Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2022 Mar 1.
Published in final edited form as: J Behav Cogn Ther. 2021 Mar 3;31(1):57–66. doi: 10.1016/j.jbct.2020.11.002

Pilot data from the Self-Blame and Perspective-Taking Intervention for eating disorders

Bethany J Hunt a, Whitney Smith Hagan a, Sarah Pelfrey a, Susan Mericle a, Jessica A Harper a, Jayme M Palka a, Carrie McAdams a
PMCID: PMC8195261  NIHMSID: NIHMS1710137  PMID: 34124699

Abstract

Eating disorders (EDs) are characterized by altered eating behaviors and valuation of self-image, as well as difficulty establishing supportive social relationships. This pilot study evaluated feasibility, acceptability, and clinical responses to a novel and brief group-therapy intervention for EDs, the Self-Blame and Perspective-Taking Intervention (SBPI). The SBPI consisted of four sessions of experiential art therapy activities in conjunction with psychoeducation targeting interpersonal attributions and mentalization. Twenty-four outpatient, treatment-seeking women with EDs participated in the SBPI, with 87.5% completing the intervention and 94% rating their participation positively. ED symptoms, depression, anxiety, self-attribution bias, and self-esteem were assessed before (T1) and after participation (N = 20 at T2; N = 18 at T3). Separate repeated measures MANOVAs were performed to assess these clinical and self-concept variables. Relative to baseline, participants demonstrated significant improvements in two all self-concept measures: self-attribution bias, trait self-esteem and state self-esteem at T2. ED, depression, and anxiety symptoms were significantly decreased at both T2 (1-4 weeks post) and T3 (3-5 months post). The SBPI altered self-concept targets acutely and led to sustained clinical improvements. Future work is needed to evaluate how self-concept and social constructs are related to clinical symptom expression in EDs.

Keywords: anorexia nervosa, bulimia nervosa, binge eating disorder, self-evaluation, social cognition, interpersonal psychotherapy

Introduction

Eating disorders (EDs) are serious psychiatric illnesses characterized by altered eating behaviors as well as impairments related to self-esteem and self-perception. Eating disorders impact approximately 10% of women during adolescence and young adulthood (Smink, van Hoeken, & Hoek, 2012) and have a high mortality rate (Arcelus, Mitchell, Wales, & Nielsen, 2011; Keel & Brown, 2010). Recovery from EDs in adults, including both anorexia nervosa (AN) and bulimia nervosa (BN), is unpredictable and prolonged with current treatment methods (Eddy et al., 2017). ED treatments focus first on altering the disease-defining symptoms which include restrictive, purging, and binge-eating behaviors as well as targeting cognitions related to body image and self-worth. Acute recovery is defined by changing ED-related cognitions and behaviors. However, improvements in social relationships and self-esteem are more consistently associated with sustained recovery than changes in cognitions about either physical appearance or food (Harper et al., 2017; Smink et al., 2012; Tozzi, Sullivan, Fear, McKenzie, & Bulik, 2003). Indeed, individuals in recovery from an ED report that recovery is largely influenced by their beliefs about themselves and their connections to others (Linville, Brown, Sturm, & McDougal, 2012).

Difficulties in social and emotional function are observed in adults with EDs. Individuals with EDs, including AN, BN, and BED, have a reduced ability to recognize and understand emotions (Aloi, Rania, Caroleo, De Fazio, & Segura-Garcia, 2017; Hason Rozenstein et al., 2018; Westwood et al., 2017). Anorexia nervosa was associated with impaired performance in theory of mind (ToM) tasks (Kuipers & Bekker, 2012; Russell, Schmidt, Doherty, Young, & Tchanturia, 2009). Women with AN provided shorter spontaneous social narratives and had difficulty recognizing social cues (Oldershaw, DeJong, Hambrook, & Schmidt, 2018). Individuals with BN have difficulty recognizing emotions in facial expressions (Sacchetti et al., 2019) and have reduced ToM (Laghi et al., 2017). Similarly, individuals with BED have reduced performance on ToM tasks with performance influenced by disordered eating pathology (Legnani et al., 2017).

Functional magnetic resonance imaging allows examination of the neural circuitry involved in social perception. During mentalization about social stimuli, both adults and adolescents with AN showed reduced utilization of regions in the default mode network, including the temporoparietal junction, precuneus, and medial prefrontal cortex (McAdams & Krawczyk, 2011; Schulte-Ruther, Mainz, Fink, Herpertz-Dahlmann, & Konrad, 2012). In both AN and BN, the dorsal anterior cingulate was differently engaged during self-relevant social perspective-taking (McAdams & Krawczyk, 2013, 2014), with strongest differences for weight-recovered women with AN (McAdams et al., 2016). Within these participants, modulation of the insula during social processing was also correlated with the tendency to negatively self-attribute situations (McAdams, Harper, & Van Enkevort, 2018). The neural differences observed in comparison of weight-recovered and ill AN cohorts provided the initial motivation for this intervention, suggesting that mentalization might offer a cognitive compensation for trait-related impairments in social perceptions.

We considered that this bias towards a negative self-concept may be exacerbated by negative social experiences, indirectly promoting ED symptoms, and might be ameliorated with third-person perspective-taking. We developed a novel group outpatient intervention, the Self-Blame and Perspective-Taking Intervention (SBPI) for Eating Disorders (Figure 1). The SBPI included art-therapy activities in conjunction with psychoeducation on social perspective-taking and self-biases. A battery of clinical and self-concept measures were assessed before and after the SBPI, and an evaluation of the components completed after the SBPI. Participant satisfaction, attendance, and homework completion was also evaluated as a measure of feasibility and acceptability. As a primary hypothesis, we predicted that SBPI participants would show reduced global ED symptomology and more positive self-attributions relative to baseline (T1) during the 1-4 week post-intervention follow-up (T2). We also predicted, as an exploratory hypothesis, that these changes would be sustained at the second follow-up (T3), 3-5 months after participation.

Figure 1.

Figure 1.

Open in a new tab

A. On upper left, a model of how differences in social perceptions may lead to more negative social interactions, and how those in turn promote ED cognitions (low self-esteem, preoccupation with body shape/size) and behaviors (restriction, binge-eating, and purging) at the top. On upper right, the SBPI was designed to both (1) provide information about social processing to counter existing biases, and (2) offer a structured group interaction that is positive, to counter prior negative experiences with others. B. Components of the SBPI. First photo panel TrustDraw, upper is completed set with original on left and copy on right; the content is accurate, but spacing changed in copy; lower, more in-process image showing copying. Second panel WeDraw, upper shows device copying the image on card, and lower is finished product after tracing and coloring. Third panel CubeStack, top shows strips of paper and goal; bottom completed stacked boxes matching each perspective. Fourth Puzzle Story, top left shows the accuracy in copying by individuals, bottom is process of putting pieces together, top right the final product and its book title.

Method

Participants and study protocol overview

All procedures received approval from the UT Southwestern Institutional Review Board. Recruitment flyers were distributed to clinicians and eating disorder treatment programs in Texas, with target enrollment at 4-8 participants per cohort. Four separate cohorts participated in the SBPI between January 2018 and March 2019, with three to eight individuals enrolled per group. Eligible individuals included female outpatients between the ages of 18-46 who met DSM-5 criteria for AN, BN, or BED, and provided informed consent. Participants could not be currently in an inpatient, residential or partial hospital program; pregnancy, acute suicidality or psychosis, and medical instability were additional exclusions.

At screening, a trained clinician administered the Mini-International Neuropsychiatric Interview (MINI) and the Eating Disorder Assessment-5 (EDA-5) to confirm a lifetime ED diagnosis of AN, BN, or BED (Sheehan et al., 1998; Sysko et al., 2015), intelligence was assessed with the Wechsler Abbreviated Scale of Intelligence (Wechsler, 2011), and demographic information obtained. Pre (T1, 1-4 weeks before SBPI) and post-intervention assessments (T2, 1-4 weeks after SBPI; T3, 3-5 months after SBPI) included both clinician-rated and self-report measures. Clinician-rated measures were completed for both depression and anxiety and self-report measures about eating, self-esteem, and interpersonal attributions were submitted online within one week of the intervention start using REDCap electronic data capture tools hosted at the University of Texas Southwestern Medical School (Harris et al., 2009).

Self-Blame and Perspective-Taking Intervention

Overview.

The SBPI was administered during four, two-hour weekly sessions and included a combination of experiential scenarios and psychoeducation. Study facilitators (including an art therapist, psychiatry resident, psychology intern, and research assistant) used a manual to guide sessions. The manual was based on psychoeducational and art therapy sessions utilized within an ED program that included the inpatient to intensive outpatient levels of care. Discussion of specific ED behavior was not permitted; participants agreed on enrollment to avoid any commentary about eating-disorder behaviors and cognitions, including remarks about their own or others’ body size, shape, weight or food-related problems.

The experiential component utilized collaborative art tasks (developed and implemented by Susan Mericle, ATRBC) and videos (Ambwani et al., 2016) to establish shared experiences for communal group processing. In addition to challenging self-blame, art tasks required perspective-taking and cooperation. The videos, all depicting examples of both positive and negative job feedback, enabled participants to discuss their emotional responses during real-life scenarios without necessitating personal disclosures. Psychoeducation included lecture, group discussion, activities, and homework assignments. For psychoeducation, participants received a handout about social behaviors and interpersonal attributions. Each lecture included active-learning exercises, and homework that involved real-world consideration of the key social concept introduced (Figure 1).

Week One: Attributions.

The art experiential was called the Trust Draw, and psychoeducation introduced the concept of attribution biases. During the Trust Draw art task, the group divided into small groups of two to three; in each group one person drew and the others directed. The directors received a card with a complex design and gave verbal-only instructions to the drawer to recreate the design. The drawer was to remain silent but could ask for instructions to be repeated. Roles switched and new cards were provided. Post-task processing focused on communication and the limitations in language at sharing one’s own perspective to others. In the psychoeducation, attribution bias was illustrated as a cognitive short cut to determine motivations underlying behaviors. Two to four critical work feedback videos were shown, using the participants’ emotional reactions to reflect on individual attribution biases. Homework focused on recognizing attribution bias in their own lives.

Week Two: Positive and Negative Biases.

The WeDraw, an apparatus with multiple handles and a single marker, such that handles detach if the device is pulled in different directions, required the entire group to copy a complex image from a small card onto a large sheet of paper. Each person stood in a different position while holding the device, and so had a unique perspective of the image. To succeed, the group must cooperate, as well as accept the imperfections in the copy related to the complexity of maneuvering the markers. Following the drawing, the group was asked to find a song that symbolized them as a group, and wrote a lyric on the drawing. Collectively, this task targets social perspective-taking, self-blame, and perfectionism, and promotes social bonding and cooperation. The psychoeducation component outlined the biopsychosocial model of mental illness and social rewards, utilizing role-plays to model differences in one’s responses to positive and negative social interactions. Time-permitting, the videos were also shown, with a focus on identifying positive feedback. Homework involved identifying positive and negative situations and considering the social signals and biases responsible for those experiences.

Week Three: Perspectives.

The CubeStack task began by providing the group two illustrations of the same three cubes stacked in different ways, a stack of white paper strips, markers, and scotch tape. The group was told to recreate the cubes and stack them to replicate the images. Constructing the cubes required awareness of how the view of the cube changed based on its rotation requires spatial reasoning, perspective-taking, and set-shifting. Psychoeducation introduced theory of mind, false beliefs, and the use of perspective-taking as a cognitive strategy to manage social-emotional stressors. Feedback videos from week one were repeated in conjunction with different background stories to demonstrate how perspective shifts emotional reaction. Videos were discussed with a focus on hearing both positive and negative aspects of feedback and applying perspective-taking in real world situations.

Week Four: Integration and Recovery Tools.

The Puzzle Story task began with each person receiving an ambiguous painted image that they were asked to recreate accurately on a blank paper with pastels. Unbeknownst to participants, each image was cut from the same original painting. After they finished copying their images, participants were instructed to reassess their images and collectively assemble their pieces. If done correctly, the recreated images would fit into the original. Subsequent discussion focused on differences in perspective, particularly in relation to looking at big picture versus details. To conclude, the group was told to title the puzzle. Psychoeducation synthesized the information from the previous three sessions and reflected on the intervention goals including identifying interpersonal biases, recognizing other perspectives using role-play activities, and working positively in a group. The group also discussed ways to incorporate these ideas into daily life and their individual psychotherapeutic work with outside therapists.

Measures

Participant feedback.

At the end of the fourth week, participants completed a questionnaire with a 1-10 rating scale evaluating how helpful the components within each session were (10, most helpful): the art experiential, the didactic psychoeducation, other experiential activities within the didactic, and homework. A single query on degree of satisfaction with participation was also assessed at T3 using a 5-point Likert-type scale (1 = extremely dissatisfied; 5 = extremely satisfied).

Repeated measures

Self-Concept.

Participants completed three measures of self-concept at three timepoints, T1-T3. The Internal, Personal, and Situational Attributions Questionnaire (IPSAQ, Kinderman & Bentall, 1996) is a 32-item self-report, in which participants self-generate a cause (i.e. self, another person, or the situation) for 16 positive and 16 negative interpersonal events. Three types of interpersonal attribution biases are computed: negative personalizing bias (NPB), positive personalizing bias (PPB), and externalizing bias. Externalizing bias, hereon referred to as self-attribution bias, indicates the valence of internal attributions such that a more positive score is a self-serving bias. Personalizing biases are ratios capturing the relative proportion of events attributed to another person versus the situation, with scores above 0.5 reflecting a greater tendency to personal rather than situational attributions. The State Self-Esteem Scale (SSES; Heatherton & Polivy, 1991) is a 20-item self-report scale including social, performance, and appearance-based evaluations. Higher scores reflect positive attitudes about current self-worth and self-competency. Each response is provided using a 5-point Likert scale ranging from “not at all” to “extremely.” The Rosenberg Self-Esteem Scale (RSES; Rosenberg, 1965) is a 10-item self-report 4-point Likert scale (“strongly agree” to “strongly disagree”) that evaluates trait self-esteem by measuring positive and negative feelings towards the self. Responses are scored from 0-30, with a higher score indicating positive global self-regard.

Clinical symptoms.

Clinical symptoms were also assessed at T1-T3. The Eating Disorder Examination-Questionnaire 6 (EDE-Q, Luce & Crowther, 1999) is a 28-item survey administered via self-report that evaluates ED cognitions and behaviors in the prior month, yielding a global score and four subscales: restriction, eating concern, shape concern, and weight concern. The Quick Inventory of Depressive Symptomology (QIDS, Rush et al., 2003) is a 16-item clinical interview that rates depressive symptoms over the past seven days, with high internal consistency. The Structured Interview Guide for the Hamilton Anxiety Scale (SIGH-A, Shear et al., 2001) is a 14-item clinical-scored interview used to evaluate anxiety, including both cognitive and physiological experiences.

Statistical analysis

All statistical data analyses were performed using the IBM Statistical Package for Social Sciences, Version 23 (SPSS, v.23). Information about participant characteristics, attendance, homework completion, and ratings of the intervention are presented with descriptive statistics. Primary hypothesis testing examined self-concept and clinical outcomes across T1 and T2 using two repeated measures multivariate analysis of variance (MANOVAs). Exploratory MANOVAs considered the same clinical and self-concept measures across all three time points (T1, T2, and T3). For all tests, effect sizes were determined using partial eta squared (ηp2) and significant results are reported where alpha is less than 0.05. Following each statistically significant MANOVA, a Bonferroni post hoc test was conducted to determine between which two time points the significant effect occurred. Partial eta squared (ηp2) effect sizes were calculated and compared against benchmarks provided by Cohen (1969) with 0.0099 (small), 0.0588 (medium), and 0.1379 (large). Results were deemed statistically significant when p < .05.

Results

Feasibility.

As shown in Figure 2, 29 subjects consented, 26 completed screening and baseline assessments, and 24 participants attended at least one session (for attendees of SBPI, sessions M = 3.44, SD = 0.87). Among attendees, two individuals dropped out after their first session while the other 22 participants completed three or more sessions; 20 participants provided follow-up data at T2 and 18 participants provided follow-up data at T3.

Figure 2.

Figure 2.

Open in a new tab

Flow of participants through the study.

Participants.

The twenty-four attendees of the SBPI included 11 with AN, 11 with BN, and 2 with BED. Comorbidities in the participants included anxiety disorders (n = 12), mood disorders (n = 7), obsessive-compulsive disorder (n = 4), and post-traumatic stress disorder (n = 3), with eight participants having no current comorbid psychiatric illnesses. Participants age ranged from 19-41 years old (M = 28.5, SD = 5.76 years) and the average IQ was 116 (SD = 8.2). Participants identified their race as White (n = 22), Black (n = 1), or Other (n = 1); and the majority of participants reported their ethnicity as non-Hispanic (n = 20).

Acceptability.

Participant ratings for each component are shown in Table 1. Averaged across weeks, the art tasks (N = 71, M = 8.4, SD = 2.4), psychoeducation (N = 69, M = 8.6, SD = 1.7), didactic role-play activities (N = 64, M = 8.1, SD = 2.2), and homework assignments (N = 51, M = 8.5, SD = 1.6) were all highly rated. The feedback videos, utilized as prompts during psychoeducation garnered the lowest ratings (N = 45, M = 5.8, SD = 2.8). Homework completion was 89.7%. The feedback questionnaire was completed by 90.5% of participants. Of those who completed the T3 follow up (N = 18), 83.3% of participants (N = 15) reported that they were “extremely satisfied” with their participation, and 11.1% of participants (N = 2) answered that they were “somewhat satisfied,” and 5.6% (N = 1) reported being “somewhat dissatisfied.”

Table 1.

Participant ratings (1 low to 10 high) for art tasks, education, and other components in the intervention.

Week N M (SD) Min (Median) Max
Art Tasks All 71 8.4 (2.4) 1 (10) 10
 TrustDraw 1 19 8.9 (2.0) 2 (10) 10
 WeDraw 2 18 8.6 (1.6) 4 (8.5) 10
 CubeStack 3 16 7.3 (3.5) 1 (8.5) 10
 Puzzle Story 4 18 8.6 (2.6) 2 (10) 10
Psychoeducation All 69 8.6 (1.7) 2 (9) 10
 Attributions 1 19 8.4 (1.9) 2 (9) 10
 Pos/Neg Biases 2 16 8.7 (1.8) 5 (9.5) 10
 Perspective 3 16 8.6 (1.5) 5 (9) 10
 Integration 4 18 8.8 (1.6) 5 (10) 10
Other Components
 Feedback Videos Multiple 45 5.8 (2.8) 1 (7) 10
 Role-Plays Multiple 64 8.1 (2.2) 2 (9) 10
 Homeworks Multiple 51 8.5 (1.6) 5 (9) 10
Open in a new tab

Note. M = mean and SD = standard deviation

Self-concept outcomes.

In the repeated measures MANOVA comparing T1 and T2, there were significant changes in self-attribution bias (p = .02; np2 = 0.27), state self-esteem (p = .03; np2 = 0.23), and trait self-esteem (p = .01; np2 = 0.28) (Table 2). Scores for self-attribution bias, trait self-esteem, and state self-esteem increased, while trait self-esteem decreased; all of these changes are large effect sizes. Changes in self-attribution bias and state self-esteem were not significant in the exploratory MANOVA examining T1, T2, and T3; however, the change in trait self-esteem was sustained (p = .003; np2 = 0.41) (Table 3). A Bonferroni post hoc test revealed that scores for trait self-esteem significantly increased between T1 and T3 (pb = .02), but T2 scores did not differ from either T1 (pb = .07) or T3 (pb = .86).

Table 2.

Univariate outcomes across time from MANOVA

Self-Concept MANOVA
Measure T1 N = 20 M (SD) T2 N = 20 M (SD) MANOVA Results p = 0.04 Effect Sizes ηp2
IPSAQ-EB 0.55 (5.87) 3.30 (4.13) F(1,19) = 6.91, p = 0.02* 0.27
IPSAQ-NPB 0.67 (0.24) 0.60 (0.32) F(1,19) = 1.89, p = 0.19 0.09
IPSAQ-PPB 0.60 (0.28) 0.48 (0.32) F(1,19) = 2.08, p = 0.17 0.10
SSES 54.75 (16.18) 61.25 (14.29) F(1,19) = 5.58, p = 0.03* 0.23
RSES 13.00 (5.97) 15.95 (5.36) F(1,19) = 7.37, p = 0.01* 0.28
Clinical MANOVA
Measure T1 N = 20 M (SD) T2 N = 20 M (SD) MANOVA Results p = 0.01 Effect Sizes ηp2
EDE-Q 3.11 (1.56) 2.39 (1.35) F(1,19) = 13.46, p = 0.002* 0.42
QIDS 9.0 (4.92) 5.8 (3.04) F(1,19) = 7.70, p = 0.01* 0.29
SIGHA 11.7 (7.52) 8.0 (3.92) F(1,19) = 5.26, p = 0.03* 0.22
Open in a new tab

Note. T1 = pre-intervention; T2 = 1-4 week follow-up. Overall significance (p < 0.05) is notated by an asterisk. M = mean and SD = standard deviation. IPSAQ-EB, Internal, Personal, and Situational Attributions Questionnaire externalizing bias; IPSAQ-NPB, IPSAQ negative personalizing bias; IPSAQ-PPB, IPSAQ positive personalizing bias; SSES, State Self-Esteem Scale; RSES; Rosenberg Self-Esteem Scale; EDE-Q, Eating Disorder Examination Questionnaire 6; QIDS, Quick Inventory of Depression Symptomology; SIGH-A, Structured Interview Guide for the Hamilton Anxiety Scale.

Table 3.

Exploratory MANOVA results across time

Self-Concept MANOVA
Measure T1 N = 18 M (SD) T2 N = 18 M (SD) T3 N = 18 M (SD) MANOVA Results p = 0.11 Effect Sizes ηp2
IPSAQ-EB 0.89 (6.11) 3.83 (3.97) 3.56 (3.01) F(1,17) = 3.11, p = 0.10 0.16
IPSAQ-NPB 0.67 (0.24) 0.63 (0.32) 0.65 (0.24) F(1,17) = 0.07, p = 0.79 0.004
IPSAQ-PPB 0.63 (0.25) 0.51 (0.31) 0.59 (0.29) F(1,17) = 0.24, p = 0.63 0.01
SSES 55.22 (16.72) 60.78 (13.45) 61.89 (14.34) F(1,17) = 4.17, p = 0.06 0.20
RSES 13.28 (6.19) a 15.83 (5.26) 17.11 (5.58) b F(1,17) = 11.756, p = 0.003* 0.41
Clinical MANOVA
Measure T1 N = 18 M (SD) T2 N = 18 M (SD) T3 N = 18 M (SD) MANOVA Results p = 0.21 Effect Sizes ηp2
EDE-Q 3.14 (1.62) a 2.45 (1.37) b 2.38 (1.38) b F(1,17) = 7.07, p = 0.02* 0.29
QIDS 9.5 (4.83) a 6.0 (3.07) b 6.3 (3.10) b F(1,17) = 10.28, p = 0.01* 0.38
SIGHA 12.4 (7.45) 8.3 (3.87) 8.3 (5.86) F(1,17) = 5.83, p = 0.03* 0.26
Open in a new tab

Note. T1 = pre-intervention; T2 = 1-4 week follow-up; T3 = 3-5 month follow-up. Overall significance (p < 0.05) is notated by an asterisk and significantly different time-points specified by unmatched alphabetic superscripts. M = mean and SD = standard deviation. IPSAQ-EB, Internal, Personal, and Situational Attributions Questionnaire externalizing bias; IPSAQ-NPB, IPSAQ negative personalizing bias; IPSAQ-PPB, IPSAQ positive personalizing bias; SSES, State Self-Esteem Scale; RSES; Rosenberg Self-Esteem Scale; EDE-Q, Eating Disorder Examination Questionnaire 6; QIDS, Quick Inventory of Depression Symptomology; SIGH-A, Structured Interview Guide for the Hamilton Anxiety Scale.

Clinical outcomes.

All three clinical variables, including ED symptoms (p = .002; np2 = 0.42), depression (p = .01; np2 = 0.29), and anxiety (p = .03; np2 = 0.22), improved between T1 and T2 (Table 2). Results from the exploratory MANOVA comparing all three time-points (Table 3) were similar for ED symptoms (p = .02; np2 = 0.29), depression (p = .01; np2 = 0.38), and anxiety (p = .01; np2 = 0.28). A Bonferroni post hoc test revealed that global ED symptoms improved, where T1 scores were significantly higher than T2 scores (pb = .02) and T3 scores (pb = .05), while scores from T2 and T3 did not significantly differ (pb = 1.00). Similarly, the comparisons for depression showed that T1 scores were significantly higher than T2 scores (pb = .04) and T3 scores (pb = .02), with no significant difference between T2 and T3 (pb = 1.00). Although both MANOVAs for anxiety were significant overall, a Bonferroni post hoc test showed no significant differences between individual time-points.

Discussion

The Self-Blame and Perspective-Taking intervention (SBPI) was designed to address social perceptual challenges observed in EDs and provide a structured, positive social experience. This study explored its feasibility, acceptability, and impact on both self-concept and clinical symptoms. The majority attended, provided follow-up data, and rated the SBPI positively. Primary hypothesis testing, comparing baseline (T1) and initial follow-up (T2) evaluations, showed improvements in both self-concept and clinical targets with large effect sizes ranging from np2 = 0.22 to np2 = 0.42. Because the MANOVA models were not conditional on other covariates, this suggests that time spent in the intervention explains approximately 22% to 42% of the variation in self-concept and clinical symptoms. Negative self-attribution bias diminished, and both state and trait self-esteem improved following the SBPI. Notably, participants also experienced clinical improvements in eating, depression, and anxiety symptoms even though these symptoms were not directly addressed in the SBPI, supporting our model (Figure 1). Collectively, this data suggests that group interventions focused on social interactions, rather than clinical symptoms, may provide a helpful adjunctive treatment for outpatients with EDs.

Social difficulties influence illness susceptibility and maintenance for EDs (Cardi, Tchanturia, & Treasure, 2018), but few treatment approaches have explicitly targeted social perceptions. In the cognitive interpersonal model of AN, social dysfunction and interpersonal disturbances serve as maintenance mechanisms that promote isolation and disconnection, and a variety of approaches to address isolation are deployed for severe and enduring AN (Treasure et. al 2020). Interpersonal psychotherapy (IPT) and interpersonal cognitive affective therapy (ICAT) both share a theoretical framework that suggests improvements in disordered eating symptoms emerge as a secondary outcome after interpersonal aspects of the disorder have been addressed. When compared to traditional approaches such as cognitive behavioral therapy, IPT has been more consistently associated with mid- to long-term clinical improvements among BN and BED groups (Miniati, Callari, Maglio, & Calugi, 2018). Participation in ICAT, meanwhile, has been correlated with significant improvement in bulimic symptoms, global ED severity, and self-oriented cognitions among participants with BN (Wonderlich et al., 2014). In comparison to these interventions, which are structured to include 20-21 sessions, the SBPI achieved social and clinical improvements after only four sessions.

Group therapy may provide an optimal mechanism to target impairments related to social processing. One inpatient group intervention targeted perfectionism, and showed improvements in both that construct (Tchanturia, Larsson, & Adamson, 2016) with a related qualitative study examining participant perceptions described the benefits of social connection and how the group setting helped to normalize one’s own individual pathology (Larsson, Lloyd, Westwood, & Tchanturia, 2018). Across a variety of inpatient groups, social bonding has been a consistent benefit described by participants (Sparrow & Tchanturia, 2016). Patients with EDs often find the transition to outpatient from residential treatments difficult, due to decrease in structure, development of a sense of disconnection, and loss of group momentum fostered during residential treatment (Cockell et al., 2004). Few patients with EDs in our community continue group therapy as outpatients (Brodrick, Harper, Van Enkevort, & McAdams, 2019). Groups designed explicitly for outpatients with EDs, like the SBPI, may provide social connection and peer support at a crucial phase in treatment.

Humans are social creatures, and formation and maintenance of a positive sense of oneself includes both interactions as well as comparisons with other people. The SBPI provided an opportunity to both interact with others as well as to engage in comparisons about cognitive and emotional responses to the in-group shared experiences. The SBPI intentionally shifted the focus of comparisons by group members to their shared humanity and away from ED-defining pathology comparisons.

Limitations

This pilot study did not include a clinical comparison group or a non-treatment group. Since the SBPI was not designed to be implemented as an independent treatment method for EDs, many participants were concurrently receiving other treatments. The range of additional outpatient care received across the sample varied considerably across participants, ranging from no other treatment to intensive outpatient therapy. Thus, it is impossible to determine if there is a causal link between participation and the clinical improvements observed. Further, this study used a convenience sample of self-selected participants with an interest in a group intervention; these participants may have been more motivated to attend, more involved, or more likely to respond positively. The small sample size further limits the internal validity, as a power analysis suggests a minimum sample size of n = 34 is needed for a within-subjects F-test to detect moderate effect sizes (i.e., np2 = 0.0588 with power at .80 and alpha at .05) We have therefore emphasized effect sizes, with large effects were observed in targeted constructs. Acceptability feedback could not be obtained from the participants who withdrew as they did not respond. Ideally, future studies will include comparison cohorts in whom clinical symptoms are tracked over the same duration but that do not attend the SBPI. A larger sample size could also enable between-group analyses based on diagnosis, as differences between AN, BN, and BED may emerge. Finally, all participants were adult women, with most White and non-Hispanic; more diverse samples are needed to see if these results generalize across different genders, races, and ethnicities.

Future Directions

The outcomes data suggest self-concept may be a malleable feature in EDs that shifts concurrently with clinical symptoms, and can be targeted in therapeutic interventions. Future research efforts might consider the impact of the SBPI in an intensive outpatient setting in conjunction with neuroimaging evaluations to assess whether neural changes in perspective-taking can be detected. Further, we encourage clinical researchers in EDs to evaluate self-attribution bias and self-esteem in their treatment studies so that the relevance of these factors in disease progression can be established. Large longitudinal studies of patients with EDs may be able to further clarify how self-concept and social perceptions are related clinical symptom expression, illness, and recovery in EDs.

Conclusion

Outpatients with EDs demonstrated improvements in self-concept and clinical symptoms following participation in a brief, group intervention targeting social behaviors and perceptions. Relative to baseline, participation in the SBPI was associated with acute changes in negative self-attributions and state self-esteem as well as sustained improvements in ED, depression, and anxiety symptoms. There are few evidence-based ED treatments, and none focused solely on psychosocial targets for medically stable patients with EDs. Current treatments for EDs focus on clinical behaviors (i.e. restricting, binge-eating, compensatory practices), as well as ED-specific cognitions (i.e. body shape/weight concerns, low self-esteem, relationship of body/eating to self-worth). We propose that social perceptions should be approached as a possible third target domain that may be critical in a sustained recovery model in adult EDs.

Acknowledgements

Research support was received from the National Institute of Mental Health of under Award Number R01MH112927. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. Data collection was supported by Academic Information Systems grant support, CTSA NIH Grant UL1-RR024982.

Footnotes

Declaration of Interest

The authors declare no conflict of interest.

References

  1. Aloi M, Rania M, Caroleo M, De Fazio P, & Segura-Garcia C (2017). Social cognition and emotional functioning in patients with binge eating disorder. European Eating Disorders Review, 25(3), 172–178. doi: 10.1002/erv.2504 [DOI] [PubMed] [Google Scholar]
  2. Ambwani S, Berenson KR, Simms L, Li A, Corfield F, & Treasure J (2016). Seeing things differently: An experimental investigation of social cognition and interpersonal behavior in anorexia nervosa. International Journal of Eating Disorders, 49(5), 499–506. doi: 10.1002/eat.22498 [DOI] [PubMed] [Google Scholar]
  3. Arcelus J, Mitchell AJ, Wales J, & Nielsen S (2011). Mortality rates in patients with anorexia nervosa and other eating disorders. A meta-analysis of 36 studies. Archives of General Psychiatry, 68(7), 724–731. doi: 10.1001/archgenpsychiatry.2011.74 [DOI] [PubMed] [Google Scholar]
  4. Balestrieri M, Zuanon S, Pellizzari J, Zappoli-Thyrion E, Ciano R, & Res TM (2015). Mentalization in eating disorders: A preliminary trial comparing mentalization-based treatment (MBT) with a psychodynamic-oriented treatment. Eating and Weight Disorders, 20(4), 525–528. doi: 10.1007/s40519-015-0204-1 [DOI] [PubMed] [Google Scholar]
  5. Brodrick B, Harper JA, Van Enkevort E, & McAdams CJ (2019). Treatment utilization and medical problems in a community sample of adult women With anorexia nervosa. Frontiers in Psychology, 10, 981. doi: 10.3389/fpsyg.2019.00981 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Cardi V, Tchanturia K, & Treasure J (2018). Premorbid and illness-related social difficulties in eating disorders: An overview of the literature and treatment developments. Current Neuropharmacology, 16(8), 1122–1130. doi: 10.2174/1570159X16666180118100028 [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Cockell SJ, Zaitsoff SL, & Geller J (2004). Maintaining change following eating disorder treatment. Professional Psychology: Research and Practice, 35(5), 527–534. doi: 10.1037/0735-7028.35.5.527 [DOI] [Google Scholar]
  8. Cohen JA (1969). Statistical power analysis for the behavioral sciences. New York: Academic Press. [Google Scholar]
  9. Eddy KT, Tabri N, Thomas JJ, Murray HB, Keshaviah A, Hastings E, … Franko DL (2017). Recovery from anorexia nervosa and bulimia nervosa at 22-year follow-up. The Journal of Clinical Psychiatry, 78(2), 184–189. doi: 10.4088/JCP.15m10393 [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Harper JA, Brodrick B, Van Enkevort E, & McAdams CJ (2017). Neuropsychological and cognitive correlates of recovery in anorexia nervosa. European Eating Disorders Review, 25(6), 491–500. doi: 10.1002/erv.2539 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Harris PA, Taylor R, Thielke R, Payne J, Gonzalez N, & Conde JG (2009). Research electronic data capture (REDCap)--A metadata-driven methodology and workflow process for providing translational research informatics support. Journal of Biomedical Informatics, 42(2), 377–381. doi: 10.1016/j.jbi.2008.08.010 [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Hason Rozenstein M, Stein D, & Latzer Y (2018). Subjective and objective: Alexithymia and social cognition in eating disorders. Israel Journal of Psychiatry and Related Sciences, 55(1), 45–53. [PubMed] [Google Scholar]
  13. Heatherton TF & Polivy J (1991). Development and validation of a scale for measuring state self-esteem. Journal of Personality and Social Psychology, 60(6), 895–910. doi: 10.1037/0022-3514.60.6.895 [DOI] [Google Scholar]
  14. Keel PK & Brown TA (2010). Update on course and outcome in eating disorders. International Journal of Eating Disorders, 43(3), 195–204. doi: 10.1002/eat.20810 [DOI] [PubMed] [Google Scholar]
  15. Kinderman P & Bentall RP (1996). A new measure of causal locus: The Internal, Personal and Situational Attributions Questionnaire. Personality and Individual Differences, 20(2), 261–264. doi:Doi 10.1016/0191-8869(95)00186-7 [DOI] [Google Scholar]
  16. Kuipers GS & Bekker M (2012). Attachment, mentalization, and eating disorders: A review of studies using the Adult Attachment Interview. Current Psychiatry Reports, 8, 326–336. [Google Scholar]
  17. Laghi F, Cotugno A, Cecere F, Sirolli A, Palazzoni D, & Bosco FM (2014). An exploratory assessment of theory of mind and psychological impairment in patients with bulimia nervosa. British journal of psychology (London, England : 1953), 105(4), 509–523. 10.1111/bjop.12054 [DOI] [PubMed] [Google Scholar]
  18. Larsson E, Lloyd S, Westwood H, & Tchanturia K (2018). Patients’ perspective of a group intervention for perfectionism in anorexia nervosa: A qualitative study. J Health Psychol, 23(12), 1521–1532. doi: 10.1177/1359105316660183 [DOI] [PubMed] [Google Scholar]
  19. Legnani A, Martoni RM, Brombin C, Cugnata F, Porta R, de Filippis R, Erzegovesi S, and Caputi M (2017). Theory of mind in binge eating disorder: An exploratory study. European Psychiatry., 41(S1), S555–S555. doi: 10.1016/j.eurpsy.2017.01.793 [DOI] [Google Scholar]
  20. Linville D, Brown T, Sturm K, & McDougal T (2012). Eating disorders and social support: Perspectives of recovered individuals. Eating Disorders, 20(3), 216–231. doi: 10.1080/10640266.2012.668480 [DOI] [PubMed] [Google Scholar]
  21. Luce KH & Crowther JH (1999). The reliability of the Eating Disorder Examination-Self-Report Questionnaire Version (EDE-Q). International Journal of Eating Disorders, 25(3), 349–351. doi: 10.1002/(sici)1098-108x(199904)25:3<349::aid-eat15>3.0.co;2-m [DOI] [PubMed] [Google Scholar]
  22. Mainz V, Schulte-Ruther M, Fink GR, Herpertz-Dahlmann B, & Konrad K (2012). Structural brain abnormalities in adolescent anorexia nervosa before and after weight recovery and associated hormonal changes. Psychosomatic Medicine, 74(6), 574–582. doi: 10.1097/PSY.0b013e31824ef10e [DOI] [PubMed] [Google Scholar]
  23. McAdams CJ, Harper JA, & Van Enkevort E (2018). Mentalization and the left inferior frontal gyrus and insula. Eur Eat Disord Rev, 26(3), 265–271. doi: 10.1002/erv.2580 [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. McAdams CJ, Jeon-Slaughter H, Evans S, Lohrenz T, Montague PR, & Krawczyk DC (2016). Neural differences in self-perception during illness and after weight-recovery in anorexia nervosa. Social Cognitive and Affective Neuroscience, 11(11), 1823–1831. doi: 10.1093/scan/nsw092 [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. McAdams CJ, & Krawczyk DC (2011). Impaired neural processing of social attribution in anorexia nervosa. Psychiatry Res, 194(1), 54–63. doi: 10.1016/j.pscychresns.2011.06.016 [DOI] [PubMed] [Google Scholar]
  26. McAdams CJ & Krawczyk DC (2013). Neural responses during social and self-knowledge tasks in bulimia nervosa. Frontiers in Psychiatry, 4, 103. doi: 10.3389/fpsyt.2013.00103 [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. McAdams CJ & Krawczyk DC (2014). Who am I? How do I look? Neural differences in self-identity in anorexia nervosa. Social Cognitive and Affective Neuroscience, 9(1), 12–21. doi: 10.1093/scan/nss093 [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Miniati M, Callari A, Maglio A, & Calugi S (2018). Interpersonal psychotherapy for eating disorders: Current perspectives. Psychology Research and Behavior Management, 11, 353–369. doi: 10.2147/PRBM.S120584 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Rosenberg M (1965). Society and the Adolescent Self-Image. Princeton, NJ: Princeton University Press. [Google Scholar]
  30. Rush AJ, Trivedi MH, Ibrahim HM, Carmody TJ, Arnow B, Klein DN, … Keller MB (2003). The 16-Item Quick Inventory of Depressive Symptomatology (QIDS), clinician rating (QIDS-C), and self-report (QIDS-SR): A psychometric evaluation in patients with chronic major depression. Biological Psychiatry, 54(5), 573–583. [DOI] [PubMed] [Google Scholar]
  31. Russell TA, Schmidt U, Doherty L, Young V, & Tchanturia K (2009). Aspects of social cognition in anorexia nervosa: Affective and cognitive theory of mind. Psychiatry Research, 168(3), 181–185. doi: 10.1016/j.psychres.2008.10.028 [DOI] [PubMed] [Google Scholar]
  32. Sacchetti S, Robinson P, Bogaardt A, Clare A, Ouellet-Courtois C, Luyten P, … Fonagy P (2019). Reduced mentalizing in patients with bulimia nervosa and features of borderline personality disorder: A case-control study. BMC Psychiatry, 19(1), 134. doi: 10.1186/s12888-019-2112-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Shear MK, Vander Bilt J, Rucci P, Endicott J, Lydiard B, Otto MW, … Frank DM (2001). Reliability and validity of a structured interview guide for the Hamilton Anxiety Rating Scale (SIGH-A). Depression and Anxiety, 13(4), 166–178. doi: 10.1002/da.1033 [pii] [DOI] [PubMed] [Google Scholar]
  34. Sheehan DV, Lecrubier Y, Sheehan KH, Amorim P, Janavs J, Weiller E, … Dunbar GC (1998). The Mini-International Neuropsychiatric Interview (M.I.N.I.): The development and validation of a structured diagnostic psychiatric interview for DSM-IV and ICD-10. Journal of Clinical Psychiatry, 59 Suppl 20, 22–33;quiz 34-57. [PubMed] [Google Scholar]
  35. Smink FR, van Hoeken D, & Hoek HW (2012). Epidemiology of eating disorders: Incidence, prevalence and mortality rates. Current Psychiatry Reports, 14(4), 406–414. doi: 10.1007/s11920-012-0282-y [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Sparrow KA, & Tchanturia K (2016). Inpatient Brief Group Therapy for Anorexia Nervosa: Patient Experience. International Journal of Group Psychotherapy, 66(3), 431–442. doi: 10.1080/00207284.2016.1156406 [DOI] [PubMed] [Google Scholar]
  37. Sysko R, Glasofer DR, Hildebrandt T, Klimek P, Mitchell JE, Berg KC, … Walsh BT (2015). The Eating Disorder Assessment for DSM-5 (EDA-5): Development and validation of a structured interview for feeding and eating disorders. International Journal of Eating Disorders, 48(5), 452–463. doi: 10.1002/eat.22388 [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Tchanturia K, Larsson E, & Adamson J (2016). Brief Group Intervention Targeting Perfectionism in Adults with Anorexia Nervosa: Empirically Informed Protocol. Eur Eat Disord Rev, 24(6), 489–493. doi: 10.1002/erv.2467 [DOI] [PubMed] [Google Scholar]
  39. Tozzi F, Sullivan PF, Fear JL, McKenzie J, & Bulik CM (2003). Causes and recovery in anorexia nervosa: The patient’s perspective. International Journal of Eating Disorders, 33(2), 143–154. doi: 10.1002/eat.10120 [DOI] [PubMed] [Google Scholar]
  40. Treasure J, Willmott D, Ambwani S, Cardi V, Bryan DC, Rowlands K, & Schmidt U (2020). Cognitive Interpersonal Model for Anorexia Nervosa Revisited: The Perpetuating Factors that Contribute to the Development of the Severe and Enduring Illness. Journal of Clinical Medicine, 9(3). [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Westwood H, Kerr-Gaffney J, Stahl D, & Tchanturia K (2017). Alexithymia in eating disorders: Systematic review and meta-analyses of studies using the Toronto Alexithymia Scale. Jounral of Psychosomatic Research, 99, 66–81. doi: 10.1016/j.jpsychores.2017.06.007 [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Wonderlich SA, Peterson CB, Crosby RD, Smith TL, Klein MH, Mitchell JE, & Crow SJ (2014). A randomized controlled comparison of integrative cognitive-affective therapy (ICAT) and enhanced cognitive-behavioral therapy (CBT-E) for bulimia nervosa. Psychological Medicine, 44(3), 543–553. doi: 10.1017/S0033291713001098 [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES