Introduction

Representative data from the USA, Europe and Australia show that around 13% of females experience a threshold or sub-threshold eating disorder (ED) by young adulthood, which are marked by chronicity, relapse, emotional distress, functional impairment and risk of future obesity, mood, anxiety and substance use disorders, suicide and mortality [1,2,3,4,5]. It is critical that EDs, which show significant and prolonged medical and psychosocial morbidity and stronger relations to functional impairment, suicide attempts and mortality than other psychiatric conditions (comprising mood disorders and schizophrenia), carry high healthcare costs and family burden and are one of the highest-ranking causes of mental health disability [6,7,8,9,10,11,12], are successfully fought. Yet over 80% of ED cases never receive treatment [2, 3] and existing treatments only result in lasting ED symptom remission for half or less patients [13]. Thus, a top public health and research priority is to design and broadly implement efficacious ED prevention programmes. These may be effective for EDs whose peak risk period for onset is around ages 16–19 [1, 2, 4], implying that broad implantation of efficacious ED prevention programmes during adolescence could decrease the population prevalence of EDs.

At the turn of the millennium, the state of the science on the prevention of EDs was severely inadequate. The roundtable discussion on the topic convened by the National Institute of Mental Health (USA) [14] highlighted that no prevention programmes had produced meaningful reductions in ED symptomatology or future ED onset, two crucial outcomes for ED prevention. Since then, the field of ED prevention has made significant strides in successfully translating basic ED risk factor research into efficacious interventions able to reduce ED risk factors and symptomatology and future onset of EDs. This Editorial aims at briefly documenting the aforementioned progress focusing on an intervention model, i.e. the dissonance-based ED prevention approach which, according to recent meta-analytic evidence, produces the largest intervention effects compared to any other type of extant selective (e.g. targeting high-risk people) or indicated (e.g. targeting people who exhibit symptoms, but do not meet ED diagnostic criteria) prevention programmes [15]. Further, this is the only ED prevention approach that meets the American Psychological Association’s [16] designation as an efficacious intervention, meaning that it has been found to (a) yield positive effects that have been replicated by independent labs/researchers and (b) significantly outperform active and credible alternative interventions in multiple randomized controlled trials. In addition to increasing awareness in the scientific community and community stakeholders on the empirical evidence base for dissonance-based ED prevention programmes [including basic ED risk factor research underpinning the development of these programmes and rigour tests on the intervention (cognitive-dissonance) theory underlying them], this Editorial offers several important research directions, particularly in areas in which queries remain open or we see room for improvement.

Empirical evidence base for dissonance-based ED prevention programmes

Several malleable sociocultural, behavioural and attitudinal risk factors for EDs (e.g. overeating, pursuit of the thin body ideal, self-objectification and body dissatisfaction) have been detected [2, 17], implying that prevention of these pernicious psychiatric disorders could be viable. Although dozens of ED prevention programmes have been designed and tested [15, 18,19,20,21,22,23], only three have decreased ED symptom composite measures [24,25,26] and/or future onset of EDs [25,26,27]. As noted, the most empirical support has accumulated for dissonance-based ED prevention programmes, often referred to as the Body Project, which has included numerous randomized (efficacy, effectiveness, and comparative) trials and rigorous tests of the intervention theory carried out by multiple, independent research labs/teams (summarized below). In the Body Project, adolescent girls and young women voluntarily critique the thin body ideal in behavioural, written and verbal exercises, theoretically generating dissonance that causes reduced pursuit of this ultra-slim, unrealistic and unattainable body ideal because individuals align their attitudes with their publicly displayed behaviours [28]. This prevention approach is based on the social psychological principle of cognitive-dissonance theory [29], which holds that people are motivated to maintain consistency between their attitudes and behaviours, and when people engage in behaviours that are inconsistent with their attitudes, they experience psychological discomfort that prompts them to align their attitude with their behaviour. For example, in the Body Project, participants complete role-plays in which they talk facilitators out of pursuing the ultra-slim body ideal, write a letter to a younger self on how to avoid concerns and body image problems, engage in body activism acts that challenge this unrealistic ideal and verbally generate costs associated with pursuing this ideal in response to Socratic questions. For interested readers the intervention (i.e. Body Project) script is available online at: www.bodyprojectsupport.org.

According to the dual pathway model of EDs (for a detailed description, see [30]), reduced subscription to the thin body ideal (often referred to as thin-ideal internalization [31]) should reduce body dissatisfaction, unhealthy weight control behaviours, negative affect, ED symptoms, and the risk for future onset of EDs. In support, basic ED risk factor research showed that elevated thin-ideal internalization, body dissatisfaction, negative affect, and dieting predict future onset of any DSM-5 ED (see [2] and the references therein)]. Research has also revealed that thin-ideal internalization typically emerges before body dissatisfaction, which emerges before negative affect and dieting, which typically precede onset of an ED, for those people who eventually display ED onset [30]. Thus, thin-ideal internalization is at the headwaters of an etiologic cascade of risk factors that give rise to ED onset and the Body Project that targets this variable should reduce onset of ED symptoms/syndromes, as well as adverse downstream risk factors such as body dissatisfaction and negative affect. Accumulated empirical evidence summarized below supports this notion.

Efficacy trials conducted by Stice and colleagues (e.g. [25, 28, 32, 33]) have shown that the Body Project yielded greater reductions in ED risk factors (i.e. thin-ideal internalization, dieting, negative affect and body dissatisfaction) and symptoms in female samples with body image problems/concerns relative to both credible alternative interventions and assessment-only control conditions, with several effects enduring through 3-year follow-ups. Similar effects have emerged in several efficacy trials carried out by independent research teams (e.g. [34,35,36]), though not all of these trials targeted females with body image problems/concerns. Moreover, the Body Project decreased ED symptomatology and electrocardiogram-assessed cardiac risk indices among females with sub-clinical and clinical EDs [37], which is relevant since cardiac problems are among the leading mortality causes among such people. Hence, the Body Project has demonstrated efficacy when implemented (a) universally to females who were not screened for body image problems, (b) selectively to females with body image concerns, and (c) in an indicated fashion to females with ED symptomatology. In addition to the aforementioned efficacy trials examining whether preventive interventions produce effects under carefully controlled experimental conditions, multiple independent effectiveness trials confirmed that the Body Project yields greater reductions in ED risk factors and symptoms compared with educational video control conditions and educational brochure, when college counsellors or undergraduate peer educators deliver the intervention under “real world” and ecologically valid conditions, with several effects enduring through 3-year follow-ups (e.g. [38,39,40,41]).

The evidence that the Body Project has yielded positive and reproducible effects across independent trials is notable given the recently evidenced “crisis” of replication in the fields of psychology and other sciences [42]. Further, the fact that the Body Project outperforms credible alternative interventions implies that the intervention effects are not simply due to demand characteristics, expectancies or non-specific effects inherent to randomized trials. As noted, a recent meta-analytic review by Watson and colleagues [15] showed that, averaging across all randomized trials, dissonance-based ED prevention programmes produced the largest reductions in ED symptomatology compared to any other type of extant selective or indicated prevention programmes, with an average d = 0.30 when compared with non-specific control conditions and an average d = 1.06 when compared with assessment-only control conditions. Perhaps most critically, the Body Project has also been found to significantly reduce future onset of threshold and sub-threshold EDs among participants free of any ED at baseline. Specifically, the Body Project has yielded a 60% reduction in future ED onset over 3-year follow-up relative to assessment-only controls [25] and a 57% reduction in future ED onset over 3-year follow-up relative to educational video controls [43]. In addition, there is trial evidence that virtual delivery of the Body Project produced a 73% reduction in future ED onset over 2-year follow-up relative to an expressive writing control [44], implying that this novel implementation approach could dramatically extend the reach of the dissonance-based ED prevention programmes.

In support of the intervention theory, decreases in thin-ideal internalization mediated the effects of the Body Project on ED symptom reduction (e.g. [45]). Consistent with the thesis that dissonance induction contributes to intervention effects, people assigned to versions of the Body Project designed to maximize dissonance reported greater ED symptom reduction that those assigned to versions of the Body Project designed to minimize dissonance but with the same intervention content (e.g. [46]). Further, the Body Project yielded larger effects for females with initial elevated thin-ideal internalization, in line with the notion that they should experience the highest dissonance [47]. Completing the Body Project also offset the ED risk conveyed by thin-ideal internalization [48]; females with this potent ED risk factor who completed the Body Project displayed an ED incidence of 0% over a 3-year follow-up versus 50% for assessment-only controls and 18% for those who completed alternative interventions. Completion of the Body Project also eliminated the negative effects of exposure to thin models on young females’ body dissatisfaction [34]. Finally, there is objective biological evidence that the Body Project reduced functional magnetic resonance imaging-assessed reward region (caudate) response to images of thin models, indicating that this prevention approach alters neural response to thin-ideal stimuli [49], which theoretically play a key role in the development of body dissatisfaction and EDs [50, 51], and implying that females who complete it may not perceive the thin body ideal as a desirable goal.

Conclusions and future directions

Over the past 2 decades, the ED field has made significant strides in successfully translating basic ED risk factor research into preventive interventions with documented effectiveness and efficacy in reducing ED risk factors and symptomatology, as well as future ED onset (i.e. a true prevention effect) in multiple randomized controlled trial. As noted, the staunchest and most empirical support has been garnered for dissonance-based ED prevention programmes, also known as the Body Project, the only intervention that has produced positive and replicated effects in multiple American trials, and significantly larger intervention effects than several credible alternative interventions through 3-year follow-ups. Consistent with the highlighted urgent need to broadly implement efficacious ED prevention programmes in Europe [2, 8], it is important that large-scale randomized controlled trials test the hypothesis that the Body Project reduces future onset of women’s EDs in Europe, translating principles that have proven effective in another (i.e. American) culture and addressing a major public health problem. It would also be useful for future research to use follow-up periods longer than 3 years (the longest follow-up to date), and other rigorous research methods such as comparing the Body Project with credible alternative control conditions and other ED prevention programmes with a broad evidence base (see [15, 23]) and use of fidelity checks, blinded interview assessment to assess EDs and objective biological outcomes (immune to the demand characteristics inherent in randomized controlled trials).

It is worth noting that there has been variation in intervention effects, with some randomized controlled trials showing that dissonance-based ED prevention programmes yield small effects (e.g. [41]) and other showing that that they yield large effects on ED outcomes (e.g. [52]). Elucidation of features related to larger effects is important, as it can inform implementation of optimally effective versions of these ED prevention programmes [53]. As such, it would be beneficial to carry out randomized experiments that manipulate the moderators that received the strongest support in a just published meta-analysis conducted by Stice and colleagues [53] (i.e. the number of facilitators and training hours, the number of dissonance-based activities in the ED prevention programme, and whether supervision is provided) to corroborate that they are causally associated with larger intervention effects. For example, a trial comparison of the Body Project when implemented by clinicians versus peer educators showed that peer-led groups yielded significantly greater reductions in future ED onset over a 3-year follow-up compared to clinician-led groups [43]. It would be beneficial to get experimental support for the thesis that combining the mentioned moderators should yield larger intervention effects [53]. It would also be beneficial to detect the types of females who subsequently develop an ED despite completing the Body Project so that alternative interventions can be designed for such females. Despite the encouraging evidence-base for the Body Project, we feel it is important to evaluate alterations that might enhance its efficacy, such as adding (e.g. dissonance-based) intervention elements that focus on reducing self-objectification (i.e. thinking and monitoring the body’s outward appearance from an external observer’s perspective [54,55,56]), as this extremely potent and newly emerged risk factor for DSM-5 ED onset [2], which fits within the well-supported dual pathway model of EDs (see [2, 56] for details) underpinning the development of the Body Project, is not addressed in its current version. Another direction would be to design additional ED prevention programmes that target distinct ED risk factors (e.g. impaired psychosocial functioning) [17], if we are to achieve the goal of decreasing the population prevalence of female EDs.

Given accumulated empirical evidence on males, body image disturbance and EDs [4, 57,58,59,60,61,62,63,64], scholars have recently examined the efficacy of an adapted version of the Body Project for males [65]. Results showed that young men with body image concerns/problems randomized to this adapted version of the Body Project reported meaningful reductions in ED risk factors (e.g. body-ideal internalization, dieting and body dissatisfaction) and symptomatology, with the effects persisting through 1-month follow-up year relative to waitlist [65]. Despite these promising results, replication trials conducted by independent groups are needed. We think it would also be useful to examine this adapted version of the Body Project for males against active control conditions and test whether it reduces future onset of threshold/sub-threshold EDs [61].

Finally, it would be beneficial to implement efficacious prevention programmes that affect multiple major public health problems, such as obesity and EDs [66,67,68,69,70,71], which appear to share certain risk factors (e.g. negative affect, reported caloric restriction, low self-esteem and body dissatisfaction) [72,73,74,75,76,77,78,79,80]. Although several prevention programmes have sought to reduce future onset of both EDs and obesity (e.g. [68]), only the Healthy Weight programme has successfully affected both outcomes (see below). In Healthy Weight, young people with body image concerns/problems make small, permanent healthy changes to exercise and dietary intake that bring energy intake and expenditure into balance, which should decrease weight gain and risk for EDs [81]. In support, randomized efficacy trials showed that the Healthy Weight programme decreased weight increases and ED symptoms and reduced obesity onset and ED onset relative to assessment-only control conditions and alternative interventions through 3-year follow-up (see [81, 82] for details). Perhaps most critically, a recent trial found that adding activities designed to create cognitive dissonance regarding overeating, a sedentary lifestyle and excess body fat to the Healthy Weight dual obesity and ED prevention programme improved efficacy [83]; the new dissonance-based Project Health prevention programme produced a 40% reduction in overweight/obesity onset relative to the Healthy Weight intervention and a 42% reduction in overweight/obesity onset relative to an educational video control condition over 2-year follow-up, and both Project Health and Healthy Weight interventions produced a 62% reduction in future ED onset over 2-year follow-up [83]. Despite these encouraging results, replication trials conducted by independent groups are needed. In addition, it would be useful for future research to test the potential of an Internet (and cost-effective) version of the Healthy Weight and new dissonance-based Project Health prevention programmes to reduce future ED and obesity onset. Finally, there is recent trial evidence that an Internet version of the noted Body Project showed a significant reduction in ED symptoms and body mass index relative to educational video control conditions and educational brochure through a 1-year follow-up [84]. Given accumulated evidence [25, 28, 32,33,34,35,36,37,38,39,40,41] that the Body Project yields greater reductions in factors that greatly increase the likelihood of both obesity and EDs (i.e. negative affect and body dissatisfaction) [72,73,74,75,76,77,78,79,80], it would be useful for future research to test the potential of this Internet version of the Body Project to reduce future ED and obesity onset. With continued research, we hope it will be possible to broadly implement programmes that prevent both excess weight gain (i.e. overweight/obesity) and EDs.