −0.662 (0.142) −4.65 <0.001 0.108 Step 3 0.296 0.011 Self-esteem −0.004 (0.147) −0.03 0.979 000 Self-compassion × BMI −0.179 (0.119) −1.51 0.134 0.011 Eating concern,F(4, 142) = 8.58,p < .001, effect sizer = .24 Step 1 0.087 Self-esteem −0.317 (0.086) −3.71 <0.001 0.087 Step 2 0.169 0.082 Self-esteem −0.068 (0.111) −0.590 0.555 0.002 BMI 0.251 (0.086) 3.01 0.003 0.052 Self-compassion −0.321 (0.109) −3.03 0.004 0.053 Step 3 0.195 0.026 Self-esteem −0.022 (0.111) −0.20 0.845 000 Self-compassion × BMI −0.191 (0.090) −2.12 0.036 0.026 Dietary restraint,F(4, 142) = 5.00,p = .001, effect sizer = .18 Step 1 0.038 Self-esteem −0.280 (0.118) −2.38 0.018 0.038 Step 2 0.110 0.072 Self-esteem 0.068 (0.153) 0.44 0.659 0.001 BMI 0.200 (0.120) 1.68 0.096 0.017 Self-compassion −0.490 (0.150) −3.12 0.001 0.066 Step 3 0.123 0.013 Self-esteem 0.112 (0.155) 0.72 0.473 0.003 Self-compassion × BMI −0.183 (0.126) −1.45 0.149 0.013 Body image flexibility,F(4, 142) = 13.94,p < .001, effect sizer = .30 Step 1 0.152 Self-esteem 0.457 (0.089) 5.12 <0.001 0.152 Step 2 0.261 0.109 Self-esteem 0.152 (0.112) 1.36 0.178 0.009 BMI −0.295 (0.087) −3.41 0.001 0.060 Self-compassion 0.405 (0.110) 3.68 <0.001 0.069 Step 3 0.281 0.020 Self-esteem 0.101 (0.114) 0.990 0.323 0.005 Self-compassion × BMI 0.191 (0.092) 2.06 0.041 0.021 Note. F-values and effect size rs were derived from the final model in Step 3 of each hierarchical regression. Table options Preliminary Results Consistent with hypotheses, zero-order correlations revealed that BMI was negatively related to body image flexibility and positively related to all forms of eating disorder pathology. In addition, self-compassion was positively related to body image flexibility and negatively related to all forms of eating disorder pathology (see Table 1). Primary Results Global eating disorder pathology. Self-esteem made a significant negative contribution to global eating disorder pathology in Step 1, but this contribution disappeared in Step 2 once BMI and self-compassion were entered; the latter two variables were positively and negatively related to global eating disorder pathology respectively (see Table 2). In Step 3, self-compassion × BMI emerged as a significant predictor as hypothesized. Simple slopes analysis revealed that self-compassion moderated BMI's positive contribution to global eating disorder pathology, such that this relationship was significant among women with low and average self-compassion, but not significant among women with high self-compassion (see Table 3 and Fig. 1). Table 3. Simple slopes analyses for significant self-compassion × BMI Effects. β (SE) t p Global eating disorder pathology 1 SD below mean on self-compassion 0.613 (0.150) 4.08 <0.001 Mean self-compassion 0.414 (0.095) 4.35 <0.001 1 SD above mean on self-compassion 0.215 (0.122) 1.77 0.079 Weight concerns 1 SD below mean on self-compassion 0.883 (0.176) 5.03 <0.001 Mean self-compassion 0.641 (0.111) 5.77 <0.001 1 SD above mean on self-compassion 0.399 (0.142) 2.81 0.006 Eating concerns 1 SD below mean on self-compassion 0.481 (0.138) 3.49 <0.001 Mean self-compassion 0.289 (0.087) 3.32 0.001 1 SD above mean on self-compassion 0.098 (0.112) 0.88 0.381 Body image flexibility 1 SD below mean on self-compassion −0.529 (0.142) −3.74 <0.001 Mean self-compassion −0.338 (0.090) −3.77 <0.001 1 SD above mean on self-compassion −0.147 (0.115) −1.28 0.202 Table options Regression lines showing the relationship between BMI and global eating disorder ... Fig. 1. Regression lines showing the relationship between BMI and global eating disorder pathology as a function of self-compassion level. Estimates for low, average, and high levels of self-compassion and BMI were calculated using standardized scores, where low was 1 SD below the mean, average was the mean, and high was 1 SD above the mean. The graph illustrates that the positive relationship between BMI and global eating disorder pathology was weaker the higher women's level of self-compassion. Figure options Weight concerns. Similar results were obtained for weight concerns at each step of the hierarchical regression (see Table 2). Self-compassion × BMI was a significant predictor, with simple slopes analysis revealing that the positive relationship between BMI and weight concerns was significant across levels of self-compassion, but weakest among women with high self-compassion (see Table 3 and Fig. 2). Regression lines showing the relationship between BMI and weight concerns as a ... Fig. 2. Regression lines showing the relationship between BMI and weight concerns as a function of self-compassion level. Estimates for low, average, and high levels of self-compassion and BMI were calculated using standardized scores, where low was 1 SD below the mean, average was the mean, and high was 1 SD above the mean. The graph illustrates that the positive relationship between BMI and weight concerns was weaker the higher women's level of self-compassion. Figure options Shape concerns. Results from the hierarchical regression were similar to those obtained in Steps 1 and 2 for global eating disorder pathology. However, in Step 3, self-compassion × BMI was not significant (see Table 2), contrary to our hypothesis. Eating concerns. The pattern of results for eating concerns was identical to that obtained for global eating disorder pathology (see Table 2), with self-compassion moderating the strength of the positive relationship between BMI and eating concerns. Again, the relationship between these two variables was significant among women with low and average self-compassion, but not significant among women with high self-compassion (see Table 3 and Fig. 3). Regression lines showing the relationship between BMI and eating concerns as a ... Fig. 3. Regression lines showing the relationship between BMI and eating concerns as a function of self-compassion level. Estimates for low, average, and high levels of self-compassion and BMI were calculated using standardized scores, where low was 1 SD below the mean, average was the mean, and high was 1 SD above the mean. The graph illustrates that the positive relationship between BMI and eating concerns was weaker the higher women's level of self-compassion. Figure options Dietary restraint. Similar to what was found when predicting global eating disorder pathology, self-esteem contributed negatively to dietary restraint in Step 1 of the hierarchical regression. In Step 2, self-esteem was no longer significant, and self-compassion emerged as a significant negative predictor. BMI, however, was not a significant predictor. Self-compassion × BMI, entered in Step 3, was also not significant, contrary to hypotheses. Body image flexibility. In Step 1, self-esteem negatively predicted body image flexibility. In Step 2, self-esteem was no longer significant, but BMI was a negative predictor and self-compassion was a positive predictor. In Step 3, self-esteem remained non-significant, and self-compassion × BMI was a significant predictor as hypothesized (see Table 2). Simple slopes analysis revealed that BMI's negative contribution to body image flexibility was attenuated the higher women's level of self-compassion (see Table 3). As depicted graphically in Fig. 4, among women with lower or average self-compassion, there was a significant negative relationship between BMI and body image flexibility; however, this relationship was not significant among women with higher self-compassion. Regression lines showing the relationship between BMI and body image flexibility ... Fig. 4. Regression lines showing the relationship between BMI and body image flexibility as a function of self-compassion level. Estimates for low, average, and high levels of self-compassion and BMI were calculated using standardized scores, where low was 1 SD below the mean, average was the mean, and high was 1 SD above the mean. The graph illustrates that the negative relationship between BMI and body image flexibility was weaker the higher women's level of self-compassion.