Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2023 Dec 1.
Published in final edited form as: Body Image. 2022 Aug 16;43:1–7. doi: 10.1016/j.bodyim.2022.08.003

Understanding the role of positive body image during digital interventions for eating disorders: Secondary analyses of a randomized controlled trial

Jake Linardon a,b,*, Tracy L Tylka c, C Blair Burnette d, Adrian Shatte e, Matthew Fuller-Tyszkiewicz a,b
PMCID: PMC9933246  NIHMSID: NIHMS1872241  PMID: 35985097

Abstract

Despite growing interest in the possible link between positive body image and eating disorder (ED) symptoms, little is known about what role this adaptive construct plays in ED treatment. This study investigated whether: (1) interventions principally designed to target ED psychopathology also lead to improvements in positive body image indices (i.e., body appreciation, functionality appreciation, and body image flexibility); (2) changes in ED symptoms correlate with changes in positive body image, both concurrently and prospectively; and (3) baseline positive body image levels moderate the degree of symptom improvement. Secondary analyses from a randomized controlled trial on digital interventions for EDs (n=600) were conducted. Intervention participants reported greater increases in the three positive body image constructs than the control group (ds=0.15–0.41). Greater pre-post reductions in ED psychopathology and binge eating were associated with greater pre-post improvements in positive body image indices. However, earlier reductions in ED psychopathology and binge eating did not predict later improvements in positive body image at follow-up. None of the positive body image constructs at baseline moderated degree of symptom change. Standard ED interventions can cultivate a more positive body image, although this is not explained by earlier symptom reduction. Understanding the mechanisms through which ED interventions enhance positive body image is needed.

Keywords: Eating disorders, Positive body image, Body appreciation, Body image flexibility, Randomized controlled trial, Binge eating, Digital interventions, Functionality appreciation

1. Introduction

Eating disorders (EDs) are common psychiatric conditions that are associated with impairments in functioning (Klump et al., 2009). To date, intervention protocols (e.g., cognitive-behavioral therapy; CBT) that have accrued the most empirical support for their efficacy are largely devoted towards targeting the psychopathology that underpins EDs, such as body image disturbances, restrictive eating, and intolerance of mood states (Fairburn, 2008). Thus, the success of such interventions has mostly been defined in terms of the absence of — or reduction in — ED behaviors or negative body-related concerns (Williams et al., 2012). As a consequence, treatment outcome research has neglected consideration of the adoption or presence of positive aspects of psychological functioning.

Positive body image has been recently shown to protect against the onset of ED symptoms (Linardon, 2021), indicating that a better understanding of how to cultivate positive body image during treatment is warranted (Cook-Cottone, 2015). Positive body image is a multifaceted construct that can be broadly defined as an individual’s ability to love, accept, respect, be mindful of, and appreciate their body regardless of its appearance or functional ability (Webb et al., 2015). Although positive body image is theorized to be distinct from negative body image, the robust inverse association found between the two suggests that addressing one could indirectly affect the other (Tylka & Wood-Barcalow, 2015b; Linardon et al., 2022). Thus, it is crucial to understand whether positive body image increases during ED treatment, and what other roles it might play in this context (e.g., an outcome predictor, mechanism of change).

The most widely studied facets of positive body image are body appreciation, functionality appreciation, and body image flexibility. Whereas body appreciation refers to the ability for one to accept, value, and hold favorable opinions towards the body (Tylka & Wood-Barcalow, 2015a), functionality appreciation reflects a tendency to be grateful for what the body is capable of doing (Alleva et al., 2017). Both components are thought to combat eating and body image disturbances through similar underlying mechanisms. They enable a person to positively reframe the way they think about their body; this in turn prevents the person from focusing on those perceived bodily imperfections, which is known to precipitate ED symptoms (Alleva et al., 2018; Avalos & Tylka, 2006). In contrast, body image flexibility refers to the tendency for one to openly experience negative thoughts and feelings about the body without acting on or trying to change them (Sandoz et al., 2013). Body image flexibility is thought to be relevant to those ED behaviors that are directly precipitated by negative thoughts, feelings, and sensations related to the body. Individuals high in this type of flexibility are able to recognize these adverse internal experiences, attend to them nonjudgmentally, and continue to pursue meaningful life goals without trying to supress them via unhealthy coping strategies (Sandoz et al., 2013).

When studying the role of positive body image in ED interventions, three questions are worth considering. The first involves whether ED interventions can enhance positive body image. It is possible that certain therapeutic techniques common in ED interventions have much broader effects than mere symptom reduction, extending to the acquisition of a positive body image. For instance, mindfulness meditation, body scan, or acceptance-based exercises might not only combat mood dysregulation, but also help the person embrace intrusive body-related thoughts. Likewise, mirror exposure exercises might not only target body image avoidance, but also help an individual recognize and respect those unique aspects of their body that were previously avoided. Evaluating the effects of ED interventions on positive body image may have implications for the delivery of more tailored treatment plans. More specifically, if acquiring a positive body image is a desired end-state for certain individuals, then small to no change in positive body image could signal a need to modify an existing treatment plan by incorporating additional strategies known to cultivate a positive body image (e.g., Alleva et al., 2015).

The second question concerns whether changes in positive body image constructs are associated with the degree of symptom change (and vice versa). Investigating relations between changes in positive body image and symptom reduction may advance understanding of the nature of these adaptive and maladaptive processes, inform theoretical models of body image and EDs, and elucidate possible mechanisms of action (Kazdin, 2007; Murphy et al., 2009).

The third question worthy of consideration in ED intervention research concerns whether positive body image bears any prognostic importance. Testing whether the degree of symptom change can be predicted by baseline levels of positive body image could help identify certain individuals that are most (or least) responsive to that type of intervention protocol (Kraemer, 2016). In light of strong associations observed between positive body image facets and willingness to adopt healthier, unrestrained eating patterns (Linardon et al., 2021), a higher positive body image could signal greater likelihood of success, as such individuals may be more inclined to engage in those confronting – yet effective – strategies that target restrictive eating patterns (e.g., regular eating, exposure tasks, food monitoring, etc.).

Research addressing these three key questions is limited. Some uncontrolled studies have observed improvements in body image flexibility following specialized ED treatment (Bluett et al., 2016; Lee et al., 2018). Pellizzer and colleagues also found higher baseline levels of body image flexibility to predict greater symptom change (Pellizzer et al., 2018) and early improvement in body image flexibility during CBT to temporally predict greater symptom reduction (Pellizzer et al., 2019). As it stands, the role of body appreciation and functionality appreciation as predictors, mediators and outcomes during ED interventions are unknown.

The present study seeks to address these three research questions by conducting secondary analyses of a randomized controlled trial (RCT) of digital interventions for EDs. The aims of this paper are threefold: First, to assess whether improvements in three indices of positive body image are observed from digital interventions (relative to a control condition) principally designed to target ED psychopathology. Second, to examine whether the degree of change in these positive body image constructs is associated with the degree of change in ED symptoms, both concurrently and prospectively. Third, to explore whether baseline levels of positive body image constructs predict the degree of symptom change at follow-up.

2. Method

2.1. Design

This study is a secondary analysis of a larger, fully remote, three-armed RCT comparing two digital interventions for EDs against a control condition (Linardon et al., 2022). Assessments were conducted at baseline (T1), 4-weeks post-intervention (T2), and 8-weeks follow-up (T3). This study received ethics approval from Deakin University’s Human Research Ethics Committee, and was pre-registered (ACTRN12621000914864). All participants provided informed consent.

2.2. Participants and procedure

Participants were recruited in July-August 2021 via advertisements distributed throughout the first author’s psychoeducational platform for EDs. We refer readers to Linardon, Rosato, et al. (2020) for a detailed description of this platform and the demographic and clinical profile its visitors. Respondents to advertisements first completed a brief screening survey to determine eligibility. Participants were eligible if they: (1) were aged 18 years or over; (2) had access to the Internet and a smartphone; and (3) reported the presence of recurrent objective binge eating, defined as one episode every two weeks, on average, over the past three months. Participants who met eligibility criteria went onto complete the baseline assessments, after which they were randomized to either a broad digital intervention (n = 199), a focused digital intervention (n = 199), or a waitlist (n = 202). Randomization took place on a 1:1:1 ratio generated through an automated computer-based random number sequence provided in Qualtrics.

2.3. Intervention groups

Details about the design, content, digital functionality, and exercises of the two digital intervention programs have been published elsewhere (Linardon et al., 2022). Below, we briefly describe the structure of these two self-guided programs.

2.3.1. Broad program

The broad program (“Break Binge Eating”) contains four modules (60–90 minutes to complete) that are designed to target three hypothesized ED maintaining mechanisms. Content and exercises are based off established CBT protocols (Fairburn, 2008). Module 1 was psychoeducational in nature, aiming to teach the user about the nature of ED symptoms. Module 2 was designed to target restrictive eating practices via self-monitoring and establishing a regular eating pattern. Module 3 was designed to target mood dysregulation via mastery of problem solving, acceptance, and mindfulness meditation skills. Module 4 was designed to target negative body image through skills that aimed to help users broaden their scheme of self-worth and reject culturally-determined appearance ideals. Of note, there was one exercise within this module that asked users to reflect on the functionality of their body.

2.3.2. Focused program

The focused intervention (“Breaking the Diet Cycle”), also based on cognitive-behavioral principles, was designed to exclusively target restrictive eating practices. There were four sessions to this program, with each session teaching the user a broader skill designed to modify extreme eating restraint. Session 1 educated users about the risks of extreme diets and its relationship to disordered eating. Session 2 taught users how to effectively monitor their eating patterns in real-time. Session 3 taught users how to establish a pattern of regular eating as a method for combating those disordered eating behaviors precipitated by dietary restriction. Session 4 taught users how to alleviate irrational food rules and anxiety via exposure exercises.

2.4. Control group

Control participants were placed on a waitlist and completed the same assessments at baseline, post-test, and follow-up. After completing the post-test survey, control participants were given access to one of the two interventions.

2.5. Study measures

2.5.1. Participant characteristics

At baseline, participants indicated their age, gender, ethnicity, education level, current and prior mental health status, and current treatment status. Motivation and perceived confidence in ability to change were also assessed via a single item.

2.5.2. Positive body image constructs

2.5.2.1. Body appreciation.

Body appreciation was assessed via the 3-item short-form of the Body Appreciation Scale-2 (Tylka et al., 2022): the BAS-2SF. The BAS-2SF contains three items (Items 1, 6, and 7 of the BAS-2) that were derived deductively from theory. Items are rated on a 5-point scale, ranging from 1 (never) to 5 (always). Scores are averaged to produce a total score, with higher scores reflecting greater appreciation. The psychometric performance of the 3-item BAS-2SF was evaluated in five studies, totaling 3114 participants (Tylka et al., 2022). Across these studies, the 3-item BAS-2SF had a near-perfect correlation with the 10-item BAS-2 (rs =.93–.97), had a well-replicated unidimensional structure that was invariant between men and women, was internally consistent (ω =.86–.92), and demonstrated convergent and incremental validity.

2.5.2.2. Functionality appreciation.

Functionality appreciation was assessed via the 7-item Functionality Appreciation Scale (FAS; Alleva et al., 2017). Items are rated along a Likert scale, ranging from 1 (strongly disagree) to 5 (strongly agree). Scores are averaged to produce a total score, with higher scores reflecting greater functionality appreciation. The FAS has exhibited a unidimensional structure, and evidence of internal consistency, convergent validity, and 3-week test–retest reliability has been found (Alleva et al., 2017; Linardon, Messer, et al., 2020).

2.5.2.3. Body image flexibility.

Body image flexibility was assessed via the abbreviated 5-item version (Basarkod et al., 2018) of the Body Image Acceptance and Action Questionnaire (BI-AAQ; Sandoz et al., 2013). The five items are rated on a 7-point scale ranging from 1 (never true) to7 (always true), are reverse scored, and then summed to produce a total score. Higher scores reflect higher levels of flexibility. The abbreviated BI-AAQ has exhibited a unidimensional structure, is internally consistent, and has almost a near-perfect correlation with the full version (Basarkod et al., 2018; Linardon, Messer, et al., 2020).

2.5.3. Eating disorder symptoms

We used two measures of ED symptoms for the present study: objective binge eating and total levels of ED psychopathology. Objective binge eating was assessed via a single item asking participants to report the number of episodes experienced in the past 28 days. Total levels of ED psychopathology was assessed using the global score from the 28-item Eating Disorder Examination Questionnaire (EDE-Q; Fairburn & Beglin, 1994). The global score is calculated by averaging participants’ responses on the four subscales (eating concerns, weight concerns, shape concerns, and dietary restraint). Higher global scores reflect greater levels of psychopathology experienced over the past 28 days. The EDE-Q global score is one of the most widely used outcome measures in clinical trials, and its psychometric performance has been reported elsewhere (Berg et al., 2012).

2.6. Statistical analyses

Prior analyses (see Supplementary Tables 1–3) showed that the two intervention arms exhibited equivalent scores at baseline and levels of change in ED symptoms and positive body image over time. The two intervention arms did not differ in the magnitude of association between change in positive and negative body image constructs (and ED symptoms). Thus, for the purpose of this study, we combined both conditions into a single intervention arm. Subsequently, participants were retained in these revised treatment arms (intervention or wait-list control) following intention-to-treat principles to handle missing data across time-points. Missingness was handled with multiple imputation (n = 50 imputations) derived via the fully conditional specification method. Analyses (detailed below) on each imputation were subsequently pooled using Rubin’s (1987) rules.

Linear mixed models (continuous outcomes) and negative binomial mixed models (count-based outcomes) were used for all hypothesis testing. To address Research Aim 1, scores on outcomes (i.e., positive body image indices) were regressed onto time (baseline vs post-intervention), group (control vs intervention arm), and their interaction (time × group). Research Aim 2 was evaluated by constructing difference scores (e.g., T1–T2 and T2–T3) that were then used as a moderator of time effects for the particular outcome. In concurrent change models, T1 to T2 and T2 to T3 differences in the proposed moderator were paired with change in the same time-frame for the outcome variable. For prospective change models, T1 to T2 change in the proposed moderator was paired with change from T2 to T3 for the outcome variable(s). Separate models were constructed for each predictor. Research Aim 3 was tested by using baseline levels of positive body image constructs as a moderator of time and group effects on ED psychopathology and objective binge eating (2- and 3-way interactions modeled).

Effect sizes for continuous outcomes are reported as standardized mean differences calculated as per Feingold (2017), whereas incidence rate ratios were used for count-based outcomes. All analyses were undertaken using Stata version 16.

Sample size calculations were conducted pre-analysis for the primary RCT from which this secondary analysis derives (Linardon et al., 2022). However, based on a sample size of 202 participants for the control group and 398 for the combined intervention group, setting power at.80 and alpha at.05 (two-tailed), we are powered to detect small effects: group mean difference between intervention and controls > 0.24 for Research Aim 1, and a within group (intervention arm) correlation > 0.14 for Research Aims 2 and 3.

3. Results

3.1. Participant flow

Six-hundred-forty individuals expressed interest in the trial, with 600 meeting eligibility criteria and being randomized to one of the three groups. A total of 367 provided post-test data on study variables, including 155 for the control group and 212 for the combined intervention group. For the intervention group, a combined total of 175 participants provided follow-up data on study variables and 45% completed at least 50% of program content. For the specific modules/sections, 54% completed at least 50% of Module/Section 1, 35% for Module/Section 2, 26% for Module/Section 3, and 18% for Module/Section 4.

3.2. Baseline characteristics

Supplementary Table 1 presents the baseline characteristics across study conditions. The mean age of the total sample was 33.85 (SD = 9.82; minimum = 18 and maximum = 75) and most participants identified as White (87%) women (93%) who finished high school (91%). The mean EDE-Q global score at baseline was 3.96 (SD = 0.99), which falls within one standard deviation of established clinical norms (Aardoom et al., 2012). There were no significant differences between study conditions on any baseline measure, indicating that randomization was successful.

3.3. Main analyses

3.3.1. Aim 1: do digital interventions lead to improvements in positive body image?

Table 1 presents the results from the analyses comparing the combined intervention group against the control group at post-test on outcome variables. Mean change, effect sizes, and p-values of the tests of differential change for outcomes are presented. The mean change differences in body appreciation, functionality appreciation, and body image flexibility between the intervention and control group were statistically significant. In all cases, the intervention group reported greater improvements in these positive body image constructs than the control group. The effect size was strongest for body image flexibility, followed by functionality appreciation and body appreciation. The intervention group also reported significantly greater reductions in ED psychopathology and objective binge eating than the control group at post-test. Within-group effects from T2 to T3 were d = 0.48 for body appreciation, d = 0.11 for functionality appreciation, and d = 0.39 for body image flexibility, indicating further improvements in positive body image were made after the post-test period.

Table 1.

Means, standard deviations, and change scores on positive body image and eating disorder variables.

Study condition Difference in change score (T1-T2)
Control Intervention Intervention - control
Outcome n M (SD) n M (SD) M change (95% CI)^ ES between p
Body appreciation
Baseline 202 2.34 (0.75) 398 2.40 (0.71)
Post-intervention 155 2.47 (0.78) 212 2.69 (0.79) 0.11 (0.01, 0.22) 0.15 .004
Follow-up - - 175 2.93 (0.84)
Functionality appreciation
Baseline 202 3.45 (0.83) 398 3.50 (0.76)
Post-intervention 155 3.44 (0.78) 212 3.79 (0.71) 0.28 (0.16, 0.39) 0.36 < .001
Follow-up - - 175 3.85 (0.77)
Body image flexibility
Baseline 202 12.77 (5.23) 398 12.38 (5.73)
Post-intervention 155 12.81 (5.70) 212 14.73 (6.81) 2.28 (1.37, 3.20) 0.41 < .001
Follow-up - - 175 16.50 (7.61)
ED psychopathology
Baseline 202 3.92 (0.97) 398 3.50 (0.76)
Post-intervention 155 3.85 (0.96) 214 3.00 (1.19) −0.81 (−1.00, −0.63) −0.87 < .001
Follow-up 178 2.75 (1.25)
Objective binge episodes
Baseline 202 16.73 (15.73) 398 16.79 (13.58)
Post-intervention 155 17.78 (16.16) 221 10.44 (8.15) −0.50 (−0.66, −0.34) 0.61 < .001
Follow-up 185 10.66 (11.53)
Open in a new tab

M and SD values are based on non-imputed data; mean differences and effect sizes are derived from ITT analysis (n = 600) using multiple imputation; ES = effect size as standardized mean difference for continuous outcomes and incidence rate ratios for count outcomes (objective and subjective binge episodes).

^

Note that the M change variable reflects the time x condition interaction effect expressed in raw units. A positive value indicates that the intervention arm experienced greater increase in scores at post-intervention relative to the waitlist control arm, whereas a negative value indicates that the intervention arm experienced greater reduction in scores at post-intervention

At baseline, 48.76% of participants in the waitlist condition and 54.52% of participants in the combined intervention arm exceed clinical norms for the EDE-Q (Aardoom et al., 2012). At post-intervention, 44.46% of participants in the waitlist condition and 22.86% of participants in the combined intervention arm exceed these clinical norms. At the follow-up time-point, 19.21% of participants in the combined intervention arm exceed these clinical norms.

3.3.2. Aim 2: are changes in eating disorder symptoms associated with changes in positive body image (and vice versa)?

Table 2 presents the results testing whether any concurrent and prospective associations exist between changes in ED outcomes and positive body image constructs. Concurrently, greater improvements in ED psychopathology from T1 to T2 and from T2 to T3 were associated with greater improvements in positive body image constructs at the same time point (except T2–T3 changes in functionality appreciation failed to reach significance). Greater improvements in objective binge eating from T1 to T2 were also associated with greater concurrent improvements in functionality appreciation and body image flexibility. Prospectively, changes in ED psychopathology and objective binge eating from T1 to T2 were not significantly associated with T2–T3 changes in any positive body image construct. Similarly, changes in the three positive body image constructs from T1 to T2 were not significantly associated with T2–T3 changes in ED psychopathology.

Table 2.

Associations between changes in positive body image and changes in eating disorder variables.

Outcome Outcome Time point Change in predictor b (95% CI)^ p
Body appreciation T1-T2
ED psychopathology (T1-T2) 0.12 (0.07, 0.17) < .001
Objective binge episodes (T1-T2) 0.00 (0.00, 0.01) .105
T2-T3
ED psychopathology (T1-T2) −0.03 (−0.10, 0.03) .318
Objective binge episodes (T1-T2) 0.00 (−0.01, 0.01) .931
ED psychopathology (T2-T3) 0.13 (0.03, 0.22) .008
Objective binge episodes (T2-T3) −0.01 (−0.02, 0.00) .029
Functionality appreciation T1-T2
ED psychopathology (T1-T2) 0.10 (0.51, 0.16) < .001
Objective binge episodes (T1-T2) 0.01 (0.00, 0.01) .013
T2-T3
ED psychopathology (T1-T2) −0.01 (−0.07, 0.03) .483
Objective binge episodes (T1-T2) 0.00 (0.00, 0.01) .569
ED psychopathology (T2-T3) 0.06 (−0.01, 0.13) .107
Objective binge episodes (T2-T3) 0.00 (−0.01, 0.00) .512
Body image flexibility T1-T2
ED psychopathology (T1-T2) 1.81 (1.36, 2.75) < .001
Objective binge episodes (T1-T2) 0.04 (0.01, 0.08) .008
T2-T3
ED psychopathology (T1-T2) −0.31 (−0.84, 0.22) .256
Objective binge episodes (T1-T2) 0.02 (−0.02, 0.07) .343
ED psychopathology (T2-T3) 1.12 (0.42, 1.83) .002
Objective binge episodes (T2-T3) −0.03 (−0.09, 0.02) .256
ED psychopathology T2-T3
Body appreciation change (T1–T2) 0.04 (−0.16, 0.24) .685
Functionality appreciation change (T1–T2) 0.01 (−0.21, 0.22) .963
Body image flexibility change (T1 – T2) 0.01 (−0.01, 0.03) .412
Body appreciation change (T2–T3) −0.25 (−0.44, −0.06) .009
Functionality appreciation change (T2–T3) −0.18 (−0.41, 0.04) .113
Body image flexibility change (T2 – T3) −0.04 (−0.07, −0.01) .002
Objective binge episodes T2-T3
Body appreciation change (T1–T2) 0.08 (−0.14, 0.29) .486
Functionality appreciation change (T1–T2) −0.01 (−0.23, 0.21) .943
Body image flexibility change (T1 – T2) 0.00 (−0.02, 0.02) .999
Body appreciation change (T2–T3) −0.19 (−0.37, 0.00) .050
Functionality appreciation change (T2–T3) −0.11 (−0.33, 0.12) .341
Body image flexibility change (T2 – T3) −0.02 (−0.04, 0.01) .192
Open in a new tab

Note:

^

These b weights (unstandardized coefficients) reflect the moderating effect of change in these predictors (as difference scores) on the time x group (T1 to T2 change) and time effect (T2 to T3 change, since no waitlist group at T3) on DVs

3.3.3. Aim 3: do baseline levels of positive body image predict change in eating disorder symptoms?

Table 3 presents the results demonstrating whether baseline levels of positive body image constructs predict changes in ED psychopathology and objective binge eating. No significant associations were observed for any variable at any time-point.1

Table 3.

Associations between baseline levels of positive body image and change in eating disorder symptoms.

ED psychopathology (T1-T2) ED psychopathology (T2-T3)
Baseline predictor b (95% CI) p b (95% CI) p
Body appreciation 0.06 (−0.17, 0.31) .596 0.07 (−0.08, 0.24) .362
Functionality appreciation 0.04 (−0.18, 0.26) .719 0.01 (−0.14, 0.17) .869
Body image flexibility 0.01 (−0.02, 0.04) .493 0.00 (−0.02, 0.02) .944
Objective binge eating (T1-T2) Objective binge eating (T2-T3)
Baseline predictor b (95% CI) p b (95% CI) p
Body appreciation 0.04 (−0.18, 0.26) .718 0.01 (−0.14, 0.17) .864
Functionality appreciation 0.06 (−0.15, 0.26) .597 0.02 (−0.14, 0.17) .841
Body image flexibility −0.01 (−0.03, 0.02) .715 0.00 (−0.02, 0.02) .948
Open in a new tab

4. Discussion

Several noteworthy findings emerged from this study. First, digital interventions designed to target ED symptoms also result in improvements in positive body image constructs. Second, despite observing concurrent associations between changes in positive body image constructs and ED symptoms, early symptom change did not predict subsequent improvements in positive body image at follow-up. Third, levels of positive body image constructs at baseline did not predict the degree of symptom change over the course of the intervention. Together, findings shed light on what relevance positive body image might have during standard interventions for EDs.

A main finding was that positive body image constructs increased after an ED intervention. This is consistent with prior studies reporting increases in body image flexibility during face-to-face ED treatment (Bluett et al., 2016; Lee et al., 2018), but extends from them by showing that improvements (albeit somewhat smaller in strength) are also made on other positive body image facets. Importantly, these improvements were not simply a by-product of earlier symptom change, indicating that positive body image improvements are likely explained by other underlying change mechanisms. It is possible that certain techniques designed to target ED psychopathology also impact positive body image indirectly. For instance, many acceptance- and mindfulness-based techniques may not only help with distress intolerance, but could also enable an individual to refrain from acting on or suppressing intrusive thoughts about the body (i.e., body image flexibility). Alternatively, activity scheduling (i.e., a technique that encourages one to engage in new activities to reduce the degree of body image-based self-worth) might foster a sense of appreciation for the body and its functionality based on some of the chosen activities participated in.

An alternative explanation for finding no temporal associations between changes in study variables could be due to the timing of measurements. It is possible that our follow-up assessments might not align with when these variables actually influence each other, thereby underestimating true effects (for further detail, see Timmons & Preacher, 2015). Perhaps such associations are better detected at the state level, with momentary reductions in ED symptoms resulting in immediate increases in positive body image (and vice versa). Alternatively, associations may be detected over longer follow-ups. For example, a recent eight month prospective study found that increases in the three positive body image facets predicted a lower likelihood of symptom onset at follow-up (Linardon, 2021). Future research is needed to clarify the optimal timing of assessments needed to test associations between changes in these variables.

Neither positive body image construct at baseline predicted the degree of symptom change at follow-up, suggesting that responsiveness to digital interventions does not depend on an individual’s prior level of positive body image. This finding stands in contrast to recent work by Pellizzer et al. (2018) who found higher baseline levels of body image flexibility to predict greater symptom improvement. This discrepancy could be explained by differences in treatment and sample characteristics. Whereas we delivered a self-guided digital intervention to individuals with recurrent binge eating, Pellizzer et al. (2018) delivered outpatient, therapist-led, CBT to a transdiagnostic sample. Since the prognostic value of certain outcome predictors has been shown to vary as a function of intervention modality and diagnostic type (Linardon et al., 2016), it is possible that the predictive importance of positive body image is limited to more intensive treatments.

Present findings must be interpreted in light of their limitations. First, given the lengthy assessment battery administered as part of this larger trial, only three facets of positive body image were assessed. There are many other positive body image facets that could each be relevant to ED treatment in their own way, such as broad conceptualization of beauty, body pride, and body sanctification (Webb et al., 2015). Future research may benefit from better understanding the role of the various facets of positive body image during treatment, ideally by a network intervention analysis (Blanken et al., 2019). Second, the timing of follow-up assessments was limited to two time-points spaced one month apart, potentially masking associations between study variables. Incorporating brief assessments of study variables daily or at longer-term follow-up periods may have shown a different pattern of results. Third, as participants were recruited from the community, findings cannot necessarily be generalized to clinical or treatment-seeking populations. Fourth, our outcomes pertaining to eating disorder symptoms were based off the EDE-Q, which has been shown to yield questionable psychometric performance in certain circumstances (Rand-Giovannetti et al., 2020). Perhaps consideration of different eating disorder measures may have produced a different pattern of results.

4.1. Conclusion

This study was the first to examine the role of three indices of positive body image in digital interventions for EDs. Findings demonstrate that a standard intervention program for EDs can lead to concomitant improvements in positive body image constructs. However, we found evidence that this effect does not appear to be accounted for by earlier symptom reduction. Positive body image at baseline was unrelated to the degree of symptom-level change, suggesting that individual differences in these constructs may not be a marker of intervention success. Future intervention research in EDs should consider incorporating a variety of measures of positive body image to confirm these findings and to better understand what role these adaptive processes might play before, during, and after treatment.

Supplementary Material

Supplementary Material

Funding

J.L. (APP1196948) received a National Health and Medical Research Council Investigator Grant.

Footnotes

Conflict of interest statement

We wish to confirm that there are no conflicts of interest.

CRediT authorship contribution statement

JL, TT, CBB, AS, MFT: Conceptualization. JL: Funding acquisition. JL, MFT: Formal analysis. JL, TT, CBB, AS, MFT: Drafting and editing. TT, MFT: Supervision.

Ethical Statement

Procedures for this study were approved by Deakin University. Informed consent was obtained from all participants. Research was conducted in accordance to the Declaration of Helsinky.

Appendix A. Supporting information

Supplementary data associated with this article can be found in the online version at doi:10.1016/j.bodyim.2022.08.003.

1

We also tested for potential interactions between indices of positive and negative (shape/weight concerns) body image at baseline on changes in binge eating. No main or interaction effects emerged.

References

  1. Aardoom JJ, Dingemans AE, Op’t Landt MCS, & Van Furth EF (2012). Norms and discriminative validity of the Eating Disorder Examination Questionnaire (EDE-Q). Eating Behaviors, 13, 305–309. 10.1016/j.eatbeh.2012.09.002 [DOI] [PubMed] [Google Scholar]
  2. Alleva JM, Diedrichs PC, Halliwell E, Martijn C, Stuijfzand BG, Treneman-Evans G, & Rumsey N (2018). A randomised-controlled trial investigating potential underlying mechanisms of a functionality-based approach to improving women’s body image. Body Image, 25, 85–96. 10.1016/j.bodyim.2018.02.009 [DOI] [PubMed] [Google Scholar]
  3. Alleva JM, Martijn C, Van Breukelen GJP, Jansen A, & Karos K (2015). Expand Your Horizon: A programme that improves body image and reduces self-objectification by training women to focus on body functionality. Body Image, 15, 81–89. 10.1016/j.bodyim.2015.07.001 [DOI] [PubMed] [Google Scholar]
  4. Alleva JM, Tylka TL, & Kroon Van Diest AM (2017). The Functionality Appreciation Scale (FAS): Development and psychometric evaluation in US Community women and men. Body Image, 23, 28–44. 10.1016/j.bodyim.2017.07.008 [DOI] [PubMed] [Google Scholar]
  5. Avalos L, & Tylka TL (2006). Exploring a model of intuitive eating with college women. Journal of Counseling Psychology, 53, 486–497. 10.1037/0022-0167.53.4.486 [DOI] [Google Scholar]
  6. Basarkod G, Sahdra B, & Ciarrochi J (2018). Body Image–Acceptance and Action Questionnaire–5: An abbreviation using genetic algorithms. Behavior Therapy, 49, 388–402. 10.1016/j.beth.2017.09.006 [DOI] [PubMed] [Google Scholar]
  7. Berg KC, Peterson CB, Frazier P, & Crow SJ (2012). Psychometric evaluation of the Eating Disorder Examination and Eating Disorder Examination-Questionnaire: A systematic review of the literature. International Journal of Eating Disorders, 45, 428–438. 10.1002/eat.20931 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Blanken TF, Van Der Zweerde T, Van Straten A, Van Someren EJ, Borsboom D, & Lancee J (2019). Introducing network intervention analysis to investigate sequential, symptom-specific treatment effects: A demonstration in co-occurring insomnia and depression. Psychotherapy and Psychosomatics, 88, 52–54. 10.1159/000495045 [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Bluett E, Lee E, Simone M, Lockhart G, Twohig M, Lensegrav-Benson T, & Quakenbush-Roberts B (2016). The role of body image psychological flexibility on the treatment of eating disorders in a residential facility. Eating Behaviors, 23, 150–155. 10.1016/j.eatbeh.2016.10.002 [DOI] [PubMed] [Google Scholar]
  10. Cook-Cottone CP (2015). Incorporating positive body image into the treatment of eating disorders: A model for attunement and mindful self-care. Body Image, 14, 158–167. 10.1016/j.bodyim.2015.03.004 [DOI] [PubMed] [Google Scholar]
  11. Fairburn CG (2008). Cognitive behavior therapy and eating disorders. Guilford Press. [Google Scholar]
  12. Fairburn CG, & Beglin S (1994). Assessment of eating disorders: Interview or self-report questionnaire? International Journal of Eating Disorders, 16, 363–370. 10.1002/1098-108X [DOI] [PubMed] [Google Scholar]
  13. Feingold A (2017). Meta-analysis with standardized effect sizes from multilevel and latent growth models. Journal of Consulting and Clinical Psychology, 85, 262–266. 10.1037/ccp0000162-266 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Kazdin AE (2007). Mediators and mechanisms of change in psychotherapy research. Annual Review of Clinical Psychology, 3, 1–27. 10.1146/annurev.clinpsy.3.022806.091432 [DOI] [PubMed] [Google Scholar]
  15. Klump KL, Bulik CM, Kaye WH, Treasure J, & Tyson E (2009). Academy for Eating Disorders position paper: Eating disorders are serious mental illnesses. International Journal of Eating Disorders, 42, 97–103. 10.1002/eat.20589 [DOI] [PubMed] [Google Scholar]
  16. Kraemer HC (2016). Messages for clinicians: Moderators and mediators of treatment outcome in randomized clinical trials. American Journal of Psychiatry, 173, 672–679. 10.1001/archpsyc.59.10.877 [DOI] [PubMed] [Google Scholar]
  17. Lee EB, Ong CW, Twohig MP, Lensegrav-Benson T, & Quakenbush-Roberts B (2018). Increasing body image flexibility in a residential eating disorder facility: Correlates with symptom improvement. Eating Disorders, 26, 185–199. 10.1080/10640266.2017.1366229 [DOI] [PubMed] [Google Scholar]
  18. Linardon J (2021). Positive body image, intuitive eating, and self-compassion protect against the onset of the core symptoms of eating disorders: A prospective study. International Journal of Eating Disorders, 54(11), 1967–1977. 10.1002/eat.23623 [DOI] [PubMed] [Google Scholar]
  19. Linardon J, de la Piedad Garcia X, & Brennan L (2016). Predictors, moderators and mediators of treatment outcome following manualised cognitive-behavioural therapy for eating disorders: A systematic review. European Eating Disorders Review, 25, 3–12. 10.1002/erv.2492 [DOI] [PubMed] [Google Scholar]
  20. Linardon J, Messer M, Lisboa J, Newton A, & Fuller-Tyszkiewicz M (2020). Examining the factor structure, sex invariance, and psychometric properties of the Body Image Acceptance and Action Questionnaire and the Functionality Appreciation Scale. Body Image, 34, 1–9. 10.1016/j.bodyim.2020.04.006 [DOI] [PubMed] [Google Scholar]
  21. Linardon J, Rosato J, & Messer M (2020). Break Binge Eating: Reach, engagement, and user profile of an Internet-based psychoeducational and self-help platform for eating disorders. International Journal of Eating Disorders, 53, 1719–1728. 10.1002/eat.23356 [DOI] [PubMed] [Google Scholar]
  22. Linardon J, Shatte A, McClure Z, & Fuller-Tyszkiewicz M (2022). A broad v. focused digital intervention for recurrent binge eating: A randomized controlled non-inferiority trial. Psychological Medicine, 1–12. 10.1017/s0033291722001477 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Linardon J, Tylka TL, & Fuller-Tyszkiewicz M (2021). Intuitive eating and its psychological correlates: A meta-analysis. International Journal of Eating Disorders, 54, 1073–1098. 10.1002/eat.23509 [DOI] [PubMed] [Google Scholar]
  24. Murphy R, Cooper Z, Hollon SD, & Fairburn CG (2009). How do psychological treatments work? Investigating mediators of change. Behaviour Research and Therapy, 47, 1–5 10.1016/j.brat.2008.10.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Pellizzer ML, Waller G, & Wade TD (2018). Body image flexibility: A predictor and moderator of outcome in transdiagnostic outpatient eating disorder treatment. International Journal of Eating Disorders, 51, 368–372. 10.1002/eat.22842 [DOI] [PubMed] [Google Scholar]
  26. Pellizzer ML, Waller G, & Wade TD (2019). Predictors of outcome in cognitive behavioural therapy for eating disorders: An exploratory study. Behaviour Research and Therapy, 116, 61–68. 10.1016/j.brat.2019.02.005 [DOI] [PubMed] [Google Scholar]
  27. Rand-Giovannetti D, Cicero DC, Mond JM, & Latner JD (2020). Psychometric properties of the Eating Disorder Examination–Questionnaire (EDE-Q): A confirmatory factor analysis and assessment of measurement invariance by sex. Assessment, 27(1), 164–177. 10.1177/1073191117738046 [DOI] [PubMed] [Google Scholar]
  28. Sandoz EK, Wilson KG, Merwin RM, & Kellum KK (2013). Assessment of body image flexibility: The Body Image-Acceptance and Action Questionnaire. Journal of Contextual Behavioral Science, 2, 39–48. 10.1016/j.jcbs.2013.03.002 [DOI] [Google Scholar]
  29. Timmons AC, & Preacher KJ (2015). The importance of temporal design: How do measurement intervals affect the accuracy and efficiency of parameter estimates in longitudinal research. Multivariate Behavioral Research, 50, 41–55. 10.1080/00273171.2014.961056 [DOI] [PubMed] [Google Scholar]
  30. Tylka TL, Linardon J, Wood-Barcalow NL, Daníelsdóttir S, & Fuller-Tyszkiewicz M (2022). Short forms of the Body Appreciation Scale-2 (BAS-2SF): Item selection and psychometric evaluation. Body Image, 41, 308–330. 10.1016/j.bodyim.2022.04.001 [DOI] [PubMed] [Google Scholar]
  31. Tylka TL, & Wood-Barcalow NL (2015a). The Body Appreciation Scale-2: Item refinement and psychometric evaluation. Body Image, 12, 53–67. 10.1016/j.bodyim.2014.09.006 [DOI] [PubMed] [Google Scholar]
  32. Tylka TL, & Wood-Barcalow NL (2015b). What is and what is not positive body image? Conceptual foundations and construct definition. Body Image, 14, 118–129. 10.1016/j.bodyim.2015.04.001 [DOI] [PubMed] [Google Scholar]
  33. Webb JB, Wood-Barcalow NL, & Tylka TL (2015). Assessing positive body image: Contemporary approaches and future directions. Body Image, 14, 130–145. 10.1016/j.bodyim.2015.03.010 [DOI] [PubMed] [Google Scholar]
  34. Williams SE, Watts TKO, & Wade TD (2012). A review of the definitions of outcome used in the treatment of bulimia nervosa. Clinical Psychology Review, 32, 292–300. 10.1016/j.cpr.2012.01.006 [DOI] [PubMed] [Google Scholar]
  35. Linardon J, McClure Z, Tylka TL, & Fuller-Tyszkiewicz M (2022). Body appreciation and its psychological correlates: A systematic review and meta-analysis. Body Image, 42, 287–296. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Material
NIHMS1872241-supplement-Supplementary_Material.docx (35.8KB, docx)

RESOURCES