Eating disorders comprise a range of clinical presentations defined by disordered eating and weight control behaviours, often involving significant disturbances in body image (American Psychiatric Association, 2013). These disorders are serious psychological problems that commonly persist for many years (Fichter & Quadflieg, 2016) and are linked to high levels of psychological distress, reduced quality of life, and severe health risks (Fisher et al., 1995; Mitchison et al., 2015). Eating disorders affect an estimated 5.7% of adult women and 2.2% of adult men at any point in time (Galmiche et al., 2019), with recent research suggesting particularly high prevalence rates for the full spectrum of clinical and subthreshold eating disorders in adolescents (i.e., eating pathology; 32.9% of adolescent girls, 12.8% of adolescent boys; Mitchison et al., 2020). While numerous risk factors have been associated with the development and maintenance of eating psychopathology (see Pennesi & Wade, 2016, for a review), the current review focuses on the role of emotion dysregulation in the onset and maintenance of eating disorders. Emotion dysregulation is an important issue across the lifespan and there has been considerable research in adult populations implicating emotion dysregulation in psychopathology. However, as will be discussed throughout the review, the role of emotion dysregulation in eating pathology may be especially relevant during adolescence given that it is a peak period of both emotional turmoil and eating disorder onset (Nagl et al., 2016; Rapee et al., 2019). This review focuses on research across the lifespan, and proposes emotion dysregulation as a maintenance factor, and potentially a risk factor, for eating disorders, with a particular focus on adolescence as a key developmental period.

There has been an increasing interest in clarifying the relationship between emotion dysregulation and eating pathology, with several recent systematic reviews and meta-analyses emerging. These reviews have highlighted that emotion dysregulation is linked with eating pathology—both at a symptom and disorder level (Prefit et al., 2019)—that emotion dysregulation is heightened in individuals with anorexia nervosa (Oldershaw et al., 2015), and that both anorexia nervosa and bulimia nervosa are characterised by emotion dysregulation (Lavender et al., 2015). However, the theoretical and clinical implications, especially as they apply to the widely applied transdiagnostic model for eating disorders ("CBT-E model"; Fairburn et al., 2003), have not yet been explored. Further, no review to date has employed a developmental focus, with few of the aforementioned reviews including research on adolescents.

Thus, the current review aims to link existing research on the relationship between emotion dysregulation and eating pathology with the CBT-E model to provide potential modifications to the existing framework in line with empirical evidence. To highlight the role of emotion dysregulation in eating pathology, this review first provides an introduction to the concept of emotion dysregulation and its potential role in eating disorder aetiology, followed by a synthesis of the research evidence linking emotion dysregulation with specific eating disorder symptoms. Finally, this review will discuss the implications of this body of research for the CBT-E model. Specifically, we propose a potential interaction between emotion dysregulation and weight/shape concerns that may contribute to the maintenance, and potentially onset, of eating disorders. We close by discussing potential implications for treatment, and areas where future research is needed to further clarify the role of emotion dysregulation in eating disorders.

The CBT-E Model

The CBT-E model of eating disorders posits overvaluation of weight/shape and eating as the central underlying mechanism shared by all eating pathologies, which maintains disordered eating behaviours (Fairburn et al., 2003). The theoretical model and an associated treatment approach (enhanced cognitive behavioural therapy; CBT-E) have received considerable support empirically in both adolescents (Jones et al., 2020) and adults (Byrne et al., 2011; Fairburn et al., 2015; Hoiles et al., 2012). In addition to specific features of eating disorders (e.g., dietary restraint, over-evaluation of weight/shape), this model also includes a number of maintaining factors (perfectionism, low self-esteem, marked interpersonal difficulties, and mood intolerance).

Notably, one of the maintenance factors is mood intolerance, which is an important aspect of emotion dysregulation. The CBT-E model defines mood intolerance as “an inability to cope appropriately with certain emotional states” (Fairburn et al., 2003, p. 157). The model therefore explicitly includes the regulatory processes of emotion dysregulation.

Mood intolerance was added to the model based primarily on research on binge eating, suggesting that binge eating may be used as an attempt to modulate mood (Kenardy et al., 1996; Lee et al., 2019). This body of research suggests that mood intolerance might lead to subsequent binge eating, whereby binge eating occurs as an attempt to regulate one’s emotions (Macht, 2008). Moreover, the CBT-E model (Fairburn et al., 2003) proposes that purging behaviour serves a similar emotional regulation purpose. Although, little research has investigated the association between mood intolerance and purging, especially in comparison to similar research on binge eating (e.g., Dakanalis et al., 2015; Lampard et al., 2013), the few studies that do exist support the proposed association between mood intolerance and purging among both clinical (Lavender et al., 2015; Pisetsky et al., 2017) and community samples (Dakanalis et al., 2014).

While the CBT-E model (Fairburn et al., 2003) offers a comprehensive maintenance model of eating psychopathology, that takes into account an important facet of emotion dysregulation, this review aims to outline that the current model does not account for a prominent role of emotion dysregulation in eating psychopathology. Firstly, it limits the scope of emotion dysregulation by focusing solely on mood intolerance (“an inability to cope appropriately with certain emotional states”; Fairburn et al., 2003, p. 157). As will be outlined below, emotion dysregulation encompasses a wide range of difficulties with emotions, ranging from emotion recognition to using regulatory strategies (Gratz & Tull, 2010; Thompson & Calkins, 1996), which are not adequately captured in the current model. Secondly, we will review evidence to suggest that emotion dysregulation is linked to all aspects of disordered eating (dietary restraint, binge eating, purging, and excessive exercise), suggesting that additional paths are needed. To support this further, we will first provide a brief background on emotion dysregulation and it’s development before summarising current empirical support for the association between emotion dysregulation and eating disorder behaviours to discuss the implications these findings have on the current conceptualisation of the CBT-E model.

Emotion Dysregulation within Psychopathology

Emotion Dysregulation Theory and Development

Emotion dysregulation refers to the failure to employ effective emotion regulation strategies and involves a range of complex cognitive and behavioural components (Gratz & Roemer, 2004; Keenan, 2000). Similarly, emotion regulation broadly refers to the ability to deal with specific emotions, both positive and negative, through the use of specific strategies (Gross, 1998). Emotional responses are thought to occur when external events trigger an ‘emotion program’ leading to the activation of ‘response tendencies’ (changes in behaviours, physiology, or subjective experience), before being expressed as observable emotional responses (see Gross & Muñoz, 1995 for review). Theories of emotion regulation highlight the use of both intrapersonal regulation mechanisms (e.g., cognitive change; Gross, 1998) and interpersonal regulation mechanisms (e.g., use of a safety person; Hofmann, 2014). However, to successfully implement any of these emotion regulation strategies, individuals must first be able to monitor and understand their emotional experience (Thompson & Calkins, 1996). Thus, the concept of emotion dysregulation goes beyond an inability to employ effective emotion regulation strategies, also capturing difficulties with recognising and accepting one’s emotions (Gratz & Roemer, 2004). Specifically, Gratz and Roemer (2004) propose that emotion dysregulation includes six distinct but related dimensions: lack of awareness of emotional responses, lack of clarity of emotional responses, non-acceptance of emotional responses, limited access to emotion regulation strategies, difficulties controlling impulses when experiencing negative emotions, and difficulties engaging in goal-directed behaviours when experiencing negative emotions. While theories of emotion regulation focus mainly on the type of strategies employed, which can be either adaptive or maladaptive, emotion dysregulation refers to broader difficulties experienced with emotion regulation, placing greater emphasis on functionality rather than the use of specific cognitive or behavioural processes.

Emotion regulation skills are developed throughout childhood through observation, parenting practices, and family emotional climate (Morris et al., 2017). Research has shown that maladaptive emotional parenting behaviours, parental emotion dysregulation, and insecure attachment are prospectively associated with child emotion dysregulation (Boldt et al., 2020; Breaux et al., 2018; Brumariu & Kerns, 2013; Halligan et al., 2013). These skills are first acquired during early childhood, around 3–4 years of age (Dennis & Kelemen, 2009). However, problems with emotion dysregulation can emerge at any time during development and are often associated with psychopathology. For example, during early-mid childhood emotion dysregulation is commonly associated with externalising disorders, such as oppositional defiant disorder (Cavanagh et al., 2014; G. M. Mitchison, Mond, et al., 2019; Mitchison, Liber, et al., 2019). During adolescence, which is commonly considered the main period of onset for “social-emotional disorders” (Rapee et al., 2019), emotion dysregulation is associated with anxiety and mood disorders (Abravanel & Sinha, 2015; Folk et al., 2014; McLaughlin et al., 2011). Furthermore, emotion dysregulation is frequently proposed as a core component in the development as well as the maintenance of these disorders (e.g., Hofmann et al., 2012; Mennin et al., 2002). Indeed, research has suggested that emotion dysregulation is linked with both externalising and internalising problems among children and adolescents (Winsper et al., 2020; Zeman et al., 2002). Moreover, longitudinal research has shown that emotion dysregulation predicts later externalising problems (Halligan et al., 2013; Kim & Cicchetti, 2010), internalising problems (Schneider et al., 2020), as well as overall psychopathology (Weissman et al., 2019), positing emotion dysregulation as a risk factor for the development of psychopathology as well as a correlate. Taken together, the literature suggests that emotion dysregulation is a transdiagnostic risk factor for a range of psychopathologies (Aldao et al., 2016), that when combined with more disorder-specific risk factors (e.g., positive beliefs about worry as a disorder-specific risk factor for generalised anxiety symptoms; Shihata et al., 2017) might lead to the development of these pathologies. We turn now to consider the role of emotion dysregulation in eating pathology specifically and discuss the potential interplay of emotion dysregulation with weight/shape concerns.

Emotion Dysregulation in Eating Disorders

Emotion dysregulation has been identified as an important transdiagnostic correlate of eating disorders (Aldao et al., 2010; Oldershaw et al., 2015; Prefit et al., 2019), with evidence of greater emotion dysregulation compared to healthy participants in those with a diagnosis of anorexia nervosa (see Oldershaw et al., 2015 for review), bulimia nervosa (see Lavender et al., 2015 for review), and binge eating disorder (see Leehr et al., 2015 for review). Supporting this, a recent meta-analysis found that a wide variety of emotion regulation strategies were associated with disordered eating and eating pathology across eating disorder diagnostic groups (Prefit et al., 2019), including important components of emotion dysregulation such as lack of awareness and lack of clarity, which are important components of emotion dysregulation (Gratz & Roemer, 2004). Additionally, a recent qualitative meta-synthesis reported that individuals with eating disorders frequently report experiencing emotions as overwhelming, struggling to manage their emotions, and using their eating disorder to manage and control their emotions (Henderson et al., 2019). However, no review to date has looked at emotion dysregulation with regard to eating disorder onset specifically or has extended this discussion to adolescents. As noted by Prefit et al. (2019), due to the clinical focus of their meta-analysis, which addressed the link between emotion regulation strategies and eating disorder symptoms, most of the included studies were conducted with adult women. Similar limitations were reported for the qualitative meta-synthesis of qualitative studies, which examined emotions in eating disorders by Henderson et al. (2019), where only one of the 16 included studies included adolescents.

It is important to consider how these findings may apply to adolescents as well as adults, as eating disorders typically develop and are most prevalent during late adolescence (Hudson et al., 2007; Nagl et al., 2016), making this a critical period for early intervention. Further, there is emerging evidence that emotion dysregulation is associated with eating pathology across genders (Lavender & Anderson, 2010), and plays a role in related body image disorders that are more common in male samples (e.g., muscle dysmorphia), including adolescent boys (Cafri et al., 2006). However, much of the existing research has focused primarily on women and further research is needed to understand these associations in men, and during adolescence.

While there is converging evidence to suggest that emotion dysregulation is associated with eating pathology among adolescents (McLaughlin et al., 2011; Weinbach et al., 2018), there is emerging evidence to suggest that emotion dysregulation may contribute to eating disorder onset as well. As such, examining the role of emotion dysregulation in eating disorders among adolescents is vital for hypothesising about potential developmental trajectories and time-ordering of emotion dysregulation and eating psychopathology and their nexus. While children start to develop emotion regulation skills from an early age (see Morris et al., 2007 for a review), eating disorders typically do not develop until mid-to-late adolescence (Hudson et al., 2007), indicating that emotion dysregulation may precede the onset of eating disorders and positioning it temporally as a risk factor. Moreover, emotions are experienced more intensely during adolescence compared to other stages of development (Bailen et al., 2018), making adolescents more vulnerable to emotion dysregulation issues. However, few studies have examined longitudinally whether emotion dysregulation is associated with later eating disorder onset.

Of the few longitudinal studies, the strongest evidence for the role of emotion dysregulation in the onset of eating disorder symptoms comes from a longitudinal study of adolescents by McLaughlin et al. (2011), which found that emotion dysregulation predicted subsequent eating disorder symptoms after controlling for baseline levels of eating disorder symptoms, as well as other presenting psychopathology. This finding indicates that emotion dysregulation uniquely predicted the development of eating disorder symptoms in adolescence over and above the high rates of comorbidity with other disorders, such as depression and anxiety (Hofmann et al., 2012; Mennin et al., 2002). Similar findings have been reported when examining specific eating disorder symptoms. Shriver et al. (2019) found that high levels of emotion regulation at age 15 predicted lower dietary restraint at age 16 in a community sample of adolescents. Similarly, Goodwin et al. (2014) found that emotion dysregulation predicted increases in compulsive exercise one year later among both adolescent boys and adolescent girls in a community sample. Considering adolescent girls specifically, Nolen-Hoeksema et al. (2007) found that maladaptive emotion regulation (i.e., rumination) predicted increases in bulimic symptoms four years. Thus, there is evidence to suggest that emotion dysregulation may predict eating disorder symptoms in community samples of adolescents, implicating it not only as a maintenance factor but also as a potential risk factor.

This is supported by cross-sectional community-based and treatment research. Sim and Zeman (2006) found that poor emotion regulation skills had a significant association with both higher levels of disordered eating and higher body dissatisfaction among adolescent girls in a community sample; albeit stronger with disordered eating. Further, a study among Swedish adolescents by Hansson et al. (2017) found a significant relationship between emotion dysregulation and concurrent disordered eating for adolescent girls, but not boys. Adolescents in treatment for an eating disorder have also been found to experience significantly higher levels of emotion dysregulation compared to normative samples (Anderson et al., 2018; Weinbach et al., 2018), although a limitation of this line of research is that the lack of representation of adolescent boys with eating disorders given low rates of males in clinical settings (Bohrer et al., 2017; Coffino et al., 2019; Forrest et al., 2017). Taken together with the mixed findings regarding gender in the aforementioned community samples (Hansson et al., 2017; McLaughlin et al., 2011), further research is needed to better understand these associations in boys. Overall, the extant literature indicates that emotion dysregulation is significantly associated with eating disorders among adolescents, both in clinical and non-clinical populations.

The potential role of emotion dysregulation in the onset of eating disorders is further indicated through its association with attachment security. Attachment insecurity has been frequently linked with eating disorders (O’shaughnessy & Dallos, 2009; Ward et al., 2000), as well as emotion dysregulation (Colle & Del Giudice, 2011). As outlined previously, one of the key factors in developing healthy emotion regulation during childhood is a secure attachment with caregivers (Morris et al., 2007) and research has consistently shown an indirect effect of attachment insecurity on eating pathology through emotion dysregulation (see Cortés-García et al., 2019 for review). Thus, there is evidence to suggest that emotion dysregulation may account for the relationship between attachment security and eating pathology, further implicating it as a potential risk factor in the development of eating disorders during adolescence.

Additionally, there is emerging experimental data linking emotion dysregulation with eating disorder symptoms supporting the notion that emotion dysregulation may be a maintenance factor of eating disorders. In a study by Hilbert et al. (2011), adult women with bulimia nervosa or binge eating disorder showed greater reactivity to sadness compared to healthy controls in a stress exposure paradigm. Similarly, a recent study by Monteleone et al. (2020) found that women with anorexia nervosa or bulimia nervosa reported heightened anxiety and heightened body dissatisfaction compared to healthy controls following a stress induction. Taken together, these studies suggest that women with eating disorders experience heightened emotional response to acute stressors, supporting the larger base of correlational research indicating that individuals with eating disorders report emotion regulation difficulties when faced with emotional stressors (Lavender et al., 2015). Furthermore, Wildes et al. (2012) showed that adults with anorexia nervosa who were randomized to a negative mood induction showed significantly greater increases in eating disorder cognitions as well as urges to engage in disordered eating behaviours compared to those in the neutral mood induction. Moreover, Fitzpatrick et al. (2019) showed that emotion regulation strategies, such as mindfulness and cognitive reappraisal, significantly reduced body dissatisfaction (but not eating disorder urges), during a negative mood induction among adults with eating disorders (i.e., anorexia nervosa, bulimia nervosa, or other specified feeding or eating disorder). Experimental research also indicates that women with anorexia nervosa or bulimia nervosa are more likely to employ maladaptive emotion regulation strategies compared to healthy controls when uninstructed (Naumann et al., 2016), indicating that individuals with eating disorders may experience problems employing effective emotion regulation strategies. Taken together, these experimental studies show that adults with eating disorders, mainly women, experience greater emotion reactivity, reported heightened eating disorder symptoms when distressed, and show a reduction in symptoms when using adaptive emotion regulations strategies, supporting the notion that emotion dysregulation may be a maintenance factor within eating pathology. However, no such research has included adolescents and few studies include men.

Emotion Dysregulation and Body Image Concerns

Given that adolescence is the peak period of onset for eating disorders (Hudson et al., 2007; Nagl et al., 2016), it is notable that the major eating disorder-specific risk factor, weight/shape concerns, is also a frequent problem during this period of development (Bartholdy et al., 2017; Bucchianeri et al., 2013). Research among adults and young males has found that emotion dysregulation is associated with heightened weight/shape concerns (Lavender & Anderson, 2010; Racine & Wildes, 2013; Rufino et al., 2018). Other research among young adult males has further shown that emotion dysregulation is associated with dysmorphic appearance concerns (Cunningham et al., 2018) and muscle dysmorphia symptoms (Cunningham et al., 2020), indicating that emotion dysregulation is linked to weight/shape concerns that centre on both thinness and muscularity.

While few studies have examined this relationship in adolescents, one such study in adolescent girls by Sim and Zeman (2006) found that poor emotional awareness, a dimension of emotion dysregulation, was associated with body dissatisfaction. Notably, the study found that this association was weaker than the association between emotional awareness and disordered eating. Furthermore, once other risk factors (e.g., negative affect), were accounted for, emotional awareness did not remain significantly associated with body dissatisfaction but did remain significantly associated with disordered eating. Similarly, recent research on the CBT-E model in adolescents with eating disorders found that mood intolerance was associated with both disordered eating and weight/shape concerns (Jones et al., 2020). These results are preliminary indicators to suggest that the relationship between emotion dysregulation and weight/shape concerns may be more complex than previously thought. Specifically, there appears to be a stronger relationship between emotion dysregulation and eating disorder behaviours then between emotion dysregulation and weight/shape concerns. As has been proposed and demonstrated in the development of other forms of psychopathology in adolescence, emotion dysregulation often interacts with other more disorder-specific and proximal risk factors to potentiate the development of a disorder. Thus, it is conceivable that emotion dysregulation could interact with weight/shape concerns, leading to the development of eating disorder behaviours, particularly during adolescence. Dimensional frameworks of psychopathology suggest that symptoms manifest based on a combination of both transdiagnostic risk factors and narrower syndrome-specific moderators (Forbes et al., 2016; Nolen-Hoeksema & Watkins, 2011). As such, emotion dysregulation could be considered a transdiagnostic risk factor for general psychopathology that may interact with weight and shape concerns specifically to predict disordered eating symptoms. Such an interactive effect has been found among adult women when examining the transdiagnostic factor of negative urgency (a concept closely related to emotion dysregulation; Juarascio et al., 2020) and disordered eating, whereby the association between negative urgency and disordered eating was stronger at high levels of body dissatisfaction (Racine & Martin, 2017; Racine et al., 2017). Similarly, a recent study by Liebman et al. (2020) found that high behavioural inhibition (the tendency to withdraw when faced with negative emotions) strengthened the relationship between weight/shape concerns and purging among adolescent girls and young adult women. However, these studies have not examined these relationships longitudinally, and no evidence to date has examined a similar interactive link for the broader concept of emotion dysregulation. We suggest that this would be an important endeavour for future research, especially among cohorts of adolescents, to consider how the two risk/maintenance factors of weight/shape concerns and emotion dysregulation interact to predict the onset and maintenance of eating disorders.

Summary

Overall, the emerging literature reviewed above suggests that emotion dysregulation, which is already recognised as a transdiagnostic factor in psychopathology broadly, may also extend to be an important maintenance factor, and potential risk factor, for eating disorders. Importantly, research suggests that emotion dysregulation may potentially interact with the specific risk factor of weight/shape concerns, especially among adolescents. To fully understand the relationship between emotion dysregulation and eating pathology it is crucial to examine this relationship at a symptom level, to better understand how emotion dysregulation fits in with the CBT-E model. This is particularly important given the shared symptomology (e.g., binge eating can occur across anorexia nervosa, bulimia nervosa, and binge eating disorder) and high level of diagnostic crossover between discrete eating disorder diagnoses (e.g., 25–45% of people with anorexia nervosa, bulimia nervosa or binge eating disorder crossover into another eating disorder diagnosis during their illness; Stice et al., 2009; Tozzi et al., 2005). To further investigate the link between emotion dysregulation and eating disorders, we now turn to review the empirical evidence regarding the relationship between emotion dysregulation and the core eating disorder behaviours examined in the CBT-E model (dietary restraint, driven exercise, binge eating, and purging). As will be seen below, most of the research linking specific eating disorder symptoms and emotion dysregulation has been conducted with adults. This line of research, however, offers important insights into the role of emotion dysregulation eating pathology during adolescence, which we will highlight throughout the review.

Empirical Support for the Role of Emotion Dysregulation in Specific Eating Disorder Behaviours

The Association Between Emotion Dysregulation and Binge Eating

The relationship between emotion dysregulation and binge eating has received considerable attention in previous research. Notably, one of the early theories of binge eating proposes an affect regulation model (Polivy & Herman, 1993), whereby individuals engage in binge eating to regulate negative emotions. In suppport of this theory, numerous studies have reported a positive association between emotion dysregulation and binge eating both in clinical (Kenny et al., 2017; Lampard et al., 2013; Monell et al., 2018) and community samples (Burton & Abbott, 2019; Dakanalis et al., 2014; Eichen et al., 2017; Whiteside et al., 2007). Importantly, this association between emotion dysregulation and binge eating has been found among both women (Eichen et al., 2017) and men (Dakanalis et al., 2014; Kukk & Akkermann, 2019). Furthermore, research by Berg et al. (2013) in adult women with bulimia nervosa found that negative affect increased prior to binge eating and decreased after episodes of binge eating, providing support for the affect regulation model. Similarly, Svaldi et al. (2019) showed that attempts at emotion regulation drastically reduced the probability of a binge episode among adults with binge eating disorder. However, a meta-analysis of ecological momentary assessment studies examining the affect regulation model found contrary evidence, as participants reported levels of negative affect increased following binge eating episodes compared to prior to their binge eating episodes (Haedt-Matt & Keel, 2011). Similarly, a study by Pisetsky et al. (2017) found no association between emotion dysregulation and binge eating among adult women with eating disorders.

Mixed evidence is also found in experimental research. In support of the association between emotion dysregulation and binge eating, Rosenberg et al. (2013) found that adults with binge eating disorder reported more sweet craving and heightened desire to binge following a stress induction compared to controls. However, similar findings were not reported in a similar study among adolescent girls (Laessle & Schulz, 2013).

Preliminary work has also found evidence linking emotion dysregulation and binge eating in adolescents. For example, in a study of male and female adolescents Laghi et al. (2018) found that emotion dysregulation was significantly associated with binge eating. Similarly, in one of the few longitudinal studies conducted, Goldschmidt et al. (2017) found that among adolescent girls, emotion dysregulation was associated with concurrent binge eating, as well as subsequent binge eating. Findings showed that emotion dysregulation at age 17 predicted binge eating at age 18. However, as mentioned above, Laessle and Schulz (2013) found no effect of negative mood on eating behaviours among adolescent girls in an experimental setting.

Taken together, this body of research suggests that emotion dysregulation is associated with binge eating both at clinical and sub-clinical levels. This supports previous research showing elevated levels of emotion dysregulation among individuals with eating disorders that are characterised by binge eating (i.e., bulimia nervosa and binge eating disorder; Brockmeyer et al., 2014; Svaldi et al., 2012).

The Association Between Emotion Dysregulation and Dietary Restraint

The relationship between emotion dysregulation and dietary restraint has been primarily investigated in women with anorexia nervosa—an eating disorder that is characterized by significant dietary restriction relative to needs (American Psychiatric Association, 2013). Theoretical models of anorexia nervosa, such as the emotion avoidance model (Wildes et al., 2010) suggest that individuals with anorexia nervosa engage in restrictive eating to prevent emotional experiences. However, such models are constrained by their focus on anorexia nervosa specifically, despite the presence of dietary restraint across other eating disorders. Recent research has indicated that individuals with anorexia nervosa may use dietary restraint as an emotion regulation strategy to deal with negative emotions (Meule et al., 2019), as negative affect has been found to precede dietary restriction (Engel et al., 2013). This is in line with qualitative research, wherein adult women with a diagnosis of anorexia nervosa have reported using dietary restraint, among other disordered eating behaviours, to regulate their emotions (Espeset et al., 2012).

In addition to the body of research among women with anorexia nervosa, in a sample of women with a range of eating disorders (anorexia nervosa, bulimia nervosa, and binge eating disorder), Svaldi et al. (2012) found that emotion dysregulation was positively associated with dietary restraint across disorder categories. Furthermore, research investigating the relationship between emotion dysregulation and dietary restraint among community samples found that emotion dysregulation is associated with restrictive eating in a transdiagnostic framework (Burton & Abbott, 2019; Dakanalis et al., 2014). In a community sample study among young adults Burton and Abbott (2019) found that emotion dysregulation was significantly associated with dietary restraint. Further, a study by Dakanalis, et al. (2014) examined the proposed associations within the CBT-E model of eating disorders (Fairburn et al., 2003) among adult men, who had been recruited in a community setting. Findings suggested an additional significant path from mood intolerance (which in this study was assessed using a measure of emotion dysregulation) to dietary restraint, providing support for the notion that emotion dysregulation and strict dieting (a form of dietary restraint) are associated among men as well as women.

While there has been limited research examining the relationship between emotion dysregulation and dietary restraint among adolescents, there is evidence to suggest that similar patterns may emerge. For example, in line with findings among adults (Engel et al., 2013), negative affect has been found to precede dietary restraint among adolescents with anorexia nervosa (Pila et al., 2019). Simiarly, in a clinical sample of adolescents Jones et al. (2020) found mood intolerance to be linked with dietary restraint directly. Furthermore, there is limited experimental research examining the link between emotion dysregulation and dietary restraint. While there is research suggesting that adults with anorexia nervosa experience heightened eating disorder symptoms after a negative mood induction (Wildes et al., 2012), no research has looked at dietary restraint in particular.

Taken together, this body of research suggests that emotion dysregulation is associated with dietary both at clinical and sub-clinical levels. This supports previous research showing elevated levels of emotion dysregulation among individuals with eating disorders that are characterised by dietary restraint (i.e., anorexia nervosa; Brockmeyer et al., 2014; Svaldi et al., 2012). However, more experimental research is needed to further explore this relationship.

The Association Between Emotion Dysregulation and Purging

While not as commonly examined as binge eating, several studies have also found an association between emotion dysregulation and purging in cross-sectional studies among both clinical (Lavender et al., 2014; Pisetsky et al., 2017) and community samples (Burns et al., 2012; Dakanalis et al., 2014). These studies indicate that emotion dysregulation is linked with more frequent purging. However, to the best of our knowledge, no studies to date have examined this relationship among adolescents and there is no longitudinal research examining the direction of effects. Furthermore, there is limited experimental research examining the link between emotion dysregulation and purging.

The Association Between Emotion Dysregulation and Driven Exercise

Research examining the relationship between emotion dysregulation and driven exercise is scarce and conflicting. One study by Lavender and colleagues (2014) examined the relationship between emotion dysregulation and eating disorder symptoms among (mainly female) adults seeking treatment for bulimia nervosa. Results revealed that one particular dimension of emotion dysregulation – greater difficulty with goal-directed behaviour in the context of distress—was significantly related to greater driven exercise frequency, but the other dimensions of emotion dysregulation were not associated with driven exercise. This was extended in a study by Monell et al. (2018), who utilised a clinical database of adults with various eating disorders to examine the relationship between emotion dysregulation and eating pathology, and showed difficulty with goal-directed behaviour as well as difficulty remaining in control of one's behaviour in the context of distress were both associated with driven exercise. By contrast, Pisetsky et al. (2017) examined emotion dysregulation among adults in treatment for an eating disorder (mainly women) and found that driven exercise was not associated with emotion dysregulation.

Only one study to date has examined this relationship among adolescents, providing support for the notion that emotion dysregulation is linked with driven exercise in this age group (Goodwin et al., 2014). The study found that among high school students, emotion dysregulation was significantly associated with concurrent driven exercise for both boys and girls. Furthermore, there is limited experimental research examining the link between emotion dysregulation and driven exercise both among adults and adolescents.

Summary

Overall, the reviewed research suggests that emotion dysregulation is associated with eating disorders at both the symptom- and diagnosis-level, highlighting its importance in the development of eating pathology, for both adolescents and adults. Notably, this line of research supports the regulatory theory of eating disorders, most commonly associated with binge eating (Polivy & Herman, 1993), whereby individuals engage in eating disorder behaviours to regulate negative emotions (Foye et al., 2019). However, as outlined above, much of the existing research focuses on adults rather than adolescents. While others have argued that eating disorder aetiology is likely to be similar in adolescents and adults (Christian et al., 2019), more research among adolescents is needed to investigate how emotion dysregulation relates to eating disorder symptoms at this critical stage of development. Furthermore, there is limited longitudinal research providing information on the temporal relationship between emotion dysregulation and eating disorder symptoms. In particular, with few exceptions (e.g., Racine & Wildes, 2015), most studies have not considered the potential bi-directional relationship between emotion dysregulation and eating pathology. As such, most cross-sectional studies presume that emotion dysregulation would have a causal effect on eating disorder symptoms but neglect to consider whether eating disorder symptoms might also have a causal effect on emotion dysregulation. Taken together, the link between emotion dysregulation and specific behavioural symptoms has important implications for both the CBT-E model as well as clinical practice, which we now turn to discuss.

Implications

The findings on emotion dysregulation reviewed above have implications for conceptualising the onset and maintenance of eating disorders as outlined in the CBT-E model, which may be particularly relevant for adolescents. Efforts to incorporate emotion dysregulation into theoretical models of eating disorders are not new, with multiple theories proposing that emotion dysregulation plays a key role in the aetiology of anorexia nervosa (e.g., Haynos & Fruzzetti, 2011), bulimia nervosa (e.g., Wonderlich et al., 2008), and binge eating disorder (e.g., Leehr et al., 2015), for example. However, while these models highlight the role of emotion dysregulation in the development of specific eating disorder diagnoses, this specificity understates the apparent transdiagnostic nature of the relationships. As aforementioned, emotion dysregulation is transdiagnostic across many mental disorders (Aldao et al., 2016). It is thus not surprising that it also appears to be a transdiagnostic feature across eating disorders, both in adolescents and adults. Further, eating pathology is largely transdiagnostic, and there is a known high rate of transition between specific diagnoses (Stice et al., 2009; Tozzi et al., 2005). As outlined in "The CBT-E model" section, the CBT-E model of eating disorders (Fairburn et al., 2003) is very widely used and promoted (Byrne et al., 2011; Fairburn et al., 2015; Hay et al., 2014), and offers a valuable opportunity to conceptualise the central role of emotion dysregulation in the development of eating pathology in a transdiagnostic framework, including among adolescents. Based on the reviewed evidence, one aspect of the model, in particular, could be broadened to reflect the current state of the literature: the concept of mood intolerance.

The CBT-E Model of Eating Disorders

As aforementioned, the CBT-E model of eating disorders provides a theoretical framework for eating disorder maintenance, which posits overvaluation of weight/shape as the central risk factor and underlying mechanism shared by all eating disorder pathologies (Fairburn et al., 2003). The maintenance factor ‘mood intolerance’ in the model is bidirectionally linked to both binge eating and purging, which was based on the evidence available at the time the model was developed. Thus, there is an explicit link between regulatory processes of emotion dysregulation and eating pathology proposed in this dominant model. However, given evidence that has emerged since the publication of the model – summarised in this review – it seems that the full extent of the relationship between emotion dysregulation and the key symptoms of eating disorders may not be captured in this model.

While the CBT-E model defines mood intolerance as “an inability to cope appropriately with certain emotional states” (Fairburn et al., 2003, p.157), emotion dysregulation encompasses a wide range of difficulties with emotions, from impaired recognition of emotions to the use of maladaptive regulatory strategies (Gratz & Roemer, 2004). Thus, while some dimensions of emotion dysregulation (i.e., difficulties with accepting emotions) are integrated, other dimensions (e.g., difficulties with clarity of emotions or engaging in goal-directed behaviour) are not captured in the current CBT-E model. This is problematic as it does not encompass the full scope of the relationship between emotion dysregulation and eating pathology, which may have important implications both in terms of understanding and treating eating disorders.

Given the evidence summarised above in the "Emotion dysregulation in eating disorders" section demonstrating a reliable relationship between eating disorders at the clinical level and emotion dysregulation, a replacement of “mood intolerance” with “emotion dysregulation” may be an update to the model worthy of consideration. Furthermore, the previously reviewed empirical evidence in the "The association between emotion dysregulation and binge eating" to "The association between emotion dysregulation and driven exercise" sections suggests that emotion dysregulation is linked to the full range of specific eating disorder behaviours, as well as the full spectrum of eating disorders (Brockmeyer et al., 2014; Svaldi et al., 2012). There is especially robust evidence for a link with dietary restraint and emerging evidence for the same with excessive exercise, which indicates that adding pathways to these behaviours (i.e., in addition to the current pathways linking mood intolerance to binge eating and purging) could be considered, see Fig. 1.

Fig. 1
figure 1

Proposed changes to the transdiagnostic model

Full size image

We also propose that emotion dysregulation may interact with weight/shape concerns as a specific path to the development and maintenance of eating disorder behaviours, particularly during adolescence, in line with the research outlined in the "Emotion dysregulation in eating disorders" section. In addition to clarifying the relationships within the CBT-E model, this conceptualisation would also be able to account for both the high levels of weight/shape concerns experienced in the general population compared to the smaller rate of diagnosable eating disorder (Fallon et al., 2014), as well as the higher proportion of adolescents who experience emotion dysregulation compared to those with eating disorders. Furthermore, this conceptualisation would have important implications for treatment approaches, which will be discussed below. As there is currently no sufficient empirical evidence to support this adaptation, this proposed adaptation is presented to stimulate further research into the mechanisms linking emotion dysregulation with eating pathology.

Treatment Approaches

One of the key implications of this review is the impact on treatments for eating disorders. Eating disorders typically run protracted courses (Fichter & Quadflieg, 2016) and some individuals with eating disorders benefit greatly from early intervention (Treasure & Russell, 2011). However, evidence-based treatments for eating disorders have varying levels of success in achieving remission (around 38% of adolescents with anorexia nervosa and 40% with bulimia nervosa; Lock & Le Grange, 2019) with many researchers and clinicians searching to improve outcomes (Murray, 2019). The current review suggests that emotion dysregulation is a core factor in eating pathology. Thus, clinical practice might benefit from targeting emotion dysregulation in the treatment of eating disorders. Treatment approaches that target emotion dysregulation, such as dialectic behavioural therapy (DBT) and other mindfulness-based therapies have been found to be effective treatments for eating disorders (Brown, Cusack, Anderson, et al., 2019a, 2019b; Turgon et al., 2019). Similarly, emotion regulation skills training has shown promise as a successful adjunct to regular eating disorder treatment (Holmqvist Larsson et al., 2020). Moreover, for adolescents integrating DBT with empirically-supported treatment approaches for adolescents (e.g., family-based treatment; FBT) has been effective in treating adolescents with eating disorders (Johnston et al., 2015; Murray et al., 2015; C. B. Peterson et al., 2017; C. M. Peterson et al., 2020).

The focus on emotion dysregulation in treatment of eating disorders may be particularly relevant for the treatment of anorexia nervosa, restrictive subtype – a particularly severe eating disorder characterised by self-starvation and often requiring re-feeding in hospital. The current CBT-E model proposes links between mood intolerance and binge eating as well as purging, and thus provides mood-based treatment interventions for eating disorders characterised by these symptoms in CBT-E. However, the links between mood intolerance and strict dieting and other weight-control behaviour (e.g., driven exercise) currently do not exist in the CBT-E model. As such, CBT-E for the restrictive subtype of anorexia nervosa may neglect targeting emotion dysregulation. This is despite increasing evidence that emotion dysregulation may be a key factor of anorexia nervosa broadly (Haynos & Fruzzetti, 2011) and dietary restraint specifically (Svaldi et al., 2012). Furthermore, if emotion dysregulation and weight/shape concerns interact with one another, as we stipulated above, effective treatments for eating disorders may need to target both mechanisms to successfully address eating pathology. Importantly, previous research has shown that women with anorexia nervosa undergoing weight restoration (the primary goal early in the treatment of anorexia nervosa) showed no improvements in emotion dysregulation (unlike eating disorder symptoms), suggesting that treatment could be improved by specifically targeting emotion dysregulation (Haynos et al., 2014). Further supporting this suggestion is research by Racine and Wildes (2015), which showed that emotion dysregulation predicted changes in eating pathology, supporting the notion that emotion dysregulation could be targeted in treatment to improve outcomes.

Furthermore, given the transdiagnostic nature of emotion dysregulation (Keenan, 2000; Sloan et al., 2017), it might be an effective target for early intervention programs, both for eating disorder prevention and general mental health prevention (see Forbes et al., 2019 for review). A recent meta-analysis on the effectiveness of programs targeting emotion dysregulation (e.g., DBT, mindfulness training) found that these types of programs are effective in reducing eating disorder behaviours among at-risk populations (Linardon et al., 2017). Similarly, experimental studies have shown that using emotion regulation strategies can reduce negative affect as well as eating disorder behaviours (Fitzpatrick et al., 2019; Svaldi et al., 2019), indicating that these strategies could be used to prevent engagement in eating disorder behaviours and provide relief from psychological distress, which typically rises in the early phase of treatment while regular eating and weight is being restored.

Given the growing evidence of intervention programs using emotion dysregulation based interventions among adults with eating disorders (Linardon et al., 2017), and the evidence summarised in this review regarding the role of emotion dysregulation in adolescents with eating disorders, the question remains as to whether these treatments may be similar if not more effective among adolescents. However, at present, the evidence base for such interventions is in its infancy, as outlined in a recent review by Reilly et al. (2020) that found that studies using DBT to treat eating disorders in adolescents were few and limited to small samples. Similarly, while there is increasing evidence among adults suggesting that changes in emotion dysregulation are associated with improvements in eating disorder symptom severity (Brown, Cusack, Berner, et al., 2019a, 2019b; Racine & Wildes, 2015; Sloan et al., 2017), it is unknown whether this extends to adolescents. This represents a crucial area for future research, as early interventions for eating disorders are critical in reducing the long-term impact of these disorders (Treasure & Russell, 2011).

Conclusions

Gaps and Future Directions

In conclusion, there is mounting evidence suggesting links between emotion dysregulation and broad eating disorder pathology, weight/shape concerns, and specific symptoms. However, the majority of studies have relied on cross-sectional data. While there is cumulative evidence suggesting that emotion dysregulation is a core factor of eating pathology, more longitudinal and experimental research would be helpful to better understand the nature of the relationships. Specifically, while the current review focuses on emotion dysregulation within the CBT-E model, primarily a maintenance model of eating pathology, longitudinal research should investigate whether emotion dysregulation is also implicated in the development of eating disorders. As with the original CBT-E model, while our hypothesised amendments focus on a maintenance framework, such a framework could be applied to disorder onset and should be examined empirically.

Secondly, while this review suggests that emotion dysregulation may be particularly important among adolescents who develop eating disorders, more studies are needed to investigate these relationships among adolescents to further enhance our understanding of the association between emotion dysregulation and weight/shape concerns in the onset of eating disorders. Recent research on the CBT-E model indicates that mood intolerance is a particularly important feature among adolescence and is associated with weight/shape concerns (Jones et al., 2020); further research is now needed to examine the temporal nature of this relationship preceding and during the course of eating disorders. Furthermore, examining the relationship between emotion dysregulation and eating pathology at various developmental stages of adolescence may provide further information regarding specific critical developmental periods.

Thirdly, while the current review focuses on emotion dysregulation as a maintenance factor of eating pathology, few studies have examined the bi-directionality in the relationship between emotion dysregulation and eating disorders, with most research examining only the effect of emotion dysregulation on the development or exacerbation of eating disorder symptoms. However, such research does not provide adequate evidence on the transactional processes hypothesised in maintenance models, such as the CBT-E model. Of the research that exists, findings are mixed. Some research among adolescents indicates that eating disorder symptoms are associated with subsequent emotion dysregulation (Goldschmidt et al., 2017; Nolen-Hoeksema et al., 2007), while other studies have found no link between eating disorder symptoms and subsequent emotion dysregulation (McLaughlin et al., 2011). Further, when examining symptom change over time in women with anorexia nervosa, Racine and Wildes (2015) found that while changes in emotion dysregulation predicted changes in eating disorder symptoms, changes in eating disorder symptoms did not predict changes in emotion dysregulation. Thus, further research should examine the interplay between emotion dysregulation and eating disorder symptoms to further understand the underlying transactional processes.

Lastly, as is common in the eating disorder field, much of the research reviewed here has utilised clinical samples of adult women with eating disorders. While this information is valuable in terms of informing treatment approaches, this population might not be representative of all people with eating disorders in the community. Treatment-seeking for eating disorders is very low, especially among adolescents (around 10–25%; Fatt et al., 2019; Forrest et al., 2017) and males, and those seeking treatment often differ from their non-treatment-seeking peers in terms of demographics and eating disorder severity (Fatt et al., 2019; Forrest et al., 2017; Trompeter et al., 2020). It is thus crucial to investigate the relationship between emotion dysregulation and eating pathology in community samples to see whether the results generalise to people who are not seeking treatment, and to individuals with sub-clinical levels of disordered eating/body dissatisfaction.

Conclusion

In conclusion, there is sufficient evidence to suggest that emotion dysregulation plays an important role in the development of transdiagnostic eating pathology, and we propose that this should be recognised in the CBT-E model of eating disorders. Given the association of emotion dysregulation with psychopathology broadly, we also propose that weight/shape concerns may act as a moderator that combines with emotion dysregulation to predict eating pathology specifically. Future work in this area will strengthen our understanding of the nature of these relationships and could ultimately influence treatment recommendations and client outcomes—in particular among adolescents.