Introduction
In the context of eating disorders, there are few empirically-supported treatments, defined as specific treatments shown to be effective in controlled research trials (Chambless & Hollon, 1998). High quality systematic reviews have demonstrated that specific forms of cognitive-behavioral therapy (CBT) are efficacious for a range of eating disorder presentations in the short and long-term (e.g., Brownley et al., 2016, National Institute of Clinical Excellence, 2017). There is also evidence that there are no statistically significant outcome differences between CBT and interpersonal psychotherapy (IPT) at long-term follow-up periods (Linardon, Wade, De la Piedad Garcia, & Brennan, 2017a). International clinical guidelines for eating disorders now recommend the use of psychological treatments that show strong empirical support, although some recommendations are also non-evidence based and likely reflect the particularities in healthcare systems (e.g., availability of outpatient services, amount of therpists trained in a particular theoretical orientation etc.; see Hilbert, Hoek, & Schmidt, 2017). From eight available clinical guidelines that recommend psychological treatments for eating disorders, all recommend CBT for bulimia nervosa (BN) and binge eating disorder (BED), and six recommend CBT for anorexia nervosa (AN). Four clinical guidelines recommend IPT for BN and BED, and two recommend IPT for AN. Family-based therapy, particularly for adolescents, is recommended by six and four guidelines for AN and BN, respectively. Other interventions recommended less frequently by clinical guidelines include psychodynamic therapy and MANTRA (see Hilbert et al., 2017).
Although the efficacy of specific psychological treatments, such as CBT, IPT, and FBT, has been demonstrated in numerous randomized controlled trials (RCTs), there is still room for improvement in treatment retention and outcomes. For example, attrition, relapse, and/or partial response is common in RCTs evaluating CBT and IPT (e.g., Agras, Walsh, Fairburn, Wilson, & Kraemer, 2000), although there is evidence to suggest that treatment outcome and retention rates are improving when new and enhanced versions of CBT (CBT-E) are delivered (Fairburn et al., 2015).1 Some authors have therefore argued that a broader range of effective eating disorder treatments are needed (Wonderlich et al., 2014). The “third-wave” behavioral therapies have been suggested as potential alternatives for the treatment of eating disorders (Juarascio, Manasse, Schumacher, Espel, & Forman, 2017).
In general, while third-wave behaviour therapies have retained many of the same components as “second wave” CBT (e.g., self-monitoring, exposure and response prevention), they also use new methods and assumptions to achieve improvements in psychological functioning and clinical change (Hayes, 2004). Whereas CBT directly targets the content and validity of cognitive processes, third-wave therapies target the function or awareness of cognitions and emotions (Hofmann & Asmundson, 2008). Consequently, third-wave therapies emphasise strategies that foster acceptance, mindfulness, metacognition, and psychological flexibility, and reduce experiential avoidance (Hayes, Villatte, Levin, & Hildebrandt, 2011). This means that third-wave therapies target response-focused emotion regulation strategies, i.e., strategies that modulate the expression or experience of emotion regulation after its initiation, whereas CBT targets antecedent-focused emotion regulation strategies, i.e., strategies that prevent the emotion response from being activated (Hofmann & Asmundson, 2008).
There are some differences of opinion regarding the therapeutic interventions that fall under the category of third-wave behaviour therapies (Kahl, Winter, & Schweiger, 2012). However, a general consensus is that acceptance and commitment therapy (ACT), dialectical behaviour therapy (DBT), compassion mind training/compassion-focused therapy (CFT), mindfulness-based interventions (MBI), functional analytic therapy (FAP), schema therapy (ST), and metacognitive therapy (MT)2 all fall under the third-wave behaviour therapy umbrella (Hayes, 2004, Hayes et al., 2011, Öst, 2008). These specific therapeutic interventions will therefore form the basis of this review.
Numerous systematic reviews and meta-analyses have examined the efficacy of third-wave therapies for several common mental health conditions. Dimidjiian and colleagues recently synthesised the evidence from all the available meta-analyses (k = 26) of third-wave therapies (Dimidjian et al., 2016). Most meta-analyses were based on third-wave therapies for mood and anxiety disorders, with only a small number considering personality, substance abuse, and eating disorders. From their synthesis, Dimidjiian and colleagues concluded that specific third-wave treatments such as ACT, DBT, MBIs, and BA are supported by numerous RCTs, which, when combined, demonstrate a large within-groups effect size (i.e., pre-post symptom change), and a moderate between groups effect size (using mainly wait-list controls or treatment as usual as a comparison). Meta-analyses have also been performed comparing ACT to CBT, and these meta-analyses have reported no significant outcome differences between these treatments for anxiety disorders, general mental health conditions, and depressive symptoms (A-tjak et al., 2015, Bluett et al., 2014, Hayes et al., 2006, Ruiz, 2012).
The efficacy of third-wave therapies for eating disorders is much less clear. Two meta-analyses of specific third-wave therapies have been conducted. First, Lenz, Taylor, Fleming, and Serman (2014) evaluated the effectiveness of DBT for BED and BN by calculating within- (pre-post change) and between-groups (comparing DBT to wait-lists or TAU conditions only) effect sizes for eating disorder behaviours. Large within-groups (k = 4, d = 1.43) and between-groups (k = 4, d = 0.82) effect sizes were observed, leading the authors to conclude that DBT is a potentially effective treatment for eating disorders. Second, Godfrey, Gallo, and Afari (2015) reviewed studies that administered any form of MBI to treat binge eating in BED and non-clinical samples. Nine MBI studies, 6 DBT studies, and 4 ACT studies were included, and their meta-analysis was based on all interventions combined. Large (g = 1.12) within-groups and moderate (g = 0.70) between-groups effects favouring MBIs over wait-lists or TAU conditions were observed. Overall, these findings suggest that specific third-wave therapies such as DBT and MBIs are potentially effective treatments for BN and BED, at least in comparison to wait-list or TAU.
Despite the limited evidence of third-wave therapies for eating disorders, research has shown that clinicians are using third-wave techniques at least as often as they are using techniques derived from evidence-based therapies (e.g., CBT) to treat eating disorders. For example, Cowdrey and Waller (2015) found that the percentage of clients with eating disorders who reported that their therapist utilized mindfulness (77%) was typically larger than the percentage who reported their therapist used CBT-specific techniques such as food monitoring records (53%), weekly weighing (39%), and regular eating (82%). The use of third-wave therapies rather than empirically supported treatments raises concerns that those seeking treatment are not being provided with the most effective therapies. Therefore, a critical synthesis of the available literature on all third-wave eating disorder treatments studied to date is timely and pertinent.
This study therefore aims to examine the efficacy of third-wave therapies for eating disorders by (1) computing pre- to post-treatment and pre-treatment to follow-up effect sizes, and (2) comparing third-wave therapies to wait-lists, active controls, and empirically supported eating disorder treatments (i.e., CBT and IPT). Based on the available literature, we aim to investigate whether each specific third-wave therapy meets the criteria required for an empirically-supported treatment for eating disorders proposed by Chambless and Hollon (1998). Chambless and Hollon (1998) differentiated between (a) empirically-supported treatments that are specific in their mechanisms of action, i.e., therapy outperforms a pill or alternative evidence-based treatment in multiple RCTs conducted by different research teams, (b) efficacious therapies, i.e., therapy outperforms no treatment in multiple RCTs conducted by different research teams, and (c) possibly efficacious therapies, i.e., therapy outperforms no treatment in one study or by more than one study conducted by the same team.
The original criteria for empirically-supported treatments proposed by Chambless and Hollon (1998) were selected over more recent criteria (e.g., proposed by Tolin, McKay, Forman, Klonsky, & Thombs, 2015). As newer criteria have been criticised (for a full commentary, see Chambless, 2015), and the Chambless and Hollon (1998) criteria are still the most commonly implemented in psychological treatment research (e.g., Steinert, Munder, Rabung, Hoyer, & Leichsenring, 2017), we used the original criteria for establishing the empirical status of the third-wave therapies.
Section snippets
Method
This review was conducted in accordance to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines (Moher, Liberati, Tetzlaff, & Altman, 2009).
Study characteristics
Table 1, Table 2 present the characteristics of the RCTs and non-RCTs, respectively. In total, 15 used a transdiagnostic sample and nine a BED sample, two a BN sample, and one study sampled individuals with AN. Fourteen studies evaluated DBT, six evaluated MBIs, three evaluated CFT, two evaluated ACT, and two evaluated ST. In Table 3, we describe the underlying theoretical model of each of these third-wave therapies that have been tested in individuals with eating disorders. Thirteen studies
Discussion
This systematic review examined the empirical standing of the third-wave behaviour therapies for the treatment of eating disorders. Findings show that while third-wave therapies resulted in symptom improvements and were more efficacious than wait-list controls, third-wave therapies were general not superior to active psychological comparisons. Each third-wave therapy resulted in moderate to large improvements in eating disorder and general psychological symptoms from pre-treatment to
Conclusion
This study was the first to evaluate the empirical standing of third-wave therapies for the treatment of eating disorders, by both qualitatively synthesising the available findings of third-wave therapies in RCTs and quantitatively estimating the size of these treatment effects. Although there is promising preliminary evidence of the potential efficacy of specific third-wave therapies for certain eating disorders, no third-wave therapy currently meets formal criteria for an empirically
Conflict of interest
We wish to confirm that there are no conflicts of interest with this publication.
Role of funding sources
This project is supported through the Australian Government's Collaborative Research Networks (CRN) program. The CRN program had no involvement in the study design, collection, analysis or interpretation of the data, writing the manuscript, or the decision to submit the paper for publication.
Contributors
The first and last author was responsible for conceptualising the paper, conducting the statistical analyses and drafting and editing the manuscript. Author 2 was responsible for conceptualising the paper, and drafting and editing the manuscript. Authors 3 and 4 were also responsible for drafting and editing the paper. We confirm that the manuscript has been read and approved by all named authors and that there are no other persons who satisfied the criteria for authorship but are not listed.
Acknowledgements
We would like to thank Professor John Gleeson, Dr. Keong Yap, and Ms. Kylie Murphy for their valuable feedback on earlier versions of this manuscript. This project is supported through the Australian Government's Collaborative Research Networks (CRN) program. CGF holds a Principal Research Fellowship from the Welcome Trust (046386).
Recommended articles (6)
© 2017 Elsevier Ltd. All rights reserved.