Relationship of depression, impulsivity, distress intolerance and coping styles with maladaptive eating patterns in bariatric candidates

The study aimed to investigate the problematic eating patterns and understand their relationship to psychological constructs, including stress intolerance, coping mechanisms and impulsivity, and psychiatric symptoms among bariatric surgery candidates. The bariatric candidates were evaluated by psychiatric interview and standard scales assessing maladaptive eating behaviors (Eating Attitudes Test (EAT), Bulimia Investigatory Test-Edinburgh (BITE), Dutch Eating Behavior Questionnaire (DEBQ)), depression (Beck Depression Inventory (BDI)), psychiatric symptoms (Brief Symptom Inventory (BSI)), and psychological constructs (Distress Intolerance Index (DSI), Coping Styles Scale (CSS), UPPS Impulsive Behavior Scale(UPPS)). More than half (57.8%) had maladaptive eating behaviors, and 23.6% had binge-eating behavior. Depression and anxiety predicted EAT, BITE, and DEBQ emotional and external eating sub-scale scores; distress intolerance, helpless coping style, and impulsivity predicted maladaptive eating behaviors in bariatric candidates. Maladaptive eating patterns play an essential role in the failure to lose weight and regain weight and are predicted by depression, anxiety, and psychological constructs in this study. Evaluation of pathological trait characteristics besides discrete psychiatric syndromes should be recommended in the pre-operation process to plan relevant interventions in the long-term management of weight. Level III, evidence obtained from well-designed cohort analytic studies.


Introduction
Bariatric surgery is an effective method for most patients with obesity; however, it is known that many individuals fail to lose weight or regain a substantial amount of the lost weight a few years after the surgery [1,2]. Reported weight regain or insufficient weight loss prevalence depends on the definition of these concepts and varies according to the procedures used; and generally reported to be between 20 and 40% [2]. The prevalence of failure to success is reported in a 1 3 wide range due to the differences in the definition of weight regain, different procedures used in bariatric surgery, and the post-op follow-up time, but one study reported that only 40% of surgery patients maintained at least a 30% weight loss after 12 years [3].
There have been efforts to identify the possible factors for the failure to lose weight and sustain weight loss in people with obesity. An important research area has been the prevalence and effect of problematic eating behaviors and psychopathology in obese patients on weight regain or insufficient weight loss in the long run.
Although obesity is not classified as an eating disorder, various disordered eating behaviors, including emotional eating, binge-eating behavior, and night eating syndrome, have been documented in obese individuals [4,5]. The prevalence rates of binge-eating disorder in individuals with obesity varied substantially due to different methods to evaluate this behavior but generally were in the range of 10 to 30% [6]. Some studies have shown that binge-eating behavior is related to the severity of obesity and psychopathology [7]. Maladaptive eating behaviors were identified as some of the factors associated with weight regain after bariatric surgery [8,9]. Bariatric surgery candidates have higher rates of eating problems and especially more severe binge-eating behaviors than obese individuals seeking traditional weight loss methods [10][11][12]. Prevalence of another dysfunctional eating behavior, emotional eating, defined as the eating behavior responding to negative emotions [13], was between 25 and 38% in some bariatric candidate groups indicating that emotional eating is also widespread in this group [14]. Therefore, it is crucial to understand which factors or mediators contribute to disordered eating behaviors in the bariatric population. Some mental health problems, such as depression and anxiety, and psychological constructs, including impulsivity, distress intolerance, and maladaptive coping mechanisms, could be among these mediators and relate to each other, contributing to disordered eating patterns.
Studies have revealed a relationship between obesity and mental health problems. Much of the work about the relationship between obesity, and psychopathology has focused on discrete psychiatric diagnoses, such as depression and bipolar disorder. Depression is the most studied and prevalent mental health problem in obese individuals [15]. The exact mechanisms linking obesity and depression were not clearly understood, but there have been hypotheses on the issue, as depressed individuals may develop obesity through dysregulated stress systems or unhealthy lifestyles and maladaptive eating behaviors, such as emotional eating. However, the direction of the relationship between obesity and depression is questionable. One meta-analysis reported that obesity also increased the risk of depression, while depression was a predictor of obesity [16]. Previous research has suggested that depressive symptoms and eating disorders, especially binge eating, are interrelated in people with obesity [17]. Some studies have reported high rates of psychopathology in bariatric surgery candidates and found them related to disordered eating behavior [18]. However, studies did not find a consistent association between eating disorders and general psychopathology [9]. The conflicting evidence on the relationship between pre-operative psychopathology and post-surgery success has also been revealed in a metaanalysis [15]. Therefore, identifying possible mediators of the psychopathology and eating disorders relationship in persons with obesity may provide valuable insight into determining the patients at risk. Marek and colleagues also argued that the focus on the relationship between discrete diagnoses and post-bariatric outcomes may have failed to explain psychological constructs shared across diagnoses; these shared psychological characteristics, such as impulsivity, emotion dysregulation, and the use of pathological coping mechanisms, might contribute to the maladaptive eating behaviors and obesity [19]. Thus, the mechanisms linking disordered eating behavior, general psychopathology, and obesity have recently been an important research area. Impulsivity and disinhibited behavior are known to be common in mood disorders, binge eating, and substance use disorders; lack of impulse control over food may contribute to weight gain and obesity. This common construct may explain the relationship between obesity and mental health problems, like disordered eating and mood disorders. Cox and Brode investigated disinhibition in a group of bariatric candidates and found that disinhibition was a critical mediator of the relationship between depressive symptoms and binge-eating in this group [20]. In another study on bariatric surgery candidates, emotional eating, depression, and all facets of impulsivity were significantly correlated with each other [21]. Schafer and colleagues have subtyped bariatric candidates based on regulative temperament, emotion dysregulation, and disinhibited eating. They found that different subtypes were associated with varying general and eating disorder psychopathology levels. Moreover, the patients with pre-bariatric deficits in self and emotional control had an increased risk of showing these deficits in the post-operative period [22].
Psychological stress and the use of maladaptive coping strategies used to deal with stress have been hypothesized to be among the factors contributing to maladaptive eating behaviors [23,24]. A review on this topic suggests that the way individuals respond to or cope with stressors is the best predictor of weight maintenance success, not the number of stressful events in patients with obesity seeking non-surgery treatments [25]. Coping mechanisms are the behavioral and cognitive strategies used to manage stressful conditions and crises. Coping skills are not considered inherently adaptive or maladaptive. Therefore, they depend on the context they are used. However, problem-focused coping is generally 1 3 associated with better adjustment, whereas emotion-focused and avoidant coping are generally conceptualized as maladaptive and associated with psychopathology, including eating disorders [26]. Generally, a difference was found to exist between people with or without an eating disorder regarding the coping strategies used: individuals with eating disorders demonstrate more emotion focused and avoidant coping responses than individuals without problematic eating behavior; thus, disordered eating may reflect some kind of an avoidant coping strategy [23]. However, while the literature is rich in coping styles for obesity and eating behaviors, only a few studies investigated coping in bariatric candidates. In one of these, McGarrity and colleagues reported that approach coping predicted less and avoidance coping predicted more disordered eating in a group of bariatric candidates [27]. Another study revealed that bariatric candidates used problem-focused coping [28]. Moreover, another study reported that bariatric patients with more psychological resilience used approach-coping mechanisms post operatively [29].
In sum, pathological eating behaviors are common in bariatric candidates, and pathological eating patterns could be related to psychological constructs and psychiatric symptoms. Despite the rich literature on the relationship between eating disorders and those mentioned above, the findings were mainly retrieved from adolescents from the community or the individuals with obesity. Studies on bariatric candidates were exceedingly limited and inconsistent. Therefore, this study aimed to investigate the problematic eating patterns and understand their relationship to some psychological constructs, including impulsivity, stress coping mechanisms, and tolerance to emotional strain and psychiatric symptoms, such as depression and anxiety, among a group of bariatric surgery candidates. We hypothesized that depression and maladaptive coping behaviors, stress intolerance, and impulsivity would be positively related to disordered eating behaviors in bariatric candidates. If the factors related to maladaptive eating patterns could be identified, effective interventions for patients with disordered eating behaviors would be developed, at least for the modifiable ones. This paper presents the results of the study sample's baseline assessment; this sample is being followed-up for the bariatric surgery results in a prospective study.

Participants
The population of this retrospective study comprises the patients who have applied for bariatric surgery. The Bariatric Surgery Board of University Hospitals has evaluated them since 2013, and their medical records were available for data extraction. As a rule, the Bariatric Surgery Board, composed of the physicians from Departments of Internal Medicine-Endocrine Department, General Surgery, Thoracic Surgery, Cardiology, and Psychiatry, evaluated all the bariatric candidates. For the board evaluation of all bariatric surgery candidates, two psychiatrists, a dietitian, and a clinical psychologist identified any psychiatric contraindications to the surgery, such as active psychosis, severe depression, suicidal behavior, and substance abuse. All bariatric candidates received brief psychoeducation on the dietary and behavioral challenges of the surgery. The candidates needing a more advanced intervention were provided with psychiatric care (medication, brief psychotherapy for 3-6 months especially targeting maladaptive eating behaviors) before the approval.
A total of 465 participants' hospital records were recruited. None of these candidates had an absolute contraindication to the surgery, and all participants were approved for the surgery. After excluding the patients who left blank more than 10% of the items of any scales selected to collect data for the procedure, the study sample was reduced to 410 patients. All the statistics were done for this group, except the presentation of sociodemographic data including all participants. University Ethics Committee approved the study.

Measures
Each bariatric surgery candidate referred to the Psychiatry Department by the Bariatric Surgery Board of the Hospital was evaluated for their suitability for surgery, with a psychiatric interview, various clinical scales, and a personality inventory recommended by the expert guidelines [30]. In addition to the scales, a clinical data form questioning the patients' body mass index, knowledge and expectations about surgery, psychiatric history, current mental complaints, and a sociodemographic data form were also administered.
Within the scope of this study, each patient was evaluated with standard scales assessing eating attitudes and screening for abnormal eating behaviors and symptoms of various eating disorders. Depression and anxiety symptoms and psychological constructs (impulsivity, stress intolerance, and coping styles) thought to be related to eating attitudes and eating disorders were also evaluated by standard scales.

Eating attitudes test (EAT)
EAT-40 is a 4-point Likert-type scale developed to evaluate eating behaviors and attitudes and to screen for possible abnormal eating behaviors in individuals [31], which is valid in Turkish [32]. The possible range of scores is 0-120, and an increase in scores indicates the abnormality in eating attitudes. The internal reliability coefficient is 0.75 for EAT in our sample.

Bulimia investigatory test, Edinburgh (BITE)
It is a self-report scale aiming to evaluate bulimic behaviors and binge-eating symptoms [33] and is valid in Turkish [34]. A score of 20 or more indicates binge-eating syndrome, 10-19 indicates abnormal eating behavior, and 0-10 indicates normal eating behavior. The internal reliability coefficient is 0.78 for BITE in this study.

Dutch eating behavior questionnaire (DEBQ)
It consists of a total of 33 items and three sub-scales measuring Emotional, External, and Restricted Eating Behaviors [35]. The internal reliability coefficient is 0.88 for DEBQ in this study. The scale was validated in Turkish [36].

Beck depression inventory (BDI)
It is a self-report scale consisting of 21 items in which physical, emotional, and cognitive symptoms observed in depression are evaluated, and increasing values correspond to increasing severity [37]. A validity and reliability study of the Turkish version has been conducted [38]. The internal reliability coefficient is 0.85 for BDI in this study.

Brief symptom inventory (BSI)
It is a 53-item short form of the Symptom Check List (SCL-90) [39,40]. The total score obtained from the scale varies between 0 and 212. The Turkish version comprises five sub-dimensions: anxiety, depression, negative self-concept, somatization, and hostility [41]. The internal reliability coefficient is 0.97 for BSI in our sample.

The UPPS impulsive behavior scale (UPPS)
The UPPS is a self-report scale with values between 1 and 4, Likert type scoring of 45 items [42]. Its four subscales are Urgency, Lack of Premeditation, Lack of Perseverance, and Sensation Seeking. Higher scores indicate higher levels of impulsivity. The scale was adapted to Turkish [43]. The internal reliability coefficient is 0.78 for UPPS in this study.

Distress intolerance index (DII)
The DII is a 10-item self-report measure designed to assess the inability to tolerate negative states [44], with higher scores indicating greater distress intolerance. The internal reliability coefficient is 0.90 for DII in our study. The index was adapted to Turkish [45].

Coping style scale (CSS)
The scale is designed [46] to measure the two main coping styles; effective style (problem-oriented) with three sub-dimensions of seeking social support, self-confident approach, and optimistic approach and ineffective style (emotion oriented) with two sub-dimensions of helpless and submissive approach. Individuals who can cope effectively with stress use the "self-confident" and "optimistic" approaches, while those who cannot cope with stress use the "helpless" and "submissive" approaches often. The Turkish validity and reliability study was conducted [47]. The internal reliability coefficient is 0.76 for CSS in this study.

Statistical analysis
All analyses were conducted with Statistical Package for the Social Sciences (SPSS) software version 18.0 (IBM Corp., Armonk, NY, USA). The normality of the data was checked with analytical (Kolmogorov-Smirnov test) and visual (histograms, probability graphs) methods. Descriptive statistics were expressed as means, standard deviations, median and minimum-maximum values for continuous variables. Cronbach Alpha coefficient was used to estimate the internal reliability of the scales. Correlation analyses between variables, correlation coefficients, and statistical significance were calculated with Pearson Product-Moment Correlation Coefficient Test.
Hierarchical multiple regression analyses were used to evaluate the independent effects of depression, anxiety, and the other psychopathology and psychological constructs, including impulsivity, stress intolerance, and coping styles on the scales assessing eating attitudes and abnormal eating behaviors. The statistical significance was set at a p-value < 0.05.

Sample characteristics
A total of 465 bariatric surgery candidates, 77.4% (n = 360) female and 22.6% (n = 105) male, were included in the study. A 71.6% of the sample had at least high school education. Most were married (n = 284, 61.1%), 25.2% single, 9.3% (n = 43) was either widowed or divorced. The participants' mean age was 37 BMI value of 50.00-59.99 kg/m 2 , and 18 (%3.9) with a BMI value of 60.00 kg/m 2 and above.

Abnormal eating behaviors
The total and sub-scale scores of the three eating scales (DEBQ, BITE, and EAT) and the scores of other scales used to evaluate psychopathology and psychological characteristics are in Table 1. A total of 92 persons (23.1%) had scores of 30 and above, implying abnormal eating behavior from EAT. Seventy-five participants (18.6%) had scores of 0-9 from BITE, while 233 participants (57.8%) had scores of 10-19 depicting abnormal eating behavior, and finally, 95 participants (23.6%) had scores of 20 and above, suggesting binge eating.

Relations between eating behaviors and psychopathology
Beck Depression Inventory mean scores were 11.42 ± 7.46 for the whole sample; 96 participants (23.4%) had BDI scores of 17 and above, indicating clinical depression.
Pearson Product-Moments Correlation Coefficients between eating behavior scales and psychiatric symptoms (BDI and BSI) are presented in Table 2, while the correlation coefficients between eating behaviors scales and tests of psychological constructs (COPE, DII, and UPPS) are presented in Table 3.

Eating test scores and psychiatric symptoms
The predicting effects of depression and other psychopathology first and the psychological constructs on the eating attitudes were examined separately for each eating test and their sub-scales using multiple regression analyses while statistically controlling for the sex and age of the participants. Therefore, these two variables were entered into the equation in the first step to control their possible confounding effect on the predictors.

3
A series of hierarchical multiple regression analyses were executed as the eating test scores assigned as the dependent variable (criterion), and the BDI total and BSI sub-scale scores assigned as the independent variable (predictor) while statistically controlling for the sex and the age of the participants at the first step. Then, the BDI score was entered into the regression in all the analyses due to its being an established and well-known variable related to depression, obesity, and abnormal eating behaviors from many studies. The four sub-scale scores of BSI, except the depression sub-score, were entered into the regression at the next step. Regression analysis results are depicted in Table 4.
As can be seen from the Table, after controlling for sex and age of the participants, the BDI scores were significant (ΔR 2 = 0.17, ΔF (1355) = 73.95, p < 0.001) in predicting the EAT score as the dependent variable. The other variables entered into the regression at the third step did not have predictive power on EAT scores.
For the regression analysis of the DEBQ external eating sub-scale assigned as the dependent variable, BDI score entered into the regression at the second step significantly (ΔR 2 = 0.05, ΔF (1,363) = 19.81, p < 0.001) predicted the dependent variable at a 5% level. Only the BSI's negative self-concept sub-scale score entered the regression at the third step significantly (ΔR 2 = 0.03, ΔF (1,362) = 12.06, p < 0.001) predicted DEBQ external eating scores beyond depression scores' contribution.

Eating test scores and psychological constructs
A series of hierarchical multiple regression analyses were executed as the eating tests' scores were assigned as the dependent variables (criterion), and DII, CSS sub-scale, and UPPS sub-scale scores were assigned as the independent variables (predictor) while statistically controlling for the sex and the age of participants at the first step. The results are presented in Table 5. Table 3 Correlations between eating behavior tests and psychological constructs   As seen from the table, in the analyses with EAT scores as the dependent variable while controlling for sex and age of participants, CSS-helpless approach (ΔR 2 = 0.08, ΔF (1,295) = 26.38, p < 0.001), UPPS-lack of perseverance sub-scale (ΔR 2 = 0.03, ΔF (1,294) = 8.20, p < 0.01) and DII (ΔR 2 = 0.03, ΔF(1,293) = 8.48, p < 0.01), significantly predicted the dependent variable.
In regression analyses with DEBQ-external eating subscale as the dependent variable, U P P S -u rg e n c y s u b -s c a l e ( Δ R 2 = 0 . 0 8 , Δ F (1,302) = 24.82, p < 0.001) and UPPS-lack of premeditation sub-scale scores (ΔR 2 = 0.01, ΔF (1,301) = 4.57, p < 0.05) had a statistically significant predictive power on the dependent variable in negative and positive ways subsequently.

Eating test scores, psychological constructs, and psychiatric symptoms
A series of hierarchical multiple regression analyses were executed as the eating tests' scores were assigned as the dependent variable (criterion). The results are presented in Table 6. Some scores of psychological constructs and psychiatric symptoms had predictive power on the eating test scores in previous analyses assigned as independent variables (predictor) while statistically controlling for the sex and the age of participants at the first step.
As seen from the Table, in the analyses with EAT scores as the dependent variable while controlling for sex and age of participants at the first step, UPPS-lack of perseverance sub-scale (ΔR 2 = 0.03, ΔF (1,341) = 14.42, p < 0.001) and DII (ΔR 2 = 0.02, ΔF (1,340) = 6.91, p < 0.01), significantly predicted the dependent variable beyond that afforded by psychiatric symptoms, namely depression score of the participants.
In the analyses with DEBQ emotional eating sub-scale as the dependent variable, only the UPPS-urgency sub-scale significantly (ΔR 2 = 0.03, ΔF (1,326) = 12.7, p < 0.001) predicted the dependent variable in the negative direction beyond that afforded by psychiatric symptoms.
In regression analyses with DEBQ-external eating subscale as the dependent variable, only the UPPS-urgency   sub-scale was statistically significant (ΔR 2 = 0.01, ΔF (1,359) = 4.8, p < 0.05) predictive power on the dependent variable in the negative direction beyond that afforded by psychiatric symptoms.

Discussion
This study investigated the maladaptive eating patterns and their relationship with psychological constructs, including coping styles, stress intolerance and impulsivity, and psychiatric symptoms in a group of bariatric surgery candidates. As expected and found in the literature, a quarter of the sample had scores implying binge eating, and more than half of the sample had scores pointing to the maladaptive eating behavior from the Bulimic Investigatory Test Edinburgh (BITE). Many bariatric surgery programs require a pre-bariatric evaluation to determine the eligibility of the candidates through evaluating the maladaptive eating behaviors as suggested in the guidelines [48,49]. Among the professionals, there is a concern shared that some patients may engage in "impression management" or "faking good" during preoperative evaluations. This condition means minimizing or hiding symptoms of psychopathology intentionally or nonintentionally to be seen in a favorable light and to avoid rejecting surgery by mental health professionals [50]. Thus, the prevalence of maladaptive eating behaviors and other psychological symptoms in this population could be detected lower than expected [51,52]. Despite this possible bias, we have still found high numbers of candidates with problematic eating behaviors in our sample.
A quarter of the sample had BDI scores of 17 and above, indicating depression; it was also compatible with many studies finding an increased prevalence of depressive disorders in bariatric surgery candidates [53]. Detecting psychopathology also varies based on the methods used and the candidates' willingness to admit that they are experiencing some mental health issues. Bariatric candidates may underreport depression or other psychiatric symptoms to avoid exclusion from the bariatric surgery procedure, as stated before [52]. In our study, the baseline psychiatric interview by the mental health professionals and the psycho-educational approach applied to all the bariatric candidates during the first evaluation might have decreased the defensive attitudes in reporting through assessment scales.
We found positive correlations between depression scores and all the eating test scores, except DEBQ restricting sub-scale, meaning that maladaptive eating behaviors are correlated with depressive symptoms. Similarly, all the eating test scores, except DEBQ restricting sub-scale, were correlated with BSI sub-scale scores of anxiety, negative self-concept, somatization, hostility, and the total BSI score. Depression predicted all pathological eating behaviors except for restrictive eating behavior. Studies disclose that the relationship between obesity and mental disorders is stronger among individuals with more severe obesity; bariatric candidates are known to have a higher prevalence of depressive syndromes [54]. Anxiety also predicted emotional eating and binge eating significantly in our study. Anxiety was less studied than depression and other affective states in obesity and bariatric surgery literature. One of the critical findings in our study suggests that anxiety is a predictive factor in pathological eating behaviors. However, the direction of the relationship between depression and other psychiatric symptoms such as anxiety and maladaptive eating behaviors is not well understood until today. Depression and anxiety may cause maladaptive eating patterns, and pathological eating behavior might trigger depressive symptoms due to obesity and related problems. This uncertainty has inspired us that the long-term psychological constructs, such as stress intolerance, coping styles, and impulsivity inherent in psychopathological conditions, should be investigated beyond psychiatric symptoms. We have investigated these constructs in different analyses, first as independent variables related to the eating patterns and then as possible mediators between psychiatric symptoms and disordered eating behaviors.
Distress intolerance was positively correlated with emotional eating, binge eating, and generally maladaptive eating behaviors in this study. Distress intolerance was a predictive factor in maladaptive eating behaviors in our study, thus being a notable psychological characteristic to be studied in obese individuals. Individuals with poor distress tolerance are characterized by a low threshold for negative emotional experiences and may have difficulty regulating emotions, leading to the inability to cope with emotions healthily [55]. In a longitudinal study, the combined effect of high internalizing symptoms and high distress intolerance was the early risk factor for the future development of eating disorder attitudes [56]. In another longitudinal study evaluating pre and post-bariatric surgery individuals, low distress tolerance was associated with disordered eating behaviors and high BMI, and patients with a higher distress tolerance were more likely to undergo bariatric surgery [57].
Moreover, stress intolerance predicted maladaptive eating behavior beyond depression in the analyses, implying that the ability to cope with stress is an independent factor on eating behavior in bariatric candidates. Despite evidence supporting relationships among stress tolerance, coping styles, and pathologic eating behaviors, most of these studies have a cross-sectional design, from which conclusions 1 3 regarding causal relationships cannot be drawn. Some longitudinal research showed significant relationships between stress intolerance and maladaptive eating behavior but an opposite direction for prediction. Rosen et al. (1993) found that eating disorder symptoms in a group of adolescent girls predicted later stress in four months [58]. Similarly, a longitudinal study assessed stress, coping, and eating behavior for six months in a large sample of young women and discovered that disordered eating weakly predicted perceived stress [59]. Despite these findings, these studies still used short periods to assess stress tolerance and coping strategies and were mostly limited to adolescents or young women, and had smaller sample sizes. These studies recruited healthy samples from the community; however, morbidly obese patients and bariatric candidates represent a different population. Hence, caution is needed in generalizing these results to the individuals with obesity seeking help from medical authorities, and longitudinal studies are needed to conclude the causality.
Stress tolerance may be a transient and relatively shorttime construct that could be affected by disordered eating. However, coping styles are long-term constructs used to confront stress throughout a lifetime, making the causal direction from coping styles to maladaptive eating patterns more reasonable. In our study, coping styles were correlated with maladaptive eating behaviors. The self-confident and optimistic approaches reflecting effective coping styles were negatively correlated with BITE scores and DEBQ total and sub-scale scores, except for restricting sub-scale. This finding implies that bariatric candidates using effective coping styles are less likely to engage in maladaptive eating behaviors.
On the contrary, the helpless approach reflecting an ineffective coping style was positively correlated with all the eating tests' scores, except for DEBQ restricting sub-scale scores. It indicates that the surgery candidates using ineffective coping styles are more likely to have maladaptive eating patterns. The literature illustrates similar results for the relationship between coping styles and eating behavior. A group of adolescents with binge eating behavior used avoidant coping strategies more than those without binge eating [60]. Using avoidant coping strategies has also been related to maladaptive eating behaviors in a group of young women [59]. In our study, the helpless approach strongly predicted the general maladaptive eating behavior, as did emotional eating and binge eating in bariatric candidates. A few studies exist in the literature on coping styles and disordered eating among bariatric candidates. In one of these, avoidance coping (an ineffective coping style) was reported to predict disordered eating in bariatric candidates and remained a relevant predictor even in regression models, including depression and anxiety [27]. The same study also reported similar results to our study: Bariatric candidates with depression and anxiety symptoms had more maladaptive eating behaviors. Thus, ineffective coping behavior, in addition to depression and anxiety could be considered as an essential predictor of disordered eating in bariatric candidates with morbid obesity levels. However, our study revealed that coping styles did not predict maladaptive eating behaviors after the effect of depression and anxiety were controlled. We hypothesized that coping styles might be essential mediators for eating behaviors beyond psychiatric symptoms. Nonetheless, our study's hypothesis was not supported, revealing that this area needs further investigation. A longitudinal study following up obese patients after treatment (either with surgery or conventional methods) found that weight gainers reduced their use of problem and emotion-focused coping. In contrast, those who lost weight decreased in emotion-focused coping and had lower distress levels two years after treatment [61]. Follow-up studies of bariatric candidates after surgery would give more information about the relationship of coping characteristics with psychiatric symptoms and their role in eating behavior in future studies.
In this study, the urgency and lack of perseverance subscales of the UPPS Impulsivity Scale were correlated with pathological eating behaviors, implying that as the impulsivity increased, the maladaptive eating behavior also increased in bariatric candidates. The urgency strongly predicted emotional eating, external eating, and binge eating, whereas lack of perseverance strongly predicted general maladaptive eating. Impulsivity was an independent predictor of all maladaptive eating behaviors beyond depression and other relevant psychiatric symptoms in our study, suggesting that impulsivity might mediate between psychiatric symptoms and disordered eating behaviors. Loss of control over eating (dietary disinhibition) played a role in obesity development [62]. Similarities between substance use disorders and binge-eating disorders regarding the loss of control over the consumption of substance or food were hypothesized, and the facets of impulsivity are thought to represent a typical trans-disease process underlying these maladaptive behaviors, contributing to the obesity development [63]. Impulsivity also predicted response to bariatric surgery [64]. Evidently, impulsivity traits are important in obesity development and the success of bariatric surgery and need to be investigated rigorously in future studies.
Our findings suggest that psychiatric symptoms (depression and anxiety) and psychological constructs (impulsivity, stress intolerance, and coping styles) predict maladaptive eating behaviors in extremely obese individuals applying for bariatric surgery. Maladaptive eating patterns play an essential role in the failure to lose weight and regain weight in the post-operative era; these characteristics should be considered in pre-surgery assessments of bariatric candidates. Even though the patients could not consume large amounts of food after the surgery, losing control over the food may evolve into another form, such as grazing, and might negatively impact weight loss management. Therefore, evaluating only the maladaptive eating patterns would not be enough. Finding the predictors and mediators of disordered eating in bariatric candidates is critical in targeting those pathological constructs. Interventions targeting both those psychopathological characteristics and maladaptive eating patterns could be planned in the pre-operation process in relevant patients. Thus, the healthcare professionals could work with these patients to decrease the maladaptive coping styles they use both in the pre-and post-surgery era; these individuals could benefit more from the bariatric interventions in the long term.

Strength and limitations
A few limitations and considerable strengths exist in this study. It is a cross-sectional study. A longitudinal evaluation of the psychological constructs, psychiatric symptoms, and eating behaviors before and after surgery would be better for demonstrating the causal directions between these variables and obesity and the success of bariatric surgery. Another limitation is the lack of a control group; patients applying for conventional treatment of obesity, especially with lower BMI values, would be a suitable comparison group for the variables investigated. Bariatric candidates were a substantially homogeneous group regarding the BMIs and age in our sample, and some vital relationships regarding age, sex, and BMI levels could not be evaluated in this study.
The detailed psychiatric interview executed for all bariatric candidates was the strong part; the individuals diagnosed with discrete eating disorders such as binge-eating disorder or defined as having certain maladaptive eating behaviors were offered a brief intervention, including dietary control and brief cognitive and behavioral therapy. The candidates diagnosed with depression or anxiety disorders during the interview were also treated with medications and brief psychotherapy before being deemed eligible for surgery.

What is already known on this subject?
Disordered eating behaviors and their relationship with psychiatric symptoms, especially depression and anxiety, were studied in obese patients seeking traditional medical treatments and bariatric surgery. Psychiatric symptoms are known to relate to disordered eating in both populations. Maladaptive eating styles and psychological constructs, such as impulsivity, distress intolerance, and coping styles, were also studied in individuals with obesity and were related to each other. However, our study is one of the few focusing on these constructs and their relationship with eating patterns while also controlling for the psychiatric symptoms in bariatric surgery candidates.

What does this study add?
Our findings suggest that the psychological constructs (impulsivity, stress intolerance, or coping styles) predict maladaptive eating behaviors, in addition to the psychiatric symptoms in extremely obese individuals applying for bariatric surgery. Our findings are critical in that, unlike other studies in this field, the psychiatric symptoms of this large sample were also controlled while investigating the relationship between eating behavior and psychological constructs. This control is crucial because psychiatric symptoms might influence all psychological constructs investigated in this study. During the regression analyses, impulsivity and distress intolerance were significantly associated with the eating behaviors over and above the effects of depression and other psychiatric symptoms, which are the most crucial finding of our research. This result shows that evaluation of the psychological constructs in addition to the psychiatric symptoms should be an important part of the pre-surgery psychiatric evaluation of bariatric candidates so that relevant interventions could be developed targeting coping, impulsive behaviors, and stress intolerance to reduce the disordered eating behaviors in the long term.
Author contributions ŞÖEA, MİY, and GZK contributed to the study conception and design. GZK, AA, and ŞÖEA performed material preparation and data collection. JK, SI, MİY, and ŞÖEA conducted data analyses. ŞÖEA wrote the first draft of the manuscript, and all authors contributed and commented on this manuscript. All authors read and approved the final manuscript.
Funding The authors declare that no funds, grants, or other support were received during the preparation of this manuscript.

Data availability
The datasets generated during and analyzed during the current study are available from the corresponding author upon reasonable request.

Conflict of interests
We wish to confirm that the authors have no relevant financial or non-financial interests to disclose.
Ethical approval Study was approved and conducted in accordance with the ethical standards and all the evaluations performed have been approved by the Hacettepe University Ethics Committee for Clinical Studies (Registration number: GO 20/660, 2020/14-20); this study was conducted in accordance with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards.

Informed consent
This study was a retrospective study evaluating the information of the participants by scanning their written and electronic hospital files.