Introduction

Obesity is recognized as a major public health concern due to its high prevalence and its short, mid-, and long-term impact on general health status [1, 2]. According to the most recent estimations from the World Health Organization, more than 1.9 billion adults are overweight worldwide (including 650 million who are obese). This prevalence increased three-fold from 1975 to 2016 [3]. At the present time, bariatric surgery appears to be the most effective treatment for obesity, compared with conventional weight-loss interventions, such as nutritional counseling, exercise, and pharmacological treatments [4, 5]. Depending on the surgical procedure performed, patients can lose up to 40–75% of their excess body weight within three years of the operation. Bariatric surgery has thus become highly popular. Recently, the number of bariatric surgeries had doubled worldwide, totaling 344,221 operations in 2008 and 685,874 in 2016 [6]. Therefore, bariatric surgery is a public health concern which involves critical health challenges.

The first reports on the efficacy of bariatric surgery revealed that patients achieve significant weight loss associated with a positive impact on their obesity-related physical and psychological co-morbidities [7,8,9]. Unfortunately, one and a half years after surgery, some patients regain weight [10, 11]. It is estimated that up to 30% of patients regain weight after bariatric surgery [12, 13]. Recent studies investigating potential mechanisms underpinning this postoperative weight gain have found there are several risk factors predictive of poor physical and psychological well-being postoperatively. These factors include return to an imbalanced diet [14], little or no physical activity [15], and psychological disorders including eating disorders [16]. These factors must be taken into consideration when assessing the success and efficacy of bariatric surgery. Indeed, the surgical intervention, which enables the patient to reduce food intake, does not resolve the patient’s emotional issues (emotional eating), nor eliminate other disorders such as depression, body image dissatisfaction, or diminished self-esteem [17,18,19,20]. Moreover, recent studies have even suggested that, physiologically, bariatric surgery per se, which alters the action of certain hormones, might explain the difficulty in sustaining weight loss and, in certain cases, even induce renewed weight gain [21, 22]. Numerous physiological effects—for instance changes in neuroendocrine gut-brain signaling pathways that impact appetite, energy expenditure, and food choices [23, 24]—have been observed after bariatric surgery. Certain hormones (leptin, insulin, adiponectin,…) that favor nutrient malabsorption by enteroendocrine cells and decrease energy expenditure, and consequently facilitate weight loss, would be involved. These appetite regulator hormones have a sustained effect on afferent gut-brain vagus nerve function, modulating neuroendocrine status after bariatric surgery. The overall effect is a modified energy intake/expenditure balance [25, 26]. In summary, the regulation of body weight depends on the complex interplay between gut, brain, and other organs involved in energy metabolism.

Better knowledge of factors favoring or hampering successful outcome before and after bariatric surgery is needed to elaborate evidence-based management practices for obese patients.

The positive impact of psychological programs, before and after bariatric surgery, is well known. It is a long-term effect, demonstrated in comparison with control groups without medically managed psychological support [27,28,29]. Recent studies have investigated the efficacy of bariatric surgery by focusing on certain risk factors such as eating disorders or signs of depression. A meta-analysis has shown that when delivered postoperatively, group cognitive behavioral therapy (CBT) has a positive effect, especially in terms of weight loss percentage. Due to differences in analytic strategies and weight loss units, 95% confidence intervals (CIs) of standardized mean differences were calculated to compare weight loss in both groups. A significant difference in weight change favoring intervention over no intervention was found with a standardized mean difference of 1.6 (95% CI = 0.8, 2.4), Z = 4.0, P < 0.01, until 12 months [30]. Many studies recommend CBT as the treatment of choice for obesity care and eating disorders [31, 32]. CBT would thus be a pertinent tool for optimizing both medical and psychological outcome after bariatric surgery. In this perspective, recent studies have emphasized the efficacy of telephone CBT sessions, a method that would have therapeutic impact equivalent to that of face-to-face group sessions for the treatment of depression, anxiety, or eating disorders [33, 34]. This might be an appropriate therapeutic intervention for those individuals unable to participate in in-person treatment and a potentially effective alternative for those with limited mobility due to the severity of their obesity [35].

Nevertheless, although CBT has been demonstrated to produce positive results when delivered as part of a global program for obesity management, all patients do not benefit from a sustained effect [36]. We hypothesized that the same pattern would be observed when CBT is delivered within the framework of bariatric surgery.

The purpose of this work was to review the literature published from 2003 to 2018 in order to identify CBT programs currently conducted in the context of bariatric surgery (follow-up or preoperative preparation).

Method

Sources and Search Strategy

We used the Academic Research Complete database, interrogating Academic One File, ERIC, FRANCIS, MEDLINE, PsycINFO, and ScienceDirect. The following keywords and Boolean operators were used: (bariatric surgery) AND (cognitive behavior therapy OR cognitive behavior therapy). We limited our search to studies published in English during the period 2003–2018. We decided to start in 2003 because the first studies investigating the relationship between CBT and bariatric surgery were published in that year.

Screening

Criteria used to include or exclude publications are presented in Table 1. We first examined the abstracts retrieved to eliminate dual citations and articles not corresponding to the inclusion criteria. We then read the entire article in order to again eliminate any publication not corresponding to the inclusion criteria. Our screening (Fig. 1) identified eleven articles.

Table 1 Inclusion and exclusion criteria for screening the literature.
Full size table
Fig. 1
figure 1

Flow diagram

Full size image

This small number of studies retained for a systematic review and their relatively short duration can be explained by several factors. First, the literature on this topic is quite recent. As well, few studies have examined, beyond the aspect of weight loss, the impact of obesity surgery on physical and psychological well-being. It should also be noted that many of these studies had a very high drop-out rate after surgery, hindering longitudinal analysis. Factors associated with attrition from bariatric aftercare in patients remain to be fully described. Certain authors reviewing the literature have reported that patient adhesion after surgery is affected by several factors including the amount of weight loss, patient age, surgical technique, medical and psychological co-morbid conditions, and distance from home to hospital [37, 38].

Data Extraction and Summary

The eleven articles identified for analysis are outlined in Table 2. Data are presented in a homogenous manner (number of patients, age, profiles). The characteristic features of the therapeutic interventions (strategies, themes discussed, duration of the program, number of sessions, program details) are also developed. Study protocols and published results (objectives, variables recorded, questionnaires used, data collection methods, results, authors’ conclusions) are presented in Table 3.

Table 2 Bariatric surgery populations reported in the literature: characteristics and interventions performed
Full size table
Table 3 Study design and results from ten articles retained in this review of the literature of cognitive behavioral therapy (CBT) in the context of bariatric surgery
Full size table

Results

General Data

The selected studies reported data from 906 patients, each of the eleven studies including 8 to 243 patients, 79.0 ± 68.4 (m ± SD). The sample size was 100 patients or more for three studies. Female gender predominated (mean 78.1% women) in all studies. Baseline body mass index (BMI) ranged from 34.1 to 53.1 kg/m2 (43.4 ± 5.89). The studies enrolled patients from Canada, Norway, Spain, and the USA.

Among the 11 studies examined here, five detailed the surgical procedures [45,46,47,48,49]. We noted that most patients (80–100%) had bypass surgery, while others underwent a sleeve intervention (2.8–20%), a revision procedure after a first bypass (11.1% in one of the studies cited in [48]), and a gastric ring procedure (5.6% in one of the studies cited in [48]).

Among these eleven studies, CBT was conducted before surgery in seven [39, 40, 42,43,44,45, 47], before and after surgery in one [49] and after surgery in four [41, 42, 46, 48]. The study groups were randomized in six studies [43,44,45, 47,48,49]. All studies were longitudinal (before/after tests for six studies) with follow-up assessments up to 4 years after surgery. Only one study conducted an individualized program [43]. Off-site sessions conducted by telephone or videoconference were used in six studies [42,43,44,45, 47, 48]. The number of therapy sessions ranged from 4 to 17 (8.6 ± 3.70). Although identified using keywords targeting CBT programs, three studies included in our analysis also conducted interventions involving nutritional education or adapted physical activities [39, 48, 49].

In general, all studies, except one [49], demonstrated that programs based on emotional, cognitive, and behavioral therapy interventions were effective. The study outcomes are summarized in Table 2.

Cognitive-Behavioral Interventions Targeting Eating Behavior

All eleven studies included in this review conducted CBT interventions targeting eating behavior and eating disorders (defective interoceptive consciousness, disrupted physiological signs of hunger and satiety, eating as an emotional regulator, eating disorders such as binge eating, etc.) as themes for discussion. Six studies demonstrated a significant decline in eating disorders (binge eating, emotional eating, cognitive restriction plus binge eating) in patients who had participated in the CBT group [39, 40, 42,43,44, 46]. Altogether, the main objective of the interventions in these eleven studies was to reduce the intensity or frequency of dysfunctional eating behavior. Among the six studies demonstrating a significant reduction in dysfunctional eating behavior, four were able to demonstrate an improvement in patients with a diagnosis of binge eating disorder (BED) [39, 40, 42, 43]. In particular, one study enrolling 110 patients found that the probability of losing more than 50% of excess weight was higher among those who had participated in twelve CBT sessions preoperatively [39]. One randomized controlled trial also reported that ten CBT sessions in preparation for surgery, compared with usual care, led to improved eating behavior in participants, with reduction in symptoms of depression and anxiety during the preoperative period [44]. Among the programs that did not demonstrate changes in dysfunctional eating behavior, one conducted work on eating behavior in its program, but finally did not evaluate eating behavior in its methodology [49]. Consequently, any conclusion concerning a real lack of efficacy of the CBT on changes in eating behavior is precluded for this study. For the other studies [45, 47, 48], the participants were surgery patients who had undergone a bariatric procedure more than 3 years earlier. In two of these studies [45, 47], the authors assumed that the weight loss and changes in eating behavior certainly resulted from the surgery per se, and not from the conducted CBT intervention. They concluded that despite the implementation of the CBT program, before or after surgery, the effects of surgery predominated during the first postoperative years, influencing outcome in terms of eating behavior. Other authors reported the same finding [48], demonstrating that the frequency of compulsive eating in the operated participants (experimental group compared with control group) did not differ in terms of weight loss. Nevertheless, these authors did demonstrate that participants presenting signs of depression at the beginning of the study achieved better weight loss results when they participated in the CBT program.

Consequently, it appears that CBT interventions focusing on eating behavior also have a positive impact on depression. This is reported by six studies in this review of the literature. Among these six studies, five reported therapeutic work on emotional regulation (e.g., relaxation or meditation) in order to cope with eating disorders. It would thus appear that working on emotional regulation as part of a CBT program is an effective way to improve obesity-related depression and anxiety. This was notably observed in three studies [44, 45, 47]. These articles revisited results of a longitudinal study that conducted CBT before bariatric surgery. Measures were taken at four time points: pre-test before the CBT program (T0), post-tests after the program (T1), at 1 year (T2) and at 4 years (T3) after surgery. In this longitudinal study, the program did not propose any specific strategy aimed at reducing symptoms of depression or enhancing emotional control. The results nevertheless demonstrated a reduction in depression-anxiety symptoms at T1 as compared with the control group. These results were sustained for 1 year, with the reduction in signs of depression persisting at T2, but not at the final time point (T3) 4 years after surgery. However, patients with symptoms of depression showed a significantly lower BMI in case of completion of the CBT program, as compared to the control group.

Cognitive Behavioral Interventions Targeting Body Image and Self-Esteem

Only two of the eleven studies retained for this review of the literature conducted interventions concerning body dissatisfaction and self-esteem [40, 41]. The first study [40] enrolled female bariatric surgery candidates. The goal was to focus on a certain number of factors involved in sustaining dysfunctional eating behavior such as compulsive eating (e.g., weight-loss expectations, and relationship between weight loss, mood, and personal values). Although the CBT interventions included work on reducing body image dissatisfaction and improving self-esteem, there were no specific measures to evaluate these two dimensions. Evaluations were based only on the presence of eating disorder episodes (e.g., BED) and their frequency as reported weekly by the participants. This 4-week intervention enabled candidates for bariatric surgery to reduce the frequency of dysfunctional eating episodes. The second study by Beaulac and Sandre [41] also targeted body image dissatisfaction and self-esteem. This was a post-operative study, conducted one and a half months after bariatric surgery. Participants, predominantly female (88%), attended eight group therapy sessions working on relational problems and depressive mood. The post-test evaluation showed that the level of psychological stress had improved, particularly the adjustment to body weight. There was however no assessment of satisfaction with body image.

Cognitive Behavioral Interventions Targeting Motivation

Four of the studies reviewed [40, 42, 43, 48] conducted working on motivation. The main arguments put forward for including this theme in the CBT program were to target attainable concrete personal goals for surgery, to explore ambivalent feelings about changes in lifestyle and to develop strategies for maintaining such changes in the long term. The objective was to prevent or remediate long-term weight regain after surgery. The participants in these studies were either candidates for bariatric surgery or postoperative patients whose weight loss after surgery was considered insufficient. Postoperative patients were eligible for the study if they had bariatric surgery 3 years before study enrollment and had lost < 50% of their excess weight before surgery [48]

The results showed an improvement in dysfunctional eating behavior (BED, emotional eating) in post-test evaluations in three studies [40, 42, 43]. There were no post-operative evaluations. Moreover, one of the studies [48] showed that greater weight loss was observed for participants in the CBT program (less than 3 years after surgery with minimal weight loss) than in the control population (patients with similar eating problems on the waiting list for surgery). This difference persisted up to 6 months after the end of the CBT program.

But these four studies did not assess the impact of these motivational interventions on lifestyle nor the patient’s capacity to sustain more adapted eating behavior in the long term.

Discussion

The purpose of this review of the literature was to study the efficacy of CBT interventions in the context of bariatric surgery.

The first observation that arises from the analysis of the selected articles is that participants in CBT programs experience significant changes compared with controls. More precisely, only one of the eleven studies failed to demonstrate a significant difference between the experimental group and the control group. Three studies demonstrated better weight-loss outcome in the CBT group. Improved eating behavior was reported by six studies, including four that demonstrated a reduction in a certain number of symptoms of bulimic hyperphagia, notably binge eating in patients who participated in the CBT program. Finally, six studies found that signs of depression and/or anxiety declined in the experimental group. Consequently, the development of CBT interventions during the care pathway for patients scheduled for or having undergone bariatric surgery appears to be a clinically pertinent proposal for improving outcome. Patient adherence to the CBT program has been relatively good in these studies that had four to seventeen group therapy sessions. Two studies evaluated patient satisfaction with the CBT program [46, 47] and found a very good satisfaction rate. Patients reported they appreciated the support provided by the group, feeling they were not alone, and learning about the experiences of others [47].

The second element highlighted in this review of the literature of CBT programs is that obesity is recognized as a chronic disease affecting an individual’s physical and psychological well-being [50]. Patients with obesity report significant dissatisfaction with their body image [51, 52]. Several studies have demonstrated that dissatisfaction with body image is associated with psychiatric disorders such as depression that favor the emergence and persistence of eating disorders [53, 54]. Other studies also present results showing that body image dissatisfaction is a risk factor for gaining weight after bariatric surgery [55, 56]. Moreover, an important decline in perceived well-being and quality-of-life was recorded in operated patients who reported they were satisfied with the medical outcome but nevertheless dissatisfied with their body image [57]. Despite these important issues concerning body image, only two studies in this review included therapeutic work on body image and self-esteem. Unfortunately, the results reported in these studies do not provide conclusive evidence concerning efficacy; none of the studies measured satisfaction with body image. This highlights the need for further long-term studies exploring the impact of strategies targeting body image satisfaction. It would be important to explore in detail the long-term impact of strategies designed to reduce body image dissatisfaction.

Altogether, the selected articles used weight loss, improvement in eating disorders, and reduction in anxiety-depression to assess the outcome of bariatric surgery. This same approach was used by earlier studies that used these same variables to demonstrate the impact of bariatric surgery [58,59,60]. It would be useful to also assess other variables in order to have a more global evaluation of the beneficial effect of CBT programs for patients scheduled for bariatric surgery [61]. Further studies on this topic should explore satisfaction with body image, self-esteem, and quality-of-life in patients participating in CBT programs within the framework of surgical care for their obesity.

In the continuity of the evaluation criteria used, we note that the studies in this review used a very heterogeneous set of methodologies, especially concerning the questionnaires used. The Eating Disorder Examination Questionnaire [62] was the most widely used, appearing in four of the eleven articles. This heterogeneity is also seen for the experimental design where the CBT programs were sometimes conducted before surgery, sometimes after. These methodological differences hinder interpretation of the overall results.

Elaboration of a harmonious and scientifically validated CBT program is a difficult challenge. For certain authors, programs have to be conducted because patients are less receptive after their surgery has been completed [63]. These authors also point out that developing the skills necessary to cope with stress or new situations takes time. Thus, early preparation would improve the chances of fully adapting to a new lifestyle in the postoperative period. But these are hypothetical considerations that remain to be tested. Other authors put forward conflicting results and conduct CBT interventions after bariatric surgery [49, 64]. Further study is thus needed to determine whether CBT should be conducted before and/or after bariatric surgery, and also to explore the clinical conditions this type of therapy should be used to treat.

Considering these observations, interventions based on CBT are increasingly popular for preparing patients for bariatric surgery or as part of their post-operative follow-up. Nevertheless, CBT is not routinely conducted as part of the care program for bariatric surgery patients, partly because medical and scientific consensus has not been reached. Randomized controlled trials should thus be undertaken to determine which criteria are essential for optimizing the long-term postoperative outcome after bariatric surgery.

Limitations

This review of the literature has certain limitations. First, a formal meta-analysis was not possible because the studies under review did not use equivalent instruments for data collection. We thus performed a descriptive analysis of eleven CBT programs.

Moreover, we focused our attention on programs specifically using the terms of cognitive and behavioral therapy, excluding by consequence certain interventions with similar strategies but not specifically called CBT. A further limitation arises directly from the studies themselves.

They examined small populations and did not provide long-term data. Moreover, none of the included patients received psychopharmacological treatment. Only four of the eleven studies examined checked for psychological or surgical contraindications for bariatric surgery such as severe depression or eating disorders. Finally, it is also noteworthy that there could be a placebo effect, which could also be termed a social desirability effect in this context.

It is generally accepted that surgery is an effective means of limiting food intake but also that it does not necessarily resolve the important issues of emotional eating, depression, body image dissatisfaction, or impaired self-esteem [20, 65, 66]. These issues are risk factors for weight regain and/or altered physical and psychological well-being, about one and a half years after bariatric surgery [67]. Most of the studies included in this review did not provide follow-up data beyond 1 year after surgery.

Conclusion

This systematic literature review provides preliminary evidence of the effectiveness of these CBT procedures. However, further studies are still needed to corroborate those findings. The beneficial effect of CBT interventions conducted within the framework of bariatric surgery suggests perspectives for future care to improve prognosis and optimize postoperative outcome. Nevertheless, we noted certain limitations of the programs presented in the available literature. The CBT interventions that appear to be the most effective are those involving therapeutic work on eating behavior, body image satisfaction, and motivation. Unfortunately, these variables have not been measured extensively. Consequently, no precise conclusion can be made concerning the impact of CBT programs, especially in terms of improving body dissatisfaction.

Further research is needed to consolidate the findings presented in previous publications, using different time points for evaluating longitudinal outcome beyond 2 years post-surgery and involving other evaluation instruments. Thus, the pertinence of CBT programs to optimize outcome after bariatric surgery could be confirmed. Future studies should also determine the most appropriate period for participation in a CBT program. The final goal is to develop a consensus on the most appropriate practices and to contribute to the elaboration of international guidelines in the field of bariatric surgery.