Interventions for improving psychological symptoms in binge eating disorder (BED) and loss of control (LOC) eating in childhood and adolescence: a systematic scoping review

Abstract

Background

Despite reports of high incidence and prevalence, relatively few studies have investigated outcomes for children and adolescents with binge eating disorder (BED) and loss of control (LOC) eating. This study aimed to scope the available literature systematically.

Methods

A systematic scoping review methodology was implemented. Five databases (Medline, PsycInfo, Embase, CENTRAL and Scopus) were searched on the 23rd of August 2024 for relevant peer-reviewed journal articles and dissertations. No beginning time point was specified, and the end time point was chosen as the 23rd of August 2024. Restrictions were placed on age (under 20), diagnosis (BED, LOC eating) and study design (quantitative).

Results

Ten quantitative studies were identified: eight randomised controlled trials and two case series. Outcome data for 2400 young people were synthesised. Most studies (70%) had a sample size of fewer than 100 participants. Treatment modalities were heterogeneous and included psychological therapies such as cognitive behavioural therapy (CBT), dialectical behavioural therapy (DBT), interpersonal therapy (IPT) and group schema therapy. One study examined the role of medication. The results suggested that six different treatment modalities (CBT (group and individual), IPT, DBT, family-based IPT (FB-IPT), and medication) were associated with a reduction in the number of binge eating episodes and LOC eating. In terms of other psychological aspects such as depression, anxiety and self-esteem, the data were sparse, and it was difficult to draw meaningful conclusions.

Conclusions

The findings highlight a paucity of evidence-based interventions in this area for young people with BED and LOC eating. This is an emerging and important field in child and adolescent eating disorders as it is now ten years since BED was introduced into the DSM-V with prevalence estimates higher than other eating disorders in this population. As the onset of binge eating often occurs in late childhood or adolescence there is a role for early intervention. Further research into the efficacy of different therapeutic options for this age group is needed.

Plain English Summary

In this review, the authors searched for studies where treatment for binge eating disorder (BED) or loss of control eating (LOC) in young people (under 20 years old) had been trialled. They wanted to see which treatments could improve the mental health of young people with these conditions. To ensure that as many studies as possible were included, five different databases were searched. Ten studies were found and the majority of these were small studies with less than 100 participants. Nine of the studies investigated the role of talking therapies, for example, cognitive behavioural therapy and one study examined how medication helped. The talking therapies used in each of the nine studies were varied. The authors found that, on the whole, talking therapy and medication helped with the number of binge eating episodes and LOC eating but it was less easy to understand if they helped with depression and anxiety which are both commonly associated with BED and LOC eating. The results of this review show that more research is needed into this area as few studies were found and BED and LOC eating are becoming more frequently diagnosed in young people.

Background

Evidence suggests that binge eating disorder (BED) is the most common eating disorder amongst adolescents [1] with recent estimates suggesting a prevalence of 1- 4% [2, 3]. It appears to be more gender-balanced with one study finding adult women were at increased odds of 2–3 times that of men in terms of being affected by the disorder [4]. This is in contrast to anorexia nervosa (AN) and bulimia nervosa (BN) where the female-to-male ratio is closer to 10:1 [5, 6]. BED has also been reported to occur more among diverse ethnic and racial backgrounds [7].

BED and binge eating are associated with both impaired physical and mental health, with increased rates of obesity [8] and co-morbid anxiety and depression [7, 9]. The psychological impact of BED is present regardless of body mass index (BMI) or body weight which highlights the importance of the distress associated with binge eating, in the clinical impairment of this disorder [10]. BED is also often recurrent, with two peaks of onset, immediately after puberty (mean age of 14 years) and late adolescence (between 18–20 years) [2]. It is therefore important to consider the role of early intervention [11, 12].

BED has been included as its own diagnostic entity in the Diagnostic Statistical Manual of Mental Disorders (DSM) since the 5th edition [13]. The aetiology of BED is hypothesised to be complex involving multiple bio-psycho-social influences [14]. However, one key theory for the maintenance of the disorder is the affect regulation model where binge eating episodes are triggered by, and act as a temporary relief from, negative affect [15]. Emotional regulation difficulties are likely key in this population and it has also been noted that interpersonal difficulties have been reported more frequently in young people with BED [16]. It has also been hypothesised that BED results from certain neurocognitive profiles such as difficulties in inhibitory control and altered reward processing [17]. Linked to this is a ‘food addiction’ hypothesis, due to neurocognitive similarities between BED and addictive substance use [7]. This is a contentious area which requires further study [18].

Treatment is most often based in psychological work. International intervention guidelines are largely lacking for the treatment of child and adolescent BED. UK NICE (2017) guidance suggests using the adult recommendations, for which guided self-help Cognitive Behavioural Therapy (CBT) is the first line recommended treatment and the most frequently used psychological approach in adults with BED [7]. Other commonly used treatment modalities with adults include Interpersonal Psychotherapy (IPT) and Dialectical Behavioural Therapy (DBT) [19]. The rationale for these interventions is based on the proposed maintenance theories of binge eating around negative affect, interpersonal difficulties (noted to be more common in people with BED and noted to be related to LOC eating [20]) and emotional dysregulation. IPT aims to enhance the quality of relationships, which in turn can improve mood, meaning that individuals are less likely to use food as a coping mechanism. DBT has been shown to be helpful in adolescents struggling with emotional dysregulation [21]. Medication (serotonin reuptake inhibitors [SSRI] or stimulants) have also been trialled as a treatment option [7] with lisdexamfetamine being the only licensed medication for BED in adults in the United States of America (USA) [22].

This systematic scoping review aims to synthesise the current evidence for interventions that target the psychological symptoms of BED and LOC eating in children and adolescents. A review exploring this has recently been published, however, it only included the adult population [7]. This review’s importance stems from the prevalence of BED in childhood and adolescence along with the impact on both physical and mental health as well as the potential for early intervention. LOC eating was included in this scoping review due to the limited available data for BED and because it is a core component of the presentation of BED [23]. LOC eating is characterized by a subjective lack of control over eating, regardless of the amount of food reportedly consumed [24]. LOC eating is the BED criterion most associated with negative psychological symptoms, for example, depression and anxiety [25]. Furthermore, research has found LOC eating behaviours during adolescence increase the risk of developing eating disorders in later life [26].

This review focused on psychological symptoms, rather than BMI or weight loss. Weight is not part of the diagnostic criteria for BED and the psychological impact of BED is present regardless of BMI or body weight [27]. This also fits with an increased drive in recent years to reduce the focus on weight loss and focus more on improvement in psychological well-being, as encapsulated in the ‘Health at Every Size’ (HAES) framework [28]. The HAES philosophy promotes the concept that an appropriate weight for an individual cannot be determined by calculating BMI or body fat percentages and that it instead should be defined by the weight a person settles at when they eat listening to internal signals of hunger/satiety and participate in regular and appropriate physical exercise [29]. This review focused on studies that examined BED or LOC eating only rather than those that merged results with treatment for BN. This was because although there is some overlap between these conditions, it was deemed important to delineate the treatment, as BED is a distinct and increasingly prevalent eating disorder amongst the adolescent population.

Method

Existing research into interventions that improve psychological symptoms for BED was explored using a systematic scoping review methodology [30]. This methodology was deemed to be most appropriate given the relative lack of studies in this area. To conduct this review, current systematic scoping review guidelines were used [30] along with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) extension for scoping reviews [31]. The search criteria were developed by one author (PLB) who synthesized the research question using the PCC (population, concept, context) framework [32] (see supplementary information, appendix 1). PLB executed the initial search strategy (see supplementary information, appendix 2), selection of studies and data extraction. This was reviewed by JB in an iterative process and a consensus reached. As per guidelines regarding scoping review methodology, a review protocol was not registered on PROSPERO [33].

Eligibility criteria

Eligibility criteria for the review are presented in Table 1.

Table 1 Eligibility Criteria

Search strategy

Five databases (PsycInfo, Medline, Embase, CENTRAL and Scopus) were searched. Searches using variations of the terms ‘binge eating disorder’/ ‘subthreshold binge eating disorder’, ‘adolescents’ and ‘intervention’ were carried out on 23rd August 2024 (see supplementary material, appendix 1 for exact search terms used). A broad search strategy was created to capture as many studies as possible.

Selection process

After the initial search was completed, the references were downloaded into Zotero software, which was then used throughout the screening process. Duplicates were identified and removed. Titles and abstracts were scanned, and papers initially included/excluded based on title and abstract. For those papers included after this initial screen, full-text articles were retrieved online, and studies were then screened again and included/excluded according to the eligibility criteria. Citation searching was then carried out to screen for further papers that may be eligible to be included. See Fig. 1 for the PRISMA flowchart.

Fig. 1

PRISMA flow diagram [34]

Data extraction and analysis

All studies included were quantitative and categorised according to three possible study types: randomised control trials (RCTs), non-randomised comparison studies and case series. Baseline data were extracted from each study (age, gender, ethnicity) as well as intervention characteristics (setting, modality, number of sessions, duration of intervention). In terms of reporting outcomes from the interventions, data extraction focused on data that reported change in psychological symptoms both pre- and post-intervention (and at follow-up if available). Body mass index (BMI) data was not extracted. Effect sizes were reported if available. As many eating disorder treatment modalities view the family as an important part of the therapeutic process [35], data regarding parental involvement during the intervention was also extracted. In line with current scoping review guidance, risk of bias assessment was not completed [33] as the main aim of this review was to map the evidence available. For consistency, if intention to treat analyses (ITT) were available they were reported. It is worth noting that during the screening process, many studies focused on BED and BN together and did not report outcomes for the disorders separately meaning BED outcomes could not be reported.

Results

Study selection and characteristics

7852 studies were initially identified from the systematic search. Once duplicates had been removed both with support from Zotero software and manually, 4200 studies remained (see Fig. 1, PRISMA flow diagram).

Once screening according to the eligibility criteria was completed, a total of ten studies were considered eligible for this review and synthesised. Data for a total of 2400 young people are reported. Most studies (70%, 7/10) had a sample size of fewer than 100 participants. 80% (8/10) were carried out in the USA with the remaining two studies carried out in Iran and Germany. 80% (8/10) were published within the last ten years (2014-) with the remaining being published in 2009 and 2010. In terms of diagnosis, 70% (7/10) of studies used LOC eating or subthreshold BED as their inclusion criteria. The definition of ‘LOC’ or ‘subthreshold’ varied between studies. Two studies included those with BED only and one study was mixed with both BED and subthreshold BED. Of the ten studies included in the review, 60% (6/10) specified in their inclusion criteria that the children or adolescents should be overweight/over a certain BMI percentile. 80% (8/10) of studies were RCT, and 20% (2/10) were case series (see Table 2). See Tables 3 and 4 for a summary of the outcomes of the studies.

Table 2 Methodologies of included studies

Table 3 Treatment modalities across the included studies

Table 4 Summary of findings for psychological interventions in BED and LOC eating in children and adolescents from included studies

Narrative synthesis

Interventions: treatment modalities, population, setting and duration

Treatment modalities varied across all ten studies ranging from adapted DBT (2 studies) to group schema therapy (1 study). See Table 3 for further details. 60% (6/10) of interventions were group interventions, 30% (3/10) were individual interventions and 10% (1/10) was an intervention involving the parent–child dyad. In terms of parental involvement, 80% (8/10) of studies did not include direct parental involvement during the intervention. The ages of the children and adolescents included ranged from 8–19. Four studies had 100% female participants, five had majority female participants (≥ 66%) and only one study had < 50% female participants (46.7%). Three studies reported that most participants were Caucasian, and four studies did not report ethnicity. All studies were carried out in an outpatient treatment setting. Duration of intervention ranged from 8–16 weeks for all included studies apart from one study where medication (lisdexamfetamine) was the intervention. In this study, treatment duration averaged 19.1 months [36]. Aside from this study, all nine other studies offered weekly sessions in the intervention phase.

Outcomes: RCTs

Eight studies used an RCT design (Table 2). Three studies randomized participants to a psychological intervention or a waitlist control [37,38,39]. Three studies randomized participants to a psychological intervention or a health education group [40,41,42]. One study randomized participants to a psychological intervention or a weight management group [43] and one to CBT and two different types of exercise groups [44].

All RCTs used different interventions making direct comparisons challenging. 63% (5/8) used group interventions, 25% (2/8) used individual interventions and one used parent–child dyads as part of a family intervention. Three RCTs had more than 100 participants [37, 40, 44]. Jones et al., (n = 105) randomized male and female adolescents from two high schools (mean age = 15.1) to either an internet-facilitated intervention, Student Bodies 2-BED (n = 52), or the waitlist control group (n = 53) [37]. To be included the adolescents were required to be greater or equal to the 85th percentile for age-adjusted BMI and to have BED or subthreshold BED. The latter was defined as exhibiting binge eating or ‘over-eating behaviours’ ≥ 1 time per week in the previous 3 months. The internet intervention was an individual semi-structured programme which included psychoeducation and cognitive behavioural interventions based on the principles of Fairburn’s CBT for BED [45]. From baseline to end of treatment, the number of objective binge eating episodes (OBEs) and subjective binge eating episodes (SBEs) decreased in both groups and this was significant in the intervention group (Mann Whitney U test (U₈₁): 565, p < 0.01). This was also true from baseline to follow-up (U₈₄: 562.5, p < 0.05). Weight and shape concerns decreased significantly among those who completed the intervention but not in the intention to treat analysis. No significant differences were noted in changes in objective overeating episodes (OOEs) or symptoms of depression in the intervention group. This was the same for OBEs, SBEs and OOEs in the waitlist group. The authors suggested that this might represent the variable nature of binge eating which makes it difficult to objectively measure, leading to measurement errors [37]. It is worth noting there were low adherence rates to this study (for example, mean number of screens viewed on the programme was 20.2 out of a possible 104).

Two other RCTs also used CBT-based interventions [38, 44]. Hilbert et al. randomized 73 adolescents (mean age: 15.3) to age-adapted CBT (n = 37) or a waiting list (WL) (n = 36). Binge eating episodes, elicited from the Eating Disorders Examination (EDE), significantly reduced in the intention to treat analysis, with a mean of 4.7 (p < 0.01) fewer monthly binge eating episodes in the CBT compared to the WL group. They also noted higher rates of abstinence from binge eating, higher rates of remission from BED and lower eating disorder psychopathology on the EDE (all p < 0.05) at post-assessment in the CBT group. Symptoms of depression, quality of life and self-esteem did not change in either group. At 6-, 12- and 24-month follow-up from the CBT intervention, binge eating episodes remained significantly reduced (p < 0.001) compared to baseline.

Mehlenbeck et al. carried out two RCTs to investigate binge eating symptoms following participation in a behavioural weight control intervention [44]. In addition to weight data, they reported psychological measures such as self-worth and social acceptance, hence inclusion in this review. Inclusion criteria for the studies were to report subthreshold binge eating symptoms and to be 20–80% overweight by BMI (study 1) and 30–90% overweight by BMI (study 2). In both RCTs, adolescents were assigned to a ‘CBT + ’ intervention exercise programme (1. CBT + peer enhanced adventure therapy (PEAT) or 2. CBT + aerobic exercise (EXER)). These involved a dietary and exercise prescription along with behavioural strategies. The behavioural topics covered included portion control, problem-solving, motivation and self-monitoring of diet and exercise. The peer-based activity was designed to increase teamwork and social skills whilst the aerobic exercise was purely focused on exercise. In study 1, the weight control intervention was based on food type and serving size. In study 2, it was based on calorie restriction (1400 -1600 cal/day). All other parts of the interventions were identical. Parents also attended a separate weekly parental group. In both studies, results indicated a significant reduction in binge eating symptoms assessed using the Binge Eating Scale [44]. In study 1, baseline global and physical appearance, self-concept and physical worth were positively associated with changes in binge eating symptoms. Global self-concept was negatively associated with changes in binge eating symptoms. In study 2, a negative relationship was observed between changes in dimensions of self-concept and binge eating symptoms, meaning improved physical appearance, global self-concept, and physical self-worth at the end of treatment were associated with a decrease in endorsement of binge eating symptoms. Decreases in binge eating scores were related to improvements in several dimensions of self-concept including global self-worth (both studies 1 and 2) and physical appearance and self-worth (study 2). This relationship is suggestive that those who binge-eat less, feel better about themselves or those who feel better about themselves are less likely to binge-eat.

Three RCTs used adaptions of interpersonal psychotherapy (IPT) as their main intervention. The largest of these studies (n = 113) was carried out by Tanofsky-Kraff et al. in 2014. 55 adolescent females were randomized to IPT and 58 to a health education (HE) group. Inclusion for the study was a BMI between the 75th and 97th percentile and LOC eating, defined as 1 episode of LOC in the past month [40]. 32.7% also reported binge eating. Results were mixed with symptoms of anxiety, depression and episodes of LOC eating at the 12-month follow-up significantly reduced in both groups (p < 0.001). However, at 12-month follow-up, IPT was more helpful than HE in reducing objective binge eating (p < 0.05). Also, girls of ethnic-racial minorities (majority African-American), if they were assigned to IPT, experienced significantly greater reductions in LOC episodes at 1-year follow-up than those assigned to HE.

In 2016, Tanofsky-Kraff et al. randomized 88 adolescent girls (aged 12–17) to IPT (n = 46) or HE (n = 42) [41]. At baseline, 6-month and 12-month follow-up, participants ate a lunch designed to simulate LOC eating in a laboratory. The treatment started after the baseline lunch and consisted of a 1.5-h individual session and 12 weekly group sessions. They found that those who participated in the IPT group had a significant decrease in pre-meal tension and anger at 1-year follow-up. The authors suggested that perhaps IPT may reduce the negative affect linked to LOC eating proposed by affect theory [15]. In terms of depressive mood, they found those randomized to IPT experienced no change in pre-meal depression whereas those in HE by follow-up at 1 year had an increase in pre-meal depressive symptoms.

In the third study, Shomaker et al. used a family-based IPT intervention (FB-IPT) consisting of parent–child dyads (n = 15) and a family-based health education group (FB-HE) as the control (n = 14). They hypothesized that underlying parent–child communication difficulties may contribute to LOC eating as the latter often occurs in response to negative affect [20]. Outcomes were measured directly post-intervention and at 6- and 12-month follow-up. The study found that a greater decrease in LOC eating at end of treatment was observed following FB-IPT compared to the FB-HE (38% versus 77%, p < 0.05) [42]. However, there were no differences between groups at either 6- and 12-month follow-up. Global disordered eating attitudes did decrease from baseline to follow-up, and at 6 months this was more pronounced with the FB-IPT group. However, at 12 months, all children had decreased global disordered eating attitudes (ps < 0.001) with no difference between the interventions (p = 0.97). They also found that the dyads who had been randomized to FB-IPT had decreased depressive symptoms (p < 0.01). This was noted in both treatments at 6-month follow-up, with no differences between treatments (p = 0.47). At 1-year follow-up, it was FB-IPT that appeared to demonstrate a long-standing impact on a reduction in depressive symptoms (p < 0.01). The FB-IPT intervention also demonstrated a significant effect on reduction in anxiety at post-treatment (p < 0.001) and at the 6-month follow-up (p < 0.01) unlike the FB-HE intervention. However, at 1 year neither intervention showed significant change from baseline in terms of anxiety reduction. The study also examined social adjustment and found that at all follow-up points, there was no significant improvement in either intervention. It is worth noting that in this study at baseline measurements, the average LOC episodes per month were small for both the FB-IPT and FB-HE groups, 3.1 and 2.6 respectively [42].

Other treatment modalities studied in the other included RCTs were group schema therapy [39] and adapted dialectical behaviour therapy (DBT). Saravi et al. examined the effects of group schema therapy (n = 15) vs a waitlist control (n = 15) for female adolescents (aged 15–17) with BMIs between 25–29.9 and who scored 17 or above on the Binge Eating Scale [46]. They noted a significant improvement in mean scores of eating attitudes both immediately post-treatment and at 3-month follow-up for those randomized to the group schema therapy intervention. They also reported a significant improvement in self-regulation and eating attitude in the schema therapy group compared to the control group (p < 0.05).

Mazzeo et al. randomized to ‘LIBER8’ (Linking Individuals Being Emotionally Real) (n = 28) or ‘2BFit’ (n = 17) for three out of five waves of their study. ‘LIBER8’ is a 12-week, weekly group intervention based on CBT principles with integrated DBT skills training and ‘2BFit’ is a behaviourally based weight management group where participants are set individualized diet and exercise goals with sessions to check adherence to goals and discuss potential barriers to achieving these [47]. For the final two waves, they did not run the weight management group (due to difficulties in recruitment). The participants met the criteria for LOC eating or BED. Both LIBER8 and 2BFit groups demonstrated significant improvements (p < 0.05) pre- and post-intervention on the Eating Disorder Examination Questionnaire (EDE-Q), eating concern, shape concern, restraint, and global score. The 2BFit group also reported significant improvements in the EDE-Q weight concern score compared to the LIBER8 group.

In terms of broader mental health symptoms, assessed by the Emotional Eating Scale for Children (EES-C) [48] in the domains of anxiety, anger and frustration, depressive symptoms and feeling unsettled, no significant differences were observed between groups. Pairwise comparisons revealed significant improvements in the 2BFit group in the areas of anxiety, anger and frustration and depressive symptoms (ps < 0.05), whereas no significant changes in any of the domains were observed in the LIBER8 group. A significant reduction in negative affect pre- and post-intervention was observed for LIBER8 (p < 0.007) but not for the 2BFit group.

Outcomes: case series

Two of the included studies were case series. One evaluated the impact of a condensed DBT skills group [49] and the other the impact of medication [36]. The latter was the only study identified that described the impact of medication on BED in adolescence and childhood. Kamody et al. [49] included 15 14–18-year-old adolescents who attended a 10-week DBT skills group. To be eligible they had to meet at least one of the DSM-V criteria for BED. Of those that completed the baseline measurements only 50% (15/30) went on to complete the intervention. Among the intervention completers, those meeting diagnostic criteria for BED decreased from six (40%) at baseline to three (20%) post-treatment. Of those who completed the 3-month follow-up (n = 11), only one met criteria for BED. The study also reported on OBE and emotional overeating (EO) via the Eating Disorder Examination Questionnaire (EDE-Q) and the Emotional Eating Scale for Children and Adolescents (EES-C) respectively. These were given to both the young people and their caregivers. OBE and EO frequency reported by both young people and their caregivers decreased following the intervention at both post-treatment and at 3-month follow-up.

In the other case series, a retrospective chart review was carried out of 25 young people (aged 10–19, mean age 16.5) with a diagnosis of BED who were prescribed lisdexamfetamine [36]. The mean duration of treatment was 19.1 months, 36% (9/25) were receiving concurrent psychotherapy for BED and 12% (3/25) were also on aripiprazole at the time of treatment. 16% (4/25) reported complete remission of their symptoms, 24% (6/25) reported an improvement in their binge eating frequency, 4% (1/25) less frequent sneaking of food and 8% (2/25) were likely to binge if lisdexamfetamine skipped.16% (4/25) reported no response to medication and 8% (2/25) reported a worsening of their symptoms.

Diagrammatic synthesis

Table 5 presents a synthesis of the different treatment modalities from the ten included studies and their impact at end of treatment on psychological symptoms in BED. The end-of-treatment effect was chosen as the endpoint due to the heterogeneity of follow-up amongst the included studies. However, it is important to note that the IPT interventions reported on follow-up data, so their results are included in the table in brackets, for comparison. 6 treatment modalities [7 studies] were suggestive of an improvement in binge eating episodes or LOC eating following their intervention, with both individual CBT studies noting an improvement following participation. As discussed above, the FB-IPT intervention noted improvement post-treatment but found no difference between groups at 6- and 12- month follow-up [42]. Table 5 also demonstrates that many different areas of psychological symptoms were explored in these studies although findings are heterogeneous.

Table 5 Diagrammatic synthesis of treatment modalities and the end-of-treatment impact on psychological symptoms in BED

Discussion

In this systematic scoping review, ten papers were identified that report on the outcomes of interventions aimed at improving psychological symptoms for children and adolescents with BED and LOC eating. The findings highlight a paucity of research in this area, both in terms of the age range studied and outcomes. The dearth of studies focusing on psychological symptoms and psychopathology as outcomes is of particular note, with most studies focusing on BMI or weight loss. Studies included were mostly small in terms of number of participants and heterogenous in terms of the treatment modalities used and the psychological symptoms assessed (see Table 5). At best, it can be said that all treatment modalities showed some evidence of improving psychological symptoms (see Table 5) but more studies with larger samples are sorely needed.

As stated in the introduction to this review, owing to the minimal studies in adolescents purely focused on BED, studies focused on LOC eating were considered also. It is also worth reiterating that in the screening process, many studies focused on BED and BN together without reporting outcomes for each group separately. Only two studies identified in this review included full-syndrome BED as an inclusion criterion. These findings highlight the importance of further research in this population, especially given BED is estimated to be the most prevalent eating disorder amongst young people [1].

The findings from this review provide very preliminary, weak evidence for 6 different treatment modalities that may be associated with a reduction in binge eating episodes and LOC eating for children and adolescents. These are CBT (group and individual), IPT, DBT, FB-IPT and medication. In terms of other psychological symptoms, such as depression and anxiety, the data were sparse, and it is difficult to draw meaningful conclusions.

Like in adult studies of BED, treatment modalities piloted with adolescents included CBT, IPT, DBT and medication. Group schema therapy was also one intervention included, which emphasizes changing maladaptive coping styles and maladaptive schemas formed in childhood [39]. In previous research, it has been reported that people in larger bodies with BED have significantly higher scores in certain schemas (abandonment/instability, emotional deprivation, and inadequate self-control/self-discipline) than those in larger bodies without BED (Moloodi et al., 2010). Saravi et al. (2020) hypothesized that when schemas are activated due to negative emotions, emotional regulation is more difficult so schema therapy can help improve self-regulation. Their results did show an improvement in eating attitudes and self-regulation over time for their schema therapy intervention [39].

Hilbert et al.’s RCT demonstrated the potential longer-term effectiveness of CBT as an intervention for BED. This is one of the only studies included where the inclusion criteria for the study was a DSM IV-TR or DSM V diagnosis of BED [50].

One intervention unique to children and adolescents used in the studies identified was FB-IPT [42]. FB-IPT developed as a treatment approach for preadolescent depression [51]. Compared to preadolescents without LOC, preadolescents with LOC eating have less healthy familial communication [52]. The FB-IPT intervention appeared to show promise in demonstrating a longstanding impact on a reduction in depressive symptoms at the 1-year follow-up but showed no significant results for BED or related difficulties after post-treatment. As family-based therapy is typically the first line treatment for other child and adolescent eating disorders, for example, AN and BN [53, 54] this is somewhat surprising. This may be more of a reflection of the need to consider how to involve families in the treatment for BED and LOC eating rather than whether to involve them at all. Family therapy interventions are unique to children and as far as the researchers are aware, have not been evaluated in adult BED. With the reported association between LOC eating and family communication, this is an important area for further research [52].

In the 8 RCTs included, 3 randomised participants to a waitlist control [37, 39, 50] and 5 randomised to active control interventions [40,41,42,43,44], which included health education, weight management and exercise groups. It is worth noting that some of these ‘active control groups’ had significant results and were not always psychological treatments. In Tanofsky-Kraff et al. (2014) symptoms of anxiety and depression and episodes of LOC eating at the 12-month follow-up significantly reduced in both the HE group and the IPT intervention [40]. For Mazzeo et al. both the DBT and 2BFit group (weight management active control group focusing on nutrition and exercise habits) demonstrated significant improvements pre- and post-intervention on eating disorder psychopathology (EDE-Q; eating concern, shape concern, restraint, global score). However, the active control group also reported significant improvements in the EDE-Q weight concern score. In terms of mood symptoms, there were significant changes in the active control group in the areas of anxiety, anger and frustration and depressive symptoms. This highlights the importance of considering the associated benefits of an intervention, for example; both the intervention and active control groups may provide social support and cohesion which is known to improve mood [41, 55].

It is apparent from reviewing the literature that there is a recent increase in interest in BED and its development and maintenance, despite the lack of reported outcome data. A few case studies were identified that focused on family therapy. Additionally, several small pilot studies also noted the role of exposure work in binge eating treatment, however, studies were excluded as they did not meet the inclusion criteria (case studies, no statistical analysis, or over age limit). Nevertheless, they suggest a burgeoning interest in novel treatments for BED and LOC eating, which is much needed.

Regarding medications for BED and LOC eating, only one study was identified by the search strategy. Several have been trialled on adults [7] with only a few case reports noted for the use of medications in childhood and adolescence [56, 57]. This was also noted anecdotally during the screening process, where papers discussing medication use were predominantly focused on adults. The medication used in the included study in this review, lisdexamfetamine, has been licensed for use in adults with BED in the USA by the U.S. Food and Drug Administration (FDA) since 2015 [22]. It is only currently licensed for Attention Deficit Hyperactivity Disorder (ADHD) in children and adults in the UK [58]. There is some evidence that ADHD and LOC eating may be related to each other in terms of impulsivity [59] and some evidence of an association between BED and ADHD [60]. Weight loss has been demonstrated in adults from use of lisdexamfetamine [61] but results regarding mood and eating disorder psychopathology are limited. Furthermore, adverse effects need to be carefully monitored [7]. The case series included in this review had a small number of participants (N = 25) [36]. Results were mixed with 16% (4/25) reporting complete remission of their symptoms, 24% (6/25) reporting an improvement in their binge eating frequency but 16% (4/25) reporting no response to medication and 8% (2/25) reporting a worsening of their symptoms.

More than half the studies (60%, 6/10) reported weight loss or change in BMI as an outcome in addition to changes in psychological symptoms. Additionally, several studies had a higher BMI as an inclusion criterion and several included weight management and exercise support as part of their main intervention, control group, or both. Given BED is a psychological, not a physical health condition, there is concern described in the literature that the inclusion of weight management interventions and/or weight loss targets only reinforces weight stigma [62]. As noted by Grucza et al. [10] the psychological impact of BED is present regardless of BMI or body weight, suggesting that it is binge eating and its associated distress that are key when considering the clinical impairment of this disorder. This is not to say that internalized weight stigma, body image concerns and psychopathology are not interrelated for many. A recent systematic review [63] focused on the impact of weight loss interventions on binge/LOC eating in children and adolescents and found that weight loss interventions did result in a significant decline in binge/LOC eating (95% of studies). However, they noted that these studies were limited in their lack of reporting of changes on an individual level and that the one study that did assess on an individual level found that whilst binge eating prevalence decreased from 24 to 3%, four patients developed new binge eating behaviours during the study [63]. This highlights that the relationship between weight and binge/LOC eating is complex and needs to be carefully considered in future research.

Given evidence for the Health at Every Size (HAES) movement [29] and continually emerging evidence of the harm and confusion caused by conflating weight loss for people in bigger bodies with psychological treatment for people with BED, LOC eating and related eating disorders, it is important to try and distinguish and better understand the interplay between weight and psychological symptoms moving forward.

Limitations

There are some key limitations to this review. Firstly, the exclusion criteria. Only English language papers were reviewed so a Western perspective taken. Furthermore, editorials, non-peer reviewed books and book chapters were excluded along with conference reports. To make the review more feasible, owing to the lack of available studies, the scope of the review had to be widened to include LOC eating. In terms of studies identified, they were small with the majority having under 100 participants. Most participants were female. Statistical analyses, heterogeneity in follow-up periods and components of psychological symptoms were varied so it was difficult to draw comparisons. Furthermore, LOC eating was defined variably throughout all studies and only one study had BED diagnosis as an inclusion criterion. While this allowed for analysis of both BED and LOC behaviours, it is important to note that these are different concepts and therefore would benefit from separate analysis. The time limit of this review meant that RCTs currently registered but not completed on the Cochrane database, CENTRAL could not be included, for example, [64]. Therefore, the findings are only up to date until 23rd August 2024. Lastly, the review did not use statistical tests and in line with current scoping review guidance did not fully assess bias [33].

Conclusions

This systematic scoping review highlights the paucity of research for interventions to improve psychological symptoms in BED and LOC eating in the child and adolescent population. The aims were achieved as the extent of the knowledge currently available in this field was mapped and the existing gaps in knowledge identified. This is an emerging and important field in child and adolescent eating disorders as it is the most prevalent eating disorder in this population, and it is now ten years since BED became its own diagnostic entity in the DSM-V. Added to this as discussed previously, its impact on the physical and mental health of those affected is significant. As the onset of binge eating often occurs in late childhood or adolescence [12] there is a role for early intervention [65].

The above limitations and small number of studies included, make conclusions from this review tentative but it still serves to raise important questions about the focus on treatment for BED and the need for further studies. Given the varied treatment modalities and limited number and size of studies, more research into the different therapeutic options would be beneficial, including the less well-known modalities such as FB-IPT and group schema therapy. Further work into the role of the family and medication would also be beneficial. Lastly, more definitive descriptors for LOC eating and BE would allow for a more defined research population.