Emotion regulation training in the treatment of obesity in young adolescents: protocol for a randomized controlled trial

Recent prevalence rates collected by the World Health Organization [150] revealed that, worldwide, 18% of the school-aged children and adolescents (5–19 years old) are diagnosed with overweight and 8% with obesity. These prevalence rates have tripled since 1975, demonstrating the growing problem of pediatric overweight and obesity [150]. This causes a great concern for both the individual and the community, as obesity has high medical and psychosocial consequences, both in the short term and long term [153].

The well-researched and currently golden standard treatment for childhood obesity is Multidisciplinary Obesity Treatment (MOT). The overall goal of the treatment is to improve the distorted balance between the energy intake and energy expenditure by changing the diet and increasing the physical activity [29, 106]. MOT incorporates cognitive behavioral therapy with active involvement of the parents to support behavior change and enhance the psychosocial well-being [29, 106]. Positive effects of MOT have already been found on weight (body mass index, BMI − 0.5 SD), physical fitness, and psychosocial well-being [104, 113]. However, the long-term effects of MOT are still limited, as weight-loss maintenance seems very hard to attain [4, 29, 96, 103, 151].

In sum, the high prevalence rates of childhood obesity; the medical, psychological, and economic consequences; and the poor long-term treatment effects emphasize the need to study in depth all mechanisms leading to childhood obesity. Herein, a psychological perspective recently has gained attention. An increasing number of studies point to “stress” as an important factor in the development and maintenance of childhood obesity [8, 74, 79, 110]. For example, families with an obese child experience more stress [102], and daily hassles seem to have a significant impact on children’s eating behavior [44, 110]. However, although the contributing role of stress in obesity is well studied, treatment options have not been developed in response. Therefore, the present study will bridge this gap by evaluating the role of treatment in an RCT study.

Stress occurs when a person perceives (goal-related) danger, attaches personal significance to the situation for his or her well-being, and the available coping resources of the person fail to deal with the situation [88]. Next to the cognitive and emotional component, stress also contains a physiological component [88]. The latter can be seen as a related but different indicator of a stress experience [143], reflecting the physiological arousal, which prepares the person to cope with the stressor by fighting or fleeing, and the restoration of homeostasis. Two main physiological stress systems have been identified [36]. The first stress system is the hypothalamic-pituitary-adrenal axis with cortisol as the end product. The second stress system is the autonomic nervous system with the catecholamines, adrenaline, and noradrenaline as end products, for which heart rate variability (HRV) is used as a noninvasive biomarker to indirectly measure cardiac parasympathetic and sympathetic activity [91].

When stress is successfully managed, the “emotional homeostasis” will return, and in the long run, the emotional stability will not be fundamentally affected (see Fig. 1: eustress). Adversely, when the stress is not successfully managed, enhanced arousal and negative affects occur like chronic experiences of tension, danger, frustration, or rejection (see Fig. 1: emotional distress) [37, 70, 86]. Emotional distress is associated with psychopathology [39, 83, 97], medical diseases [125, 127, 129], and obesity [8, 42, 58, 79, 101, 123].

For a thorough understanding on how emotional distress may be associated with weight gain and obesity, three processes need to be considered (see Fig. 1). First, when stress is not successfully managed, chronic activation of the physiological stress tonus entails, which results in an elevated cortisol production. Chronic cortisol hypersecretion is directly associated with the accumulation of fat by stimulation of the fat cell growth and dysregulation of the lipolysis. Indirectly, the elevated cortisol level increases the appetite, calorie consumption, and consequently the adiposity [21, 36, 116]. Second, emotional distress triggers a low-grade inflammatory process, which occurs directly by elevating the cytokine production and indirectly by its association with less sleep, imbalanced diet, and less physical activity that subsequently induce inflammation [71, 89]. This inflammation process may stimulate other hypothalamic pathways, resulting in increased food intake and finally obesity by resistance to the appetite-reducing factors leptin and insulin [31, 32]. The third mechanism concerns emotional or stress-induced eating, defined as “overeating in response to emotions and stress” [8, 28, 48]. When experiencing stress, people aim to neutralize the stress-related arousal by eating, even in the absence of hunger, as this gives a warm and good feeling in the short term. Especially eating comfort food (e.g., highly palatable food which is rich in sugar and fat) is known as reinforcing [90]. Moreover, it initiates an automatic learning process [48, 90, 110], facilitating emotional eating as the preferred behavior when feeling emotionally distressed. This is observed in both normal weight and overweight adults and children [24, 107, 142]. As emotional eating can foster eating beyond the saturation point, it can lead to an increased calorie intake and to overweight or obesity over time [55, 84].

Importantly, the relationship between emotional distress and obesity is not unidirectional but a vicious cycle can be assumed: obesity itself will lead to further emotional imbalance (e.g., physical and psychological well-being), which is hypothesized to further induce (a) enhanced cortisol, (b) low-grade inflammation, and (c) emotional eating (see Fig. 1) [3, 28, 92, 119, 134, 146, 149]. Furthermore, excessive body fat directly maintains the low-grade inflammation that may lead to higher stress sensitivity and a maintained emotional imbalance, as inflammatory cytokines from excessive body fat can affect the brain and its stress-related neuro-circuitry, neuroendocrine activity, and neurotransmitters [34, 75].

The above-mentioned drastic processes, resulting from unsuccessfully regulated stress, highlight the question as to why some people seem able to regulate their stress, while others are unable to do so. Although it might be that some experience more stressors [102], it is also plausible that some are less capable of regulating stress-related emotions, leading to chronic stress experiences. This brings us to the hypothesis that emotion regulation is playing a crucial role in the vicious cycle of obesity (Fig. 1).

Emotion regulation (ER) refers to the actions by which individuals try to exert an influence on the emotions they have, when and how they experience and communicate them [67]. These actions implicate the use of several ER competences which refer both to ER abilities (e.g., demonstrating compassion) and strategies (e.g., cognitive reappraisal). ER is recently found to affect different psychological problems, such as depression, anxiety, chronic pain, and ADHD [9, 73, 109, 124], and is now recognized as a transdiagnostic factor playing a role in the etiology and maintenance of different kinds of psychological problems [64]. Based on the long-term effects on affect, behavior, and psychopathology, Aldao and colleagues [5, 6, 68] suggest a classification in adaptive ER strategies and maladaptive ER strategies. Especially for obesity, studies have found that the association between stress and weight gain is stronger when more maladaptive ER strategies are used, whereas ER seems to mediate the stress-eating relationship [8, 48, 144]. Interesting, recent studies also emphasize the need to research adaptive ER strategies, as these are protective factors [27]. Commonly, in the literature, three adaptive ER strategies, “cognitive reappraisal,” “acceptance,” and “problem solving” have been researched and are found to be associated with less psychopathology (2009). Moreover, the adaptive ER abilities “acceptance,” “support yourself,” and the strategy “analyze and reappraise” are found to be trainable with positive effects in young adolescents [147, 148]. However, next to learning these ER competencies, also the sequence of applying them when feeling stressed is important [95, 111]. Therefore, Berking and Whitley [18] combined all these theoretical insights on ER competencies into a conceptual theory and training model on ER in adults, called Affect Regulation Training (ART). The ART model is in accordance with the recent literature on ER flexibility, referring to the importance of momentary flexibility in the use of different ER abilities and ER strategies across different stressful contexts taken into account the personal goals [7, 23]. Interesting, ART, both as a stand-alone intervention and on top of care as usual, is shown effective in clinical and nonclinical adult populations [15, 17, 18, 60]. Currently, our research group is evaluating the ER sequence, based on Berking and Whitley [18], in a prevention program for children and adolescents in schools [147], but it has not yet been evaluated in the context of emotional eating in a sample of obese children. However, given the evidence for the mediating role and the trainability of ER, we assume that an ER training will have good effects also for obese children. Therefore, in the preparation phase of the current RCT study, the feasibility of learning ER abilities and adaptive ER strategies in a population of obese youngsters has been investigated and well evaluated (Boelens, Debeuf, Verbeken, Volkaert & Braet: A Two-Hour Emotion Regulation Workshop in Early Adolescents with Obesity: A Feasibility Study, in preparation).

As in the presence of stress, flexible and adaptive ER is a requisite to achieve eustress and avoid the detrimental pathways to weight (re)gain and obesity, an ER training may empower the current MOT and foster long-term weight loss maintenance (see Fig. 1). The present study will therefore examine the effectiveness of a newly developed 12-week ER group training (one training session of 2 hours per week) on top of the MOT by means of a randomized controlled trial (RCT) in an inpatient treatment center for childhood obesity. The addition of an ER training to the MOT compared with MOT “as usual” is hypothesized to result in 1) an improvement in ER competencies (assessed by means of self-report questionnaires, idiosyncratic measures, and validated with physiological parameters during a lab task) (=primary outcome); 2) less emotional eating behavior (assessed by means of self-report questionnaires and in a stress paradigm); 3) better sleep quality (assessed by means of self-report questionnaires); 4) improved weight-loss and weight-loss maintenance (in terms of adjusted BMI and waist circumference) with associated change in metabolic parameters; and 5) better long-term (6-month) psychological well-being (assessed by means of self– and parent report questionnaires). The protocol was registered at ISRCTN Registry with study ID: ISRCTN 83822934.

Participants will be recruited in an inpatient treatment center for obesity in Belgium, Het Zeepreventorium, between March 2018 and September 2020. During consecutive waves, recruitment takes place at intake (T0; 4 months before starting the inpatient treatment) and both the youngsters and one of the parents should participate in the study. Inclusion criteria of the study are (1) 10–14 years old at the start of the inpatient treatment; (2) primary obesity, with minimum 60% overweight at intake in the treatment center according to the overweight index of Cole, Bellizzi, Flegal, and Dietz [38]; and (3) language proficiency, with mastery of the dominant language.

The Ethics Committee of the Ghent University Hospital approved the study design, procedure, and data collection. The national laws and the Helsinki Declaration of 1964 will be applied in all data collection procedures.

The study design concerns a two-arm randomized controlled trial, evaluating the effects of an emotion regulation training intervention versus a care-as-usual control group. Assessments in the two conditions are planned at four time points: after recruitment (T0), before the start of the intervention (T1), after the end of the intervention (T2), and at 6-months’ follow-up (T3). Primary outcome measures are emotion regulation and eating behavior, more specifically emotional eating. As secondary outcome measures, (1) weight change, (2) psychological well-being, and (3) sleep quality will be taken into account.

After giving informed consent (T0), participants and their parents will fill out questionnaires assessing the psychological well-being, ER capacities, and perceived stress of the child (parent- and self-report), which will be used for descriptive purposes or as covariates. Next, before the start of the intervention (T1), at the end of the intervention (T2), and at 6 months’ follow-up (T3), participants will fill out questionnaires on their psychological well-being, sleep quality, ER capacities, eating behavior, and perceived stress level, and they will participate in a stress paradigm. On top, at T1, the participants will also complete the Standard Progressive Matrices test to determine their overall cognitive ability. Parents of the participants will be asked again to fill out questionnaires on the psychological well-being and the eating behavior of their child at T2 and T3. Moreover, the participants in the intervention condition will be asked to fill out a diary three times a week on their stress level, affect, ER capacities, and emotional eating behavior between the ER training sessions. Filling out a diary enables a more momentary inspection on the relationship between the stress level, ER capacities, and emotional eating behavior in the naturalistic environment of the participants.

The eligibility, allocation, and assessments are illustrated in Figs. 2 and 3.

Stratified randomization is a two-stage process, used to control for the influence of covariates (baseline characteristics), in the randomization process [135]. In the current RCT study, both sex and age are identified as potential and important covariates. Therefore, separate blocks for each combination of the covariates will be generated (e.g., girls 10–12, boys 10–12, girls 13–14, and boys 13–14).

Participants who give consent to participate the study will be assigned to one of the four blocks.

Afterward, simple randomization to one of two conditions (e.g., experimental vs. care as usual condition) will be applied within each block at the individual level using dice [82]. The dice is thrown by the same researcher in all consecutive waves. Based on our a priori power calculations that 140 participants have to be included, we will recruit participants until each of the blocks has 35 participants.

We will control for multiple variables that are known as important but without specific hypothesis regarding their effect on our outcome variables, e.g., eating disorders, medication use, smoking behavior, puberty state, alexithymia, and cognitive functioning. First, eating problems, such as loss of control and binge eating are often observed as comorbidities of obesity [59, 152], associated with more eating-related and internalizing psychopathology [59]. Second, medication use is associated with a lower heart rate variability [112]. Third, smoking behavior can be a confounder in the relation between stress and weight, as it is a maladaptive ER strategy. Moreover, smoking can influence the energy imbalance [154]. Fourth, obesity is associated with early puberty onset, initiated by high levels of leptin [131]. Besides, the pubertal development influences the stress regulation by changes in the cortisol axis [94]. Fifth, alexithymia is a construct referring to difficulties in identifying and describing emotions [136]. In addition, alexithymia is associated with difficulties in discriminating between different emotions [10] and with difficulties in coping with stressful events [115]. Sixth, cognitive functioning, more specifically executive functioning, is shown to play a role in the development and maintenance of overweight and obesity [57, 65].