Influence of mental and behavioral factors on weight loss after bariatric surgery: A systematic review and meta‐analysis

Multiple factors are related to lower weight loss after bariatric surgery. This review and meta‐analysis evaluates the influence of several mental and behavioral factors on weight loss.


| Exclusion criteria
Studies that did not specify the type of bariatric-metabolic surgery or had unclear descriptions were excluded.Descriptive studies, case series, and case reports were also excluded because of their lower level of evidence.

| Study and data selection
Two reviewers, A.J. and V.M., independently conducted an initial screening of study titles and abstracts to determine their adherence to the inclusion and exclusion criteria.Subsequently, the same reviewers independently assessed the remaining full-text reports for eligibility.Data from full-text articles were extracted and subjected to double-checking.In cases of any discrepancies, consensus was reached through discussion between the two reviewers, with the availability of a third reviewer if required, though consultation was not necessary.Data pertaining to outcomes were collected and divided into separate groups for subsequent analysis.This included details regarding the type of surgery and duration of follow-up.Preoperative BMI was selected as baseline weight.When BMI was not provided, it was calculated from the mean baseline weight and mean height of the study population.Information regarding the methodologies used for assessing the moderating factors and the timing of these assessments (pre-or postoperatively) was extracted.Additional study characteristics such as the study design and the number of patients were also selected.Given the various methods for describing weight loss, data on all weight loss metrics were collected.The choice of outcome parameter for the subsequent meta-analyses, such as %EWL or %TWL, was determined by the availability of data and prioritized the parameter that was most frequently utilized in the included articles.
Authors of the studies were contacted at least twice to request any additional data required for the meta-analysis, such group sizes and standard deviations.In cases where studies did not present data for two distinct groups based on the moderator (opting instead for regression analyses), authors were contacted to acquire the necessary data for inclusion in the meta-analysis.

| Assessment of risk of bias
Two reviewers, A.J. and V.M., independently conducted assessments of the methodological quality and risk of bias for each included study.
The Newcastle-Ottawa Scale 48 was used to evaluate the quality of non-randomized studies, including cohort and case-control studies.This scale utilizes a scoring system with a maximum attainable score of nine points, distributed across three distinct domains: selection bias (four points), comparability (two points) and outcome bias (three points).The total scores were then categorized as high, medium, or low risk of bias, based on the number of points scored in each domain (Appendix 1).

| Data analysis
For each included study, patients were categorized into groups based on the presence or absence of specific moderators (e.g., patients with or without depression), in accordance with the definitions provided within the respective article (Tables 1-7).To minimize heterogeneity, separate meta-analyses were conducted for each type of bariatricmetabolic procedure and for distinct postoperative follow-up moments.Articles were only included if the standard deviation of follow-up durations fell within a range of less than 3 months.The mean difference in weight loss between groups was calculated using a random-effects model.Heterogeneity was evaluated by the I 2 statistic, for quantifying inconsistency.Interpretation of I 2 values was as follows: 0%-40% signified "might not be important," 30%-60% indicated "moderate heterogeneity," 50%-90% denoted "substantial heterogeneity," and 75%-100% represented "considerable heterogeneity." 49In cases where heterogeneity exceeded 60% (surpassing the threshold for "moderate heterogeneity"), the meta-analysis was omitted, and the relevant articles were solely described in the review.All statistical analyses were conducted using Review Manager version 5.4.1, 50and forest plots were generated.A p-value of <0.05 was considered statistically significant.

| Search results
After removing duplicates, 6408 unique articles were identified (Figure 1).Titles and abstracts of all 6408 articles were reviewed, leading to the exclusion of 6185 articles.Subsequently, 222 full-text articles were assessed for eligibility.Ultimately, 75 articles met the inclusion criteria for this review.Among these, 30 studies reported the effect of multiple moderators, as detailed in Tables 1-7.Fourteen studies provided adequate data for the conduct of at least one metaanalysis.

| Definition of weight loss
In the majority of the included studies, data on weight loss were only reported as %EWL.Consequently, %EWL was chosen as the outcome parameter for analysis.In cases where the mean and/or standard deviations of %EWL were not explicitly provided within the articles, these values were computed according to Cochrane standards to facilitate the analysis. 49

| Risk of bias
Out of the 75 articles included, 38 articles were classified as high risk of bias, 34 as low risk, and three fell within the medium risk category T A B L E 1 Overview of included studies that assessed compliance as moderator for weight loss after surgery.
Seven studies did not have sufficient data for a meta-analysis and were reviewed.Preoperative compliance had no significant correlation with weight loss in two studies. 51,61In three studies, postoperative adherence was associated with more weight loss and successful weight loss, 56,57,63 two studies found no difference, 51,61 and in one study, it was dependent on how weight loss success/failure was defined. 62One study demonstrated that attending support group meetings was associated with increased weight loss, whereas attendance to surgical follow-up appointments did not yield the same effect. 64Among these seven studies, one exhibited a low risk of bias.

| RYGB
A total of 30 studies evaluated the association between depressive symptoms and weight loss following RYGB.5][86][87][88][89][90][91][92]94,95 Other studies performed structured interviews based on the DSM-IV (Diagnostic and Statistical Manual of Mental Disorders) criteria, 28 or DSM-V criteria, 29 measured frequency and severity of symptoms, 70 or considered the use of antidepressants. 83It was unclear how depression was measured in five studies. 27,56,63,80,81The number of patients included in these studies ranged from 20 to 647, the mean BMI from 38.8 to 56.7 kg/m 2 , and the maximum duration of follow-up was 8.9 years.
Twenty-seven studies did not provide sufficient data for a metaanalysis and were consequently included in the review.One of these 27 studies had to be excluded due to the use of %TWL as outcome, or parameter, and despite multiple requests for additional information, the authors did not respond. 90In the context of preoperative ,86,[88][89][90] When assessing postoperative depressive symptoms, two studies reported lower weight loss among patients with depressive symptoms, 40,94 whereas five studies revealed no difference in weight loss outcomes. 41,76,82,95,96In one study, the impact of depressive symptoms on weight loss was found to be contingent on how successful weight loss was defined. 62Ten out of 28 studies were deemed to have a low risk of bias.
Meta-analyses involving three studies illustrated no significant difference in %EWL between patients with and without depressive symptoms before RYGB 26,30,88 (Figure 3A-C).Difference in %EWL ranged from 0.90% at 6 months to 2.56% at 3 months' follow-up.Heterogeneity (I 2 ) ranged from 0% at 3 months to 46% at 24 months.All three studies exhibited a low risk of bias.Due to high heterogeneity at 6-and 36-month follow-up (I 2 = 72% and 87%, respectively), these meta-analyses were excluded.

| SG
Six studies evaluated the relationship between depressive symptoms and weight loss following SG 27,91,93,[96][97][98] using four different questionnaires.Depressive symptoms were assessed prior to surgery in three studies 91,97,98 and post-surgery in two studies. 93,96In one study, there was a lack of clarity regarding the methodology and timing employed for the assessment of depression. 27The patient populations ranged from 34 to 300 individuals, the mean BMI from 42 to 60.4 kg/m 2 , and the maximum duration of follow-up was 4 years.
A single study provided sufficient data for a meta-analysis. 97Consequently, a meta-analysis was unfeasible due to the limited data availability.In two studies, preoperative depressive symptoms were found to have no impact on weight loss after SG. 91,97 However, one study found that depressive symptoms were associated with lower weight loss when assessed with the Beck Depression Inventory, although there was no relationship with weight loss when assessed with the Symptom Checklist-90-Revised. 98 Postoperative depressive symptoms were not associated with weight loss in two studies. 93,96e study with unclear measurement timing suggested that depressive symptoms were related to less weight loss. 27Two out of six studies had a low risk of bias.

| 3.6.1. RYGB
][106][108][109][110] Other studies performed a structured interview based on the DSM-IV criteria, 62,76,88,101,103,107 the DSM-V criteria, 29 did not specify a particular questionnaire, 31,67 assessed the frequency of binge eating, 70 or used a self-designed eating survey. 95These studies encompassed patient populations ranging from 32 to 497 individuals, mean BMI ranged from 44.4 to 56.7 kg/m 2 , and the maximum follow-up duration was 6 years.
Twenty-one studies lacked adequate data for inclusion in the meta-analysis, whereas one study possessed the requisite data for incorporation 108 ; however, this meta-analysis had to be excluded due to significant heterogeneity, necessitating the inclusion of the study in the review.Preoperative binge eating was related to reduced weight loss in seven studies, 29,63,87,88,104,107,108 and associated with increased weight loss in two studies, 65,67 while not showing a significant relationship with weight loss in eight studies. 31,43,70,91,100,101,103,106Postoperative binge eating was associated with less weight loss in two studies 42,95 and was not significantly associated with weight loss in two other studies. 76,109In one study, patients classified as successful (<30 kg/m 2 at 1-year post-RYGB) were less likely to report binge eating, although this trend disappeared when alternative definitions of successful weight loss were applied. 62Eight out of 22 studies exhibited a low risk of bias.
Meta-analysis including five studies showed that preoperative symptoms of binge eating were associated with greater weight loss at 24-and 36-month follow-up.The mean difference in %EWL was 7.97% (95% CI 2.75-13.20,I 2 = 0%) for the 24-month follow-up and 11.79% (95% CI 1.44-22.15,I 2 = 0%) for the 36-month follow-up (Figure 4A-D).No significant differences in %EWL were observed at 3 and 60 months.Due to high heterogeneity at 6-and 12-month follow-up (I 2 = 61% and 80%, respectively), these meta-analyses were excluded.Four out of five studies had a low risk of bias.
A meta-analysis including two studies illustrated that patients with postoperative binge eating symptoms experienced less weight loss compared to those without such symptoms.The mean difference in %EWL was À11.92% (95% CI À20.04 to À3.80,I 2 = 0%; Figure 5).
Both studies had a high risk of bias.

| SG
A total of five studies evaluated the association between binge eating symptomatology and weight loss after SG using validated questionnaires. 91,93,98,99,111These studies included between 34 and 117 patients, with mean BMIs ranging from 43.2 to 55.3 kg/m 2 , and had a maximum follow-up period of 4 years.
The available data were inadequate to conduct a meta-analysis.
Among these studies, one indicated that preoperative binge eating negatively impacted weight loss. 98In contrast, the other three studies reported no significant relationship between preoperative binge eating and weight loss. 91,99,111Furthermore, postoperative binge eating was not associated with weight loss in two studies. 93,111Four out of the five studies exhibited a low risk of bias.
The available data were insufficient for conducting a metaanalysis.In the case of RYGB, preoperative anxiety was found to be associated with reduced weight loss in only one study, 112 and in the F I G U R E 5 Meta-analysis of the association of postoperative binge eating symptomatology and %excess weight loss 12 months after RYGB.
context of SG, a single study reported that only the subscale of phobic anxiety had a negative impact on weight loss. 98However, the remaining studies, comprising 12 related to preoperative anxiety and three related to postoperative anxiety, did not demonstrate any significant predictive power for weight loss after RYGB or SG. 26,29,31,32,40,43,70,81,88,89,91,92,94,95Among these 16 studies, nine exhibited a low risk of bias.

| Body image
4][45] The number of patients ranged from 51 to 230, the mean BMI from 44.9 to 51.5 kg/m 2 , and the maximum follow-up duration reached 12 months.
The available data were insufficient to conduct a meta-analysis.
Two studies reported no significant association between preoperative body image and weight loss following RYGB. 43,44Similarly, no correlations were identified between the change in body image and weight loss after SG in the third study. 45All three studies had a high risk of bias.

| QoL
Four studies evaluated the impact of QoL on weight loss after RYGB, each study utilizing a different questionnaire 30,[40][41][42] (Table 7).The number of included patients ranged from 49 to 497, the mean BMI from 44.1 to 50.7 kg/m 2 , and the maximum follow-up period was 8 years.
There were insufficient data to conduct a meta-analysis.Among the two studies that evaluated preoperative QoL, one observed a positive correlation, where higher preoperative QoL was linked to increased weight loss at 5-and 6-year post-surgery, specifically for the physical health and pain scales. 30However, this relationship did not persist at 1-and 4-year follow-up. 30Conversely, the second study found no association between preoperative QoL and weight loss. 40[42] Two out of the four studies were classified as having a low risk of bias.

| DISCUSSION
This systematic review and meta-analysis aimed to provide a comprehensive overview of the mental and behavioral factors related to weight loss following primary RYGB and SG.The findings reveal that lower postoperative compliance and the presence of postoperative binge eating are associated with lower weight loss after RYGB.Additionally, preoperative binge eating symptoms are associated with higher %EWL 24 and 36 months after RYGB, while no discernible difference in weight loss was evident at 3 and 60 months.Conversely, no significant difference in weight loss after RYGB is observed when comparing patients with and without preoperative depressive symptoms.It is noteworthy that no meta-analyses could be conducted for preoperative compliance, PA, postoperative depressive symptoms, anxiety, body image, and QoL due to the lack of sufficient data.

| Compliance to follow-up
Recently updated international guidelines recommend increasing follow-up rates after bariatric-metabolic surgery, as it is associated with improved outcomes. 17This review and meta-analysis substantiate this recommendation: meta-analyses for all follow-up moments (up to 36 months after surgery) demonstrated that postoperative compliant patients achieved a higher %EWL compared to noncompliant patients.There was either moderate or no heterogeneity between studies, and all studies that were included in the meta-analysis had a low risk of bias, enhancing the comparability of studies and the validity of the conclusions.However, the data do not allow to draw definitive conclusions regarding the direction of this effect.It remains unclear whether adherence to follow-up appointments leads to more weight loss, or if patients with more weight loss are more likely to attend these appointments.A prior review suggests that follow-up rates tend to be lower in patients with less weight loss. 18Another plausible explanation is that motivated patients exhibit better compliance with consultations and lifestyle recommendations, which, in turn, results in more weight loss.This could result in a selection bias that may impact the study results.

| PA
The positive impact of compliance, especially in terms of higher weight loss among patients who engage in postoperative PA, aligns with the concept of compliance as a broader concept that encompasses various aspects of patient adherence.While this study did not provide sufficient data for a meta-analysis, most studies included in the systematic review highlighted a positive association between postoperative PA and weight loss after RYGB and SG.For instance, one study with a follow-up period of 2-5 years demonstrated a 15% greater weight loss in physically active patients. 75Regular PA is strongly recommended for individuals undergoing bariatric-metabolic surgery.Engaging in PA not only contributes to physical improvements such as weight loss, weight maintenance, enhanced cardiorespiratory fitness, and improved insulin sensitivity but also has favorable effects on QoL and other psychological outcomes. 21Therefore, consistent with previous guidelines, promoting PA should be a fundamental component of the care plan for all patients undergoing bariatric-metabolic surgery. 17The current meta-analysis has revealed an association between postoperative binge eating and reduced weight loss following bariatric-metabolic surgery.However, it is important to note that the relationship between preoperative binge eating and postoperative weight loss appears to be inconsistent across various follow-up moments.This variability can be attributed, in part, to the heterogeneity observed among the included studies.One source of this heterogeneity is the diverse array of self-reported questionnaires employed to assess binge eating symptoms.Self-reported questionnaires may not be sufficiently reliable for accurately diagnosing and assessing binge eating.Instead, a (semi-)structured interview is considered the gold standard for evaluating disordered eating patterns. 113Moreover, it is essential to recognize that the studies with long-term assessments (3-5 years) had relatively small sample sizes, ranging from 15 to 61 patients, 28,30,105 may introduce potential bias.To enhance the quality of research in this area, we recommend using standardized questionnaires to ensure more consistent and comparable outcomes.
Additionally, it is crucial to implement early detection strategies for postoperative disordered eating patterns and provide appropriate interventions to optimize patient outcomes.

| Depressive symptoms
No meta-analysis has been conducted to comprehensively assess the impact of depressive symptoms on weight loss following RYGB or any other bariatric-metabolic procedure.The findings from this study reveal that there is no discernible association between preoperative depression and weight loss at 3-, 12-and 24-month post-surgery.
However, considerable heterogeneity was observed at the 6-and 36-month follow-up moments, which ultimately led to exclusion of these meta-analyses.Only four out of 27 studies (which were not included in the meta-analysis) reported an association between preoperative depressive symptoms and weight loss following RYGB.These findings suggest that preoperative depressive symptoms are not associated with weight loss outcomes following bariatric-metabolic surgery.

| Anxiety symptoms
Fourteen out of the 16 studies that were included in the systematic review reported that symptoms of anxiety, either before or after surgery, were not significantly associated with weight loss following bariatric-metabolic surgery.These findings align with the results of another recent systematic review, which similarly concluded that there is no clear correlation between changes in BMI after bariatric-metabolic surgery and the presence of anxiety. 114Although the available data did not permit a meta-analysis in the current study, the collective evidence suggests that anxiety is unlikely to lead to reduced postoperative weight loss.Therefore, it is important to emphasize that patients with mood disorders, including depression and anxiety, should not be automatically denied from consideration for bariatric-metabolic surgery.

| Body image
All three included studies consistently revealed no significant association between preoperative body image or change in body image and postoperative weight loss.Notably, these studies had relatively brief follow-up periods, with a maximum of 12 months, and were found to have a high risk of bias.Given these limitations, it is not feasible to definitively determine the existence of a significant relationship between body image and weight loss outcomes following bariatricmetabolic surgery.

| QoL
Current review suggests that higher levels of postoperative, rather than preoperative, QoL are associated with higher weight loss after bariatric-metabolic surgery.However, it remains challenging to distinguish whether higher QoL leads to increased weight loss, or conversely, whether the weight loss achieved through bariatric-metabolic surgery results in enhanced QoL.This dynamic is complex, and it is worth noting that previous research has already well established that weight loss following bariatric-metabolic surgery is associated with improvements in QoL. 115,1168 | Treatment prior to surgery The impact of psychological factors on the outcomes of bariatricmetabolic surgery is complex and requires careful consideration.In accordance with international guidelines, it is common practice for patients with known or suspected psychiatric illness, such as severe depressive symptoms or binge eating, to undergo formal mental health evaluation before being accepted for surgery. 17It is crucial to acknowledge that the effects of psychological diagnoses on bariatricmetabolic surgery outcomes may vary between pre-and postoperative diagnoses.While preoperative treatments may positively impact patient outcomes, focusing solely on this phase fails to provide a comprehensive understanding.Therefore, the present review and meta-analysis separately analyzed pre-and postoperative psychological factors and therefore provides a more nuanced perspective on the role of psychological factors in bariatric-metabolic surgery.

| Risk of bias
Most studies exhibited a high risk of bias, primarily due to incomplete follow-up data and substantial baseline differences between compared cohorts.This disparity can be attributed to the fact that psychological factors cannot be randomized, which increases the likelihood of having different cohorts at baseline.To address this issue, case-control studies could be conducted, where patients with psychological disorders are matched with those without that disorder.It is also known that loss to follow-up rates are high among patients who have undergone bariatric-metabolic surgery, 117,118 as was affirmed in the current risk of bias assessment and could lead to data and results.To address this, prospective trials should be designed with a strong emphasis on achieving and maintaining higher follow-up rates.Despite these challenges, it is important to note that many of the included studies demonstrated a strong methodological quality with a low risk of bias, lending reliability to their results.

| Heterogeneity
The high heterogeneity observed in several meta-analyses can be attributed to the diverse methodologies used in the included studies, making direct comparisons difficult.To address this challenge in future research, the adoption of more gold-standard assessments and increased collaboration among researchers could enhance study comparability and reduce heterogeneity.

| Strengths and limitations
A significant strength of this study is the approach of conducting separate meta-analyses for each follow-up moment.Since weight loss after bariatric-metabolic surgery is strongly dependent on the time since surgery, this method allows for a precise examination of the factors that influence weight loss at different postoperative intervals.
Furthermore, RYGB and SG were analyzed separately, recognizing that these two surgical procedures lead to varying weight loss outcomes. 46However, due to the limited published literature concerning SG, the conduct of meta-analyses was only feasible for RYGB.In addition, 66 articles were excluded from this study because they did not present results independently for different types of surgery, for example, combined data for RYGB and laparoscopic adjustable gastric banding (Figure 1).
One of the limitations of this review and meta-analysis is that it solely focuses on weight loss as outcome parameter.While many studies primarily emphasize weight loss as the key outcome, it is crucial to question whether this is the most important indicator.Other outcomes, such as the resolution or improvement of associated medical conditions, medication usage, and QoL, as well as societal outcomes like absenteeism and premature death, may hold equal or even greater significance.Consequently, it is imperative to allocate more attention to these multifaceted aspects of bariatric-metabolic surgery in future research.Moreover, it is important to acknowledge that the predominant inclusion of qualitative studies (75 in total) in this study, compared to a smaller number of quantitative studies ( 14), may limit the robustness of the conclusions.Most of these studies were conducted in the past decade, a period when the use of %EWL as a standardized metric for weight loss evaluation was not as established as per current guidelines.This has inevitably led to a greater representation of qualitative research in our analysis.While qualitative studies offer valuable insights into patient experiences and perspectives, quantitative studies are typically lauded for their ability to yield more quantifiable and generalizable results.In light of this, future metaanalyses could enhance their methodological rigor by strictly adhering to contemporary guidelines for outcome reporting in bariatricmetabolic surgery, thereby ensuring a more balanced inclusion of quantitative data. 119Additionally, it is important to note that mental disorders were often diagnosed using self-report questionnaires.This approach is suboptimal for making precise diagnoses and may have introduced notable bias into the data and, consequently, the study's findings.Lastly, the presence of range restriction, wherein the signifi-

F I G U R E 2
Meta-analysis of the association of postoperative compliance and % excess weight loss after RYGB.(A) 6 months after RYGB.(B) 12 months after RYGB.(C) 24 months after RYGB.(D) 36 months after RYGB.

F
I G U R E 3 Meta-analysis of the association of preoperative depressive symptoms and % excess weight loss after RYGB.(A) 3 months after RYGB.(B) 12 months after RYGB.(C) 24 months after RYGB.

F I G U R E 4
Meta-analysis of the association of preoperative binge eating symptomatology and % excess weight loss after RYGB.(A) 3 months after RYGB.(B) 24 months after RYGB.(C) 36 months after RYGB.(D) 60 months after RYGB.

4. 3 |
Binge eating impact of bariatric-metabolic procedures on postoperative weight loss outcomes, coupled with the use of a dichotomous diagnostic variable, may have constrained the variability of our data.Consequently, this limitation could potentially obscure the detection of associations between psychological factors and postoperative weight loss, thereby influencing the comprehensive interpretation and generalizability of our findings within the larger context of the literature and clinical implications.To address range restriction, future research could adopt strategies to enhance the study's generalizability.These strategies include employing longitudinal designs with multiple assessment points in both pre-and postoperative periods, utilizing continuous (gold-standard) measures for psychiatric symptoms, and incorporating outcome measures beyond weight loss.5 | CONCLUSIONThis study aimed to comprehensively review and analyze the associations between several mental and behavioral factors and weight loss following bariatric-metabolic surgery.The literature reveals high heterogeneity between studies, particularly in the methods used to assess psychological factors, with a common reliance on self-reported questionnaires rather than the gold-standard assessments.Nonetheless, based on the findings of this study, a trend emerges suggesting that the presence of postoperative binge eating symptoms and lower postoperative compliance may be associated with less weight loss after bariatric-metabolic surgery.Additionally, preoperative depressive symptoms and binge eating do not seem to significantly impact weight loss.Predicting post-surgery outcomes solely based on preoperative mental and behavioral factors is challenging.Therefore, decisions regarding a patient's eligibility for bariatric-metabolic surgery should not be based on a single psychological diagnosis or questionnaire alone.Rather, a comprehensive evaluation conducted by a multidisciplinary team, which includes a mental health professional, should be the standard.Early detection of postoperative binge eating symptoms is advised for, as this seems to be associated with lower weight loss.
The primary factor contributing to a high risk of bias in most articles was the utilization of cohorts that were incomparable, resulting in only 22 out of the 75 articles earning both points in this domain.Additionally, inadequate follow-up was identified in 36 of the 75 studies included.The domain with the most common issue was the "selection of the non-exposed group," with most studies earning just one point out of a possible two (73 out of 75 studies).
Good Abbreviations: n.a., not available; RYGB, Roux-en-Y gastric bypass; SG, sleeve gastrectomy.aCompliance measured both pre-and postoperatively.bIncluded in meta-analysis.T A B L E 2 Overview of included studies that assessed physical activity as moderator for weight loss after surgery.
Overview of included studies that assessed anxiety symptoms as moderator for weight loss after surgery.
Poor Abbreviations: PA, physical activity; RYGB, Roux-en-Y gastric bypass; SG, sleeve gastrectomy.aPAmeasured both pre-and postoperatively.T A B L E 3 Overview of included studies that assessed depressive symptoms as moderator for weight loss after surgery.T A B L E 3 (Continued) Negative Poor Abbreviations: n.a., not available; RYGB, Roux-en-Y gastric bypass; SG, sleeve gastrectomy.aDepressionmeasured both pre-and postoperatively.T A B L E 4 Overview of included studies that assessed binge eating as moderator for weight loss after surgery.Fair Abbreviations: BED, binge eating disorder; n.a., not available; RYGB, Roux-en-Y gastric bypass; SG, sleeve gastrectomy.aBED measured both pre-and postoperatively.T A B L E 5 a Anxiety measured both pre-and postoperatively.T A B L E 7 Overview of included studies that assessed quality of life as moderator for weight loss after surgery.Positive Good Abbreviations: n.a., not available; QoL, quality of life; RYGB, Roux-en-Y gastric bypass; SG, sleeve gastrectomy.aQoL measured both pre-and postoperatively.T A B L E 6 Overview of included studies that assessed body image as moderator for weight loss after surgery.