Psychological Variables Associated with Weight Loss in Obese ...

RESEARCH Current Research

Psychological Variables Associated with Weight Loss in Obese Patients Seeking Treatment at Medical Centers RICCARDO DALLE GRAVE, MD; SIMONA CALUGI, DPsy; FRANCESCO CORICA, MD; SILVIA DI DOMIZIO, RD; GIULIO MARCHESINI, MD; QUOVADIS STUDY GROUP

ABSTRACT Background The identification of process and treatment variables associated with successful weight loss could be a pivotal strategy to reduce attrition and improve effectiveness of dietary treatment in obesity and could help find new therapeutic strategies. Objective The aim of study is to identify the psychological predictors of weight loss in patients with obesity compliant to continuous treatment at medical centers. Design Longitudinal observation of a large cohort of obese subjects entering weight-loss programs in the years 20002002. Subjects/setting Five hundred obese patients who completed 12-month weight-loss treatment by Italian medical centers offering different programs (78.8% females; age 46.2⫾10.8 years; body mass index [BMI; calculated as kg/m2] 37.3⫾5.6). Main outcome measured Measurements were obtained at baseline and after a 12-month weight-loss program. Psychological distress, binge eating, body uneasiness, and attitude toward eating were evaluated by self-administered questionnaires (Symptom Check List-90, Binge Eating Scale, Body Uneasiness Test, and Eating Inventory [Dietary Restraint, Disinhibition, and Hunger]), together with BMI changes. Weight-loss expectations and primary motivation for seeking treatment (health or improving appearance) were also recorded.

R. Dalle Grave is head and S. Calugi is doctor of psychology, Department of Eating and Weight Disorder, Villa Garda Hospital, Garda, Italy. F. Corica is associate professor, Department of Internal Medicine, University of Messina, Policlinico Universitario, Messina, Italy. S. Di Domizio is registered dietitian, Unit of Clinical Dietetics, Alma Mater Studiorum, University, Policlinico S. Orsola, Bologna, Italy. Address correspondence to: Giulio Marchesini, MD, Unit of Clinical Dietetics, Alma Mater University of Bologna, Policlinico S. Orsola, Via Massarenti, 9, I-40138 Bologna, Italy. E-mail: [email protected] Manuscript accepted: June 19, 2009. Copyright © 2009 by the American Dietetic Association. 0002-8223/09/10912-0004$36.00/0 doi: 10.1016/j.jada.2009.09.011

2010

Journal of the AMERICAN DIETETIC ASSOCIATION

Results At follow-up, mean percent weight loss was similar in males and females. Both hierarchical regression and logistic regression analysis revealed that increased dietary restraint and decreased disinhibition were the only independent psychological predictors of BMI change after controlling for age, sex, and baseline BMI (5% weight loss at 12 months: Eating Inventory Restraint (odds ratio [OR]: 1.15; 95% confidence interval [CI]: 1.09 to 1.21) and Disinhibition (OR: 0.92; 95% CI: 0.85 to 0.99); 10% weight loss: Restraint (OR: 1.11; 95% CI: 1.06 to 1.16) and Disinhibition (OR: 0.91; 95% CI: 0.85 to 0.98). Adjustment for centers did not change the results. Conclusion Successful weight loss was associated with increased dietary restraint and reduced disinhibition in obese patients seeking weight-loss treatment in several medical centers throughout Italy. J Am Diet Assoc. 2009;109:2010-2016.

T

he identification of factors associated with successful weight loss is an important research area to improve the outcomes of obesity management. A close matching between treatment features and patients’ pretreatment needs could be a pivotal strategy to reduce treatment attrition and to improve the effectiveness of therapy. Process and treatment variables associated with weight loss could help find new therapeutic strategies to improve outcomes. In the past 30 years, several research trials have investigated the baseline predictors of successful obesity treatment and the results, often inconsistent, have been summarized in review articles (1,2) and books (3,4). Among patients’ pretreatment variables, a very recent review found that little previous dieting, few weight-loss attempts, self-motivation, general efficacy, and autonomy were the best prospective predictors of successful weight management (5). Initial body mass index (BMI; calculated as kg/m2) was associated with weight loss only in studies that included participants with severe obesity (ie, mean BMI ⬎35), whereas the effects of body-image dissatisfaction, low self-esteem, unrealistic weight-loss expectations, and external locus of control remain controversial (5). Other psychological factors, such as binge eating, depression, eating disinhibition (ie, tendency to lose control, overeating when hungry or when exposed to external stimuli or negative mood states), dietary restraint (ie, perceived cognitive ability to restrain food intake in order to modify body weight and shape), perceived hunger, personality, general cognitive style, and

© 2009 by the American Dietetic Association

perceived social support, did not systematically predict the amount of weight loss (5). The most recent studies confirmed the predictive role of self-efficacy (6,7) and baseline BMI in severely obese patients (8), but not in overweight individuals (9). Finally, no association has been observed between pretreatment weight-loss expectations and end-of-program weight loss (10-14), but the attrition rate was higher in patients with higher than expected 1-year BMI loss (15). Several process variables were positively associated with the amount of weight loss, including an increase in dietary restraint (16), early weight loss, and continuous attendance to controls (3,17), higher frequency of selfmonitoring (18-20), and consistent self-monitoring of weight (21). Finally, treatment factors affecting weight loss include an increased duration of treatment (22), the prescription of higher intensity physical activity (23), the provision of food (24), and the association of behavior therapy with weight-loss drugs (20). Unfortunately, most data on the psychosocial factors associated with weight loss come from studies including mainly overweight women or obese women in the BMI range of 30 to 35, or are generated in research settings and cannot be generalized to the heterogeneous “real world” of the medical centers treating obesity. Baseline and process-related psychological variables (eg, psychological distress, body uneasiness, binge eating, dietary restraint, disinhibition, and weight-loss expectations, as well as their changes at 1-year follow-up) are expected to play a major role in weight-loss outcomes in subjects with obesity. The aim of the present analysis was to analyze the role of psychosocial factors associated with weight loss in male and female patients with obesity, entering weight-loss programs offering prescriptive diet counseling (with or without additional antiobesity drug therapy) or cognitive-behavioral treatment in a large multisite Italian observational study (25). In the absence of a robust theory supported by empirical findings on the psychological factors associated with weight loss, the study included several psychological variables potentially relevant in the care of obese patients tested in previous research studies (5,26-28). METHODS QUOVADIS Study Planning and Protocol The Quality of Life in Obesity: Evaluation and Disease Surveillance (QUOVADIS) study planning and protocol were described in detail in a previous article (25). The QUOVADIS study is an observational analysis of quality of life, psychological distress, and attitude toward eating in obese patients seeking weight-loss treatment at obesity medical centers accredited by the Italian Health Service. The centers were located in urban areas throughout Italy. The study design was observational and the centers treated patients according to their slightly different specific protocols, including dieting, cognitive behavioral therapy, and drugs. All obese subjects (BMI ⱖ30) seeking treatment at participating medical centers in the years 2000 to 2002 were eligible for the QUOVADIS study, provided they were not on active treatment at the time of enrollment, were in the age range between 25 and 65 years, and

agreed to fill in a package of self-administered questionnaires. All evaluations were carried out at baseline, approximately 1 week before the beginning of treatment, and were repeated after 12 months in 18 centers. Subjects entering protocols of bariatric surgery (⬍2% of total patients) were excluded. In the course of the study, a large dropout rate was observed and the predictors of attrition have already been reported (15,29). The present analysis is based on the longitudinal data of subjects on continuous care at 12 months (500 cases, 21.2% males and 78.8% female) (Table 1). A post hoc analysis showed only minor differences between this selected population and the general QUOVADIS cohort (see Results: Baseline Characteristics). All data were stored in a large database, accessed by each center through an extranet system and electronic forms. The general QUOVADIS protocol was approved by the committees of the various centers, after approval by the ethical committee of the coordinating center (Azienda Ospedaliera di Bologna, Policlinico S. Orsola-Malpighi). All participants gave written informed consent for participation. Measures Case Report Form. Physicians filled in the case report form at the time of enrollment by directly interviewing patients. It included demographic and weight data, a detailed diet history, expected 1-year weight loss, maximum acceptable weight, and desired weight. Expected 1-year loss was defined as “the amount of weight that patients were expecting to lose with treatment in the following 12 months.” To help subjects quantify their expectations, this value was categorized in multiples of 10 kg. Maximum acceptable weight was defined as “the heaviest body weight that patients could accept and tolerate to reach after treatment,” whereas desired weight was defined as “the body weight that they were hoping to achieve with treatment, however unrealistic it was.” The case report form also included a question about reasons for seeking treatment. For this specific purpose, patients were asked to choose the main reason for entering a weight-loss program among three different answers: improving appearance, improving future health, or improving present health. Psychosocial Measures. At baseline and at follow-up, participants completed a battery of questionnaires measuring psychological distress, binge eating, body uneasiness, and attitudes toward eating. The Symptom Checklist-90R (SCL) (30) was used to identify psychological distress. For each item, patients scored how much that problem had distressed them during the previous week, with responses ranging from 0 (not at all) to 4 (extremely). The 90 items of the test were used to compute the Global Severity Index, an indicator of overall psychological distress (30). A value ⱖ1 in SCLGlobal Severity Index or in a specific subscale is suggestive of psychopathology (1.00 to 1.49, mild; 1.50 to 1.99, moderate; ⱖ2.00, severe). The internal consistency coefficients ␣ for the nine dimensions of SCL ranged from .77 to .90 (30). The Binge Eating Scale (31) was used to measure the severity of binge eating. It examines in 16 items, both behavioral signs (eating large amounts of food) and feel-

December 2009 ● Journal of the AMERICAN DIETETIC ASSOCIATION

2011

Table 1. Demographic and clinical variables of males and females in the 500 participants of the QUOVADISa study available at the 12-month follow-up

Age (y) BMIc Historical variables BMI at age 20 Maximum BMI Age at first dieting (y) Maximum weight loss (%) Weight-loss expectations Maximum acceptable BMI Dream BMI Expected 1-year BMI loss Psychopathology scores Binge Eating Scale Symptom Checklist 90 (Global Severity Index) Body Uneasiness Test (Global Severity Index) Eating Inventory Dietary Restraint Disinhibition Hunger

Malesb (nⴝ106)

Femalesb (nⴝ394)

t Test or Mann-Whitney test

47.1⫾10.8 37.2⫾6.9

46.2⫾10.8 37.3⫾5.6

⫺0.83 ⫺0.67

0.406 0.500

26.5⫾4.8 38.5⫾7.4 32.7⫾10.6 16.9⫾10.3

24.9⫾4.3 38.5⫾6.3 28.0⫾11.2 16.2⫾9.1

⫺3.51 ⫺0.19 3.28 0.63

0.001 0.847 0.001 0.534

29.8⫾3.5 27.1⫾2.7 8.6⫾3.4

28.7⫾3.4 25.8⫾3.0 10.0⫾3.7

⫺2.89 3.99 ⫺3.51

0.004 ⬍0.001 ⬍0.001

10.0⫾7.5 0.5⫾0.4 1.0⫾0.8

15.0⫾9.6 0.8⫾0.6 1.6⫾1.0

⫺4.87 ⫺4.45 ⫺6.20

⬍0.001 ⬍0.001 ⬍0.001

8.2⫾3.7 6.8⫾3.2 5.2⫾3.6

8.9⫾4.0 8.6⫾3.4 6.3⫾3.7

⫺1.70 ⫺4.84 ⫺2.83

0.089 ⬍0.001 0.005

P value

a

QUOVADIS⫽Quality of Life in Obesity: Evaluation and Disease Surveillance. Mean⫾standard deviation. c BMI⫽body mass index; calculated as kg/m2. NOTE: Information from this table is available online at www.adajournal.org as part of a PowerPoint presentation. b

ings or cognition during a binge episode (loss of control, guilt, fear of being unable to stop eating). Scores ⱖ27 have conventionally been considered as indicative of severe binge eating and ⱕ16 as identifying mild or no binge eating (32). The scale has adequate internal consistency and validity (31), good test–retest reliability (r⫽0.87; P⬍0.001), and moderate associations with binge-eating severity as measured by food records (r⫽0.20 to 0.40; P⬍0.05) (33). The Body Uneasiness Test (34) was used to evaluate body uneasiness. The term body uneasiness was used to describe not only body dissatisfaction but also body-associated emotions, such as anxiety, alarm, trepidation, worry, mistrust, misgiving, doubt, suspicion, and embarrassment. The Body Uneasiness Test consists of two parts: Body Uneasiness Test-A comprises 34 items with a score ranging from 0 (never) to 5 (always). Scores of the 34 items are used to compute the Global Severity Index and Body Uneasiness Test-B has 37 items that look at specific worries about particular body parts or functions. For the aim of the present study, only part A was used because Body Uneasiness Test-B also evaluates worries on specific body parts (eg, mouth, moustaches, and skin) that are not modifiable with weight loss. The Body Uneasiness Test has been validated in patients with obesity and showed good internal consistency (Cronbach’s ␣ coefficient ⬎.7) (35). The Three-Factor Eating Questionnaire (36), later renamed the Eating Inventory (37), was used to evaluate attitudes toward eating. The Eating Inventory has three scales: Dietary (Cognitive) Restraint, Disinhibition, and

2012

December 2009 Volume 109 Number 12

Hunger. Dietary Restraint measures the cognitive ability to restrain food intake. Disinhibition refers to one’s tendency to lose control over eating. Hunger refers to the perception of hunger/satiety. The Eating Inventory has good internal consistency (⬎.80) (36) and the reliability and validity of its scales have been established in many studies (38). The four questionnaires have been validated in their Italian versions by the NetWorking team Group of the Italian Society for Eating Behavior Disorders or by authors (35). Weight and Height. Weight and height were measured at each medical center by trained research staff in patients with underwear without shoes by mechanical column medical scales (Seca Model 709, Hamburg, Germany; capacity 200 kg, graduation 100 g in the majority of centers) and by stadiometers. Weight changes were available at 12 months only in patients compliant to continuous treatment. Statistical Analyses All weight data (in kg) were transformed into BMI units to improve comparison between sexes. A first descriptive analysis was used to obtain a qualitative evaluation of clinical data, response to questionnaires and patients’ outcomes. Changes in clinical parameters at 12 months were tested for significance by means of t test for paired data or by ranks of Wilcoxon test for nonnormally distributed data (ie, age, BMI, BMI at age 20, maximum BMI, maximum weight loss, maximum acceptable BMI, SCL90, Body Uneasiness Test, Eating Inventory subscales), and were transformed into effect sizes for the assessment of their magnitude (39,40).

Table 2. Body mass index and psychosocial measures at baseline and at 12 months in the 500 participants of the QUOVADISa study available at follow-upb

Body mass index Body Uneasiness Test-Global Severity Index Symptom CheckList-90⫺ Global Severity Index Binge Eating Scale Eating Inventory Dietary Restraint Disinhibition Hunger

Sex

Baselinec

12-monthc

Effect sized

Fe Mf

37.3⫾5.6 37.2⫾6.9

34.4⫾5.9 33.8⫾6.2

0.66 0.70

17.3 10.0

⬍0.001 ⬍0.001

F M

1.59⫾0.99 1.35⫾1.03 0.97⫾0.84 0.72⫾0.69

0.27 0.36

5.6 3.9

⬍0.001 ⬍0.001

F M F M

0.82⫾0.60 0.71⫾0.59 0.54⫾0.41 0.46⫾0.43 14.9⫾9.7 11.2⫾8.6 10.1⫾7.5 8.1⫾6.9

0.25 0.21 0.43 0.32

5.1 2.2 9.3 3.5

⬍0.001 0.031 ⬍0.001 0.001

0.52 0.53 0.34 0.29 0.33 0.37

⫺12.0 ⫺6.4 7.2 3.1 6.8 4.1

⬍0.001 ⬍0.001 ⬍0.001 0.003 ⬍0.001 ⬍0.001

F M F M F M

8.9⫾4.0 8.2⫾3.7 8.6⫾3.4 6.8⫾3.2 6.3⫾3.7 5.2⫾3.5

11.7⫾4.2 11.1⫾4.5 7.5⫾3.6 5.9⫾3.5 5.2⫾3.4 4.1⫾3.3

t or z

P value

a

QUOVADIS⫽Quality of Life in Obesity: Evaluation and Disease Surveillance. Paired t test or Wilcoxon test and P values of differences, as well as effect sizes, are reported. c Values are mean⫾standard deviation. d Effect size: around 0.20, small; around 0.50, moderate; around 0.80, large (39,40). e F⫽female. f M⫽male. NOTE: Information from this table is available online at www.adajournal.org as part of a PowerPoint presentation. b

Spearman’s ␳ correlation was used to evaluate the linear relationship between predictors and weight change. A few independent variables assessed at baseline displayed a nonnormal distribution, warranting the use of this nonparametric technique. Also, partial correlation controlling for baseline BMI was carried out. Forward stepwise logistic linear regression analysis was used to identify the determinants of weight loss (ⱖ5% and ⱖ10%) at 12-month follow-up. Variables used in the three models were the baseline clinical parameters and their changes at 12 months (SCL-90, Body Uneasiness Test, Binge Eating Scale and Eating Inventory subscales). Age, sex, and initial BMI were forced into the models to adjust for their effects. Data are reported as mean⫾standard deviation or percentage. All analyses were performed using SPSS for Windows, Version 15.0, 2006 (SPSS Inc, Chicago, IL). RESULTS Baseline Characteristics Baseline characteristics of the whole QUOVADIS sample were described in detail in previous reports (25). A post hoc analysis showed that the group available at follow-up (500 of the 1,530 patients enrolled in the QUOVADIS study at the 18 centers) was not different from the total entry population in sex distribution and BMI, with minor differences in age (46.4⫾10.8 years vs 44.6⫾11.0 in total sample; P⫽0.001). In subjects available at follow-up (Table 1), no significant differences were present between males and females

in age, BMI, and maximum BMI. Males had a significantly higher BMI at age 20 (P⫽0.001), whereas women reported an earlier age at first dieting (P⫽0.001), a lower maximum acceptable and desired BMI (P⫽0.007 and P⬍0.001, respectively), and higher weight-loss expectations (P⬍0.001). Females had higher scores than males in most psychological variables, with the notable exception of dietary restraint (Table 1). 12-Month Changes By the end of the follow-up period, a substantial reduction in BMI, general psychopathology (SCL-Global Severity Index), binge eating (Binge Eating Scale), body image dissatisfaction (Body Uneasiness Test-Global Severity Index) and some scores of the eating inventory (Eating Inventory Disinhibition and Hunger) were observed in both females and males (Table 2), whereas the Eating Inventory Dietary Restraint increased considerably. Females had a higher reduction of Binge Eating Scale scores than males [t(490)⫽⫺2.46; P⫽0.015), but no systematic sex-related differences were observed in the other psychosocial variables and in weight-loss percentage. Baseline and historical variables weakly correlated with BMI changes; the most significant correlation was found between BMI changes and changes in Eating Inventory Dietary Restraint, before and after adjusting for baseline BMI (␳⫽.29; P⬍0.001; r⫽0.28; P⬍0.001; respectively). In particular, considering the correlation coefficients ⬎.20, a larger weight loss was associated with a larger improvement in SCL-90 and in Eating Inventory

December 2009 ● Journal of the AMERICAN DIETETIC ASSOCIATION

2013

Table 3. Correlation matrix between changes in body mass index during the study and clinical and psychological variables at baseline as well as their changes during 12 monthsa in 500 participants of the QUOVADISb study available at follow-up Changes in BMIc (nⴝ500) Unadjusted for Adjusted for basal BMI basal BMI Age (y) BMI BMI at age 20 y Body Uneasiness Test⫺Global Severity Index Symptom Checklist-90⫺Global Severity Index Binge Eating Scale Eating Inventory Dietary Restraint Eating Inventory Disinhibition Eating Inventory Hunger Changes in Body Uneasiness Test⫺Global Severity Index Changes in Symptom Checklist90⫺Global Severity Index Changes in Binge Eating Scale Changes in Eating Inventory Dietary Restraint Changes in Eating Inventory Disinhibition Changes in Eating Inventory Hunger

⫺0.13** 0.18** 0.06 0.00

⫺0.21** — 0.06 ⫺0.03

⫺0.11* ⫺0.12** ⫺0.02 ⫺0.07 ⫺0.06

⫺0.15** ⫺0.14* 0.02 ⫺0.10 ⫺0.09

0.12*

0.12*

0.15** 0.17**

0.21** 0.19**

0.29**

0.28**

⫺0.19**

⫺0.24**

⫺0.20**

⫺0.27**

a A positive coefficient indicates that more weight loss is associated with higher values of baseline clinical variables or a larger decrease of clinical variables during treatment; a negative coefficient indicates that more weight loss is associated with lower baseline values or lower changes during follow-up. b QUOVADIS⫽Quality of Life in Obesity: Evaluation and Disease Surveillance. c BMI⫽body mass index (calculated as kg/m2). *P⬍0.05. **P⬍0.01. NOTE: Information from this table is available online at www.adajournal.org as part of a PowerPoint presentation.

Dietary Restraint, a younger age and a lower increase of Eating Inventory Disinhibition and Hunger, adjusted for basal BMI (Table 3). Hierarchical regression analyses revealed that the factors more closely associated with BMI change at 12 months were changes in Eating Inventory Dietary Restraint (␤⫽.200; t⫽4.44; P⬍0.001) and changes in disinhibition (␤⫽⫺.133; t⫽⫺2.94; P⫽0.003), after controlling for age, sex, and baseline BMI. Logistic regression analysis showed that the probability to achieve a 5% weight loss (310 of 500 cases, 63.7%) significantly increased for any point increase in Eating Inventory Dietary Restraint (odds ratio [OR]: 1.15; 95% confidence interval [CI]: 1.09 to 1.21) and decrease in disinhibition (OR: 0.92; 95% CI: 0.85 to 0.99) (adjusted for age, sex, and BMI). A weight loss exceeding 10% was observed in 34.4% (n⫽172) of participants and was similarly predicted by changes in Eating Inventory Dietary Restraint (OR: 1.11; 95% CI: 1.06 to 1.16) and disinhibition (OR: 0.91; 95% CI: 0.85 to 0.98). Further adjustment for center did not change the results.

2014

December 2009 Volume 109 Number 12

DISCUSSION The study findings suggest that changes in dietary restraint and disinhibition on Eating Inventory are the factors most closely associated with BMI changes after 12 months of continuous treatment in obese patients attending weight-loss programs. In particular, an increase in dietary restraint was the principal psychological factor associated with weight loss in this setting. Dietary restraint is defined as the intention to restrict food intake in order to control body weight and shape (41). The original hypothesis considered dietary restraint as the main psychological determinant of overeating (42). This hypothesis was supported by laboratory studies in normalweight participants showing that subjects with high dietary restraint tend to overeat in response to a preload (43-46). However, other studies did not find any relationship between dietary restraint and overeating (47-50), and the dietary restraint hypothesis was also challenged by clinical studies showing less weight regain in obese participants with an increase in dietary restraint, both at medium- (51) and long-term analysis (52). In agreement with these observations, the QUOVADIS data confirm that the baseline score of Eating Inventory Dietary Restraint is not associated with weight loss (5). However, a positive correlation exists between the increase in Eating Inventory Dietary Restraint and the amount of weight loss in subjects who completed 12 months of treatment (27). BMI changes were also associated with a decrease in the Eating Inventory Disinhibition subscale. The term disinhibition may be misleading; it implies the disruption of an inhibition state, whereas loss of control over eating may occur also in people with low restraint scores in response to emotional factors or hunger (53). Previous studies found little or no association between treatmentinduced weight changes and baseline disinhibition (54), but higher levels of disinhibition were associated with higher risk of weight regain in a few studies (55,56), but not in others (27,54). In agreement with the present results, a recent study showed that a decrease in the disinhibition score of the Eating Inventory was a factor significantly associated with weight loss (8). The study findings may be influenced by a cause-andeffect association. Calorie intake was not systematically assessed at either entry into the program or at follow-up. The most likely reason for the association is that the increased dietary restraint and the reduced disinhibition achieved by attendance to treatment sessions might have led to weight loss via decreased calorie intake. This hypothesis, however, should be confirmed in future studies. Pretreatment values and treatment-induced changes in binge eating, psychological distress, and body image were not associated with successful weight loss, and also weightloss expectations did not influence the degree of weight loss in subjects on continuous treatment. These results are consistent with other weight-loss studies (5,10,11,13,14,57). However, the finding that changes in psychopathology and binge eating are not associated with BMI outcome among dieters enhances the scientific knowledge regarding weightloss interventions. Higher weight-loss expectations were also unrelated to weight regain (57), but were reported to play a crucial role in favoring attrition in patients seeking treatment in a medical setting (15). The study has two main strengths. Its observational de-

sign allows a comprehensive analysis of the effect of several psychological factors associated with weight loss in the “real world” of medical obesity centers, with heterogeneous protocols of care. It included a large sample of obese males, a group rarely considered in obesity treatment. The most critical limitation of this study is that only a subgroup of the homogeneous QUOVADIS participants, mainly including white females, was available at follow-up. Consequently, the results cannot be generalized to the whole population of obese patients enrolled by medical centers. In particular, it remains to be determined whether the increase of dietary restraint closely associated with weight loss in completers might have had negative consequences in the subgroup of participants who failed to complete the 1-year assessment (eg, favoring disinhibition). However, a deleterious effect of dieting in specific Eating Inventory categories is not likely because no differences in baseline Eating Inventory scores were demonstrated between subjects on continuous treatment and those who dropped out. Secondly, both attrition rates and weight loss were significantly different among centers. The statistical significance was maintained after controlling for center variability, potentially excluding a specific effect of individual treatment strategies, but external validation is needed. The study design also did not consider the potential effect of other possible factors (eg, quality of the weight-loss treatment). Finally, the study did not collect data on the large group of dieters who successfully achieve weight loss without professional help. The high dropout rate observed in the QUOVADIS study (33%) was already examined (15,29), but deserves a specific comment. Young age and unrealistic weight-loss expectations were previously identified as baseline predictors of dropout. In addition, practical difficulties (eg, family problems, problems at work, logistics, health problems), reported by more than half of dropouts (55%), were the leading causes of attrition identified by a structured telephone interview, together with perceived treatment failure and some psychological processes intervening during treatment (eg, lack of motivation, confidence to lose additional weight without professional help, and sense of abandonment by the therapists) (58). Interestingly, attrition was not associated with baseline Restraint and/or Disinhibition scores. The much lower attrition rate reported in most clinical trials may be explained by several factors. In clinical trials patients receive treatment for free, whereas in the QUOVADIS study, where patients have to pay for treatment, they are much more likely to drop out if the results are unsatisfactory. Two thirds of study patients had moderate-to-severe obesity (class II and III obesity) and medical comorbidities, which represent exclusion criteria in most clinical trials and are common reasons for dropout. Finally, participating centers were located in general hospitals of urban areas; most patients reported transportation problems and the difficulty attending treatment sessions during working hours as principal reasons for attrition. In general, the QUOVADIS study has useful implications for clinicians in the treatment of obesity. The pretreatment psychological examination appears to be of limited importance, once contraindications to weight-loss treatment are ruled out (eg, major depression and bulimia nervosa), because it does not provide useful data predicting weight loss

(59). Conducting a shorter psychological assessment during the enrollment process saves time, which may be used to reinforce patients’ motivation. Cognitive strategies (eg, setting realistic weight goals, addressing dysfunctional thinking that hinders weight loss) might be warranted to help patients develop a weight-control mindset characterized by healthful dietary restraint and low levels of disinhibition. Finally, attrition could be also limited by clinicians being more responsive to patients’ unrealistic expectations, practical and psychological issues, therapist “burnout,” and perception of abandonment by health care providers during long-term follow-up (58). CONCLUSION In conclusion, the findings of this study suggest that changes in cognitive restraint and disinhibition are correlated with successful weight loss in a large sample of obese patients treated at medical centers in Italy. Food and nutrition practitioners in the United States as well as in other countries should carefully consider restraint and disinhibition as pivotal factors of patients’ weight-management programs. Future studies are needed to evaluate the potential value of psychological assessments and cognitive strategies associated with fee-for-service weight-loss programs to maximize success rates for medically supervised obesity programs. STATEMENT OF POTENTIAL CONFLICT OF INTEREST: No potential conflict of interest was reported by the authors. FUNDING/SUPPORT: The QUOVADIS study was supported by an unrestricted grant from BRACCO Imaging Spa, Milan, Italy. References 1. Brownell KD. Behavioral, psychological, and environmental predictors of obesity and success at weight reduction. Int J Obes. 1984;8: 543-550. 2. Foreyt JP, Goodrick GK. Factors common to successful therapy for the obese patient. Med Sci Sports Exerc. 1991;23:292-297. 3. Wadden TA, Letizia KA. Predictors of attrition in weight loss in patients treated by moderate and severe caloric restriction. In: Wadden TA, VanItallie TB, eds. Treatment of the Seriously Obese Patients. New York, NY: The Guilford Press; 1992:383-410. 4. Wilson GT. Behavioral and psychological predictors of treatment outcome in obesity. In: Allison DB, Pi-Sunyer FX, eds. Obesity Treatment: Establishing Goals, Improving Outcomes, and Reviewing the Research Agenda. New York, NY: Plenum Press; 1995:183-189. 5. Teixeira PJ, Going SB, Sardinha LB, Lohman TG. A review of psychosocial pre-treatment predictors of weight control. Obes Rev. 2005; 6:43-65. 6. Wamsteker EW, Geenen R, Iestra J, Larsen JK, Zelissen PM, van Staveren WA. Obesity-related beliefs predict weight loss after an 8-week low-calorie diet. J Am Diet Assoc. 2005;105:441-444. 7. Wiltink J, Dippel A, Szczepanski M, Thiede R, Alt C, Beutel ME. Long-term weight loss maintenance after inpatient psychotherapy of severely obese patients based on a randomized study: Predictors and maintaining factors of health behavior. J Psychosom Res. 2007;62: 691-698. 8. Hainer V, Kunesova M, Bellisle F, Hill M, Braunerova R, Wagenknecht M. Psychobehavioral and nutritional predictors of weight loss in obese women treated with sibutramine. Int J Obes (Lond). 2005; 29:208-216. 9. Collings AS, Saules KK, Saad LR. A prospective study of predictors of successful weight maintenance by women enrolled in communitybased weight-loss programs. Eat Weight Disord. 2008;13:38-47. 10. Linde JA, Jeffery RW, Finch EA, Ng DM, Rothman AJ. Are unrealistic weight loss goals associated with outcomes for overweight women? Obes Res. 2004;12:569-576.

December 2009 ● Journal of the AMERICAN DIETETIC ASSOCIATION

2015

11. Linde JA, Jeffery RW, Levy RL, Pronk NP, Boyle RG. Weight loss goals and treatment outcomes among overweight men and women enrolled in a weight loss trial. Int J Obes (Lond). 2005;29:1002-1005. 12. Fabricatore AN, Wadden TA, Womble LG, Sarwer DB, Berkowitz RI, Foster GD, Brock JR. The role of patients’ expectations and goals in the behavioral and pharmacological treatment of obesity. Int J Obes (Lond). 2007;31:1739-1745. 13. Gorin AA, Marinilli Pinto A, Tate DF, Raynor HA, Fava JL, Wing RR. Failure to meet weight loss expectations does not impact maintenance in successful weight losers. Obesity. 2007;15:3086-3090. 14. White MA, Masheb RM, Rothschild BS, Burke-Martindale CH, Grilo CM. Do patients’ unrealistic weight goals have prognostic significance for bariatric surgery? Obes Surg. 2007;17:74-81. 15. Dalle Grave R, Calugi S, Molinari E, Petroni ML, Bondi M, Compare A, Marchesini G. Weight loss expectations in obese patients and treatment attrition: An observational multicenter study. Obes Res. 2005;13:1961-1969. 16. Vogels N, Diepvens K, Westerterp-Plantenga MS. Predictors of longterm weight maintenance. Obes Res. 2005;13:2162-2168. 17. Wing RR, Phelan S. Behavioral treatment of obesity: Strategies to improve outcome and predictors of success. In: Eckel RH, ed. Obesity: Mechanisms and Clinical Management. Philadelphia: Lippincott, Williams & Wilkins; 2003:415-435. 18. Baker RC, Kirschenbaum DS. Self-monitoring may be necessary for successful weight control. Behav Ther. 1993;24:377-394. 19. Boutelle KN, Kirschenbaum DS. Further support for consistent selfmonitoring as a vital component of successful weight control. Obes Res. 1998;6:219-224. 20. Wadden TA, Berkowitz RI, Womble LG, Sarwer DB, Phelan S, Cato RK, Hesson LA, Osei SY, Kaplan R, Stunkard AJ. Randomized trial of lifestyle modification and pharmacotherapy for obesity. N Engl J Med. 2005;353:2111-2120. 21. Butryn ML, Phelan S, Hill JO, Wing RR. Consistent self-monitoring of weight: A key component of successful weight loss maintenance. Obesity. 2007;15:3091-3096. 22. Perri MG, Sears SF Jr, Clark JE. Strategies for improving maintenance of weight loss. Toward a continuous care model of obesity management. Diabetes Care. 1993;16:200-209. 23. Jeffery RW, Wing RR, Sherwood NE, Tate DF. Physical activity and weight loss: Does prescribing higher physical activity goals improve outcome? Am J Clin Nutr. 2003;78:684-689. 24. Jeffery RW, Wing RR. Long-term effects of interventions for weight loss using food provision and monetary incentives. J Consult Clin Psychol. 1995;63:793-796. 25. Melchionda N, Marchesini G, Apolone G, Cuzzolaro M, Mannucci E, Grossi E, the QUOVADIS Study Group. The QUOVADIS study. Features of obese Italian patients seeking treatment at specialist centers. Diabetes Nutr Metab. 2003;16:115-124. 26. Foster GD, Wadden TA, Vogt RA. Body image in obese women before, during, and after weight loss treatment. Health Psychol. 1997;16:226229. 27. Foster GD, Wadden TA, Swain RM, Stunkard AJ, Platte P, Vogt RA. The Eating Inventory in obese women: Clinical correlates and relationship to weight loss. Int J Obes Relat Metab Disord. 1998;22:778785. 28. Sherwood NE, Jeffery RW, Wing RR. Binge status as a predictor of weight loss treatment outcome. Int J Obes Relat Metab Disord. 1999; 23:485-493. 29. Dalle Grave R, Melchionda N, Calugi S, Centis E, Tufano A, Fatati G, Fusco MA, Marchesini G. Continuous care in the treatment of obesity. An observational multicentre study. J Intern Med. 2005;258:265-273. 30. Derogatis LR, Cleary PA. Confirmation of the dimensional structure of the SCL-90: A study in construct validity. J Clin Psychol. 1977;33: 981-989. 31. Gormally J, Block S, Daston S, Rardin D. The assessment of binge eating severity among obese persons. Addict Behav. 1982;7:47-55. 32. Greeno CG, Marcus MD, Wing RR. Diagnosis of binge eating disorder: Discrepancies between a questionnaire and clinical interview. Int J Eat Disord. 1995;17:153-160. 33. Timmerman GM. Binge Eating Scale: Further assessment of validity and reliability. J Appl Biobehav Res. 1999;4:1-12. 34. Cuzzolaro M, Vetrone G, Marano G, Garfinkel PE. The Body Uneasiness Test (BUT): Development and validation of a new body image assessment scale. Eat Weight Disord. 2006;11:1-13. 35. Marano G, Cuzzolaro M, Vetrone G, Calugi S, Dalle Grave R, Marchesini G, Garfinkel PE, and the QUOVADIS Study Group. Further validation study of the Body Uneasiness Test (BUT) in a clinical sample of 1922 adult obese subjects. Eat Weight Disord. 2007; 12:70-82.

2016

December 2009 Volume 109 Number 12

36. Stunkard AJ, Messick S. The Three-Factor Eating Questionnaire to measure dietary restraint, disinhibition and hunger. J Psychosom Res. 1985;29:71-83. 37. Stunkard AJ, Messick S. The Eating Inventory. San Antonio, TX: Psychological Coporation; 1988. 38. Gorman BS, Allison DB. Measures of restrained eating. In: Allison DB, ed. Handbook of Assessment for Eating Behaviors and WeightRelated Problems. Thousands Oaks, CA: Sage Publishing, Inc.; 1995: 149-184. 39. Cohen J. Statistical Power Analysis for the Behavioural Sciences. New York, NY: Academic Press; 1977. 40. Kazis LE, Anderson JJ, Meenan RF. Effect sizes for interpreting changes in health status. Med Care. 1989;27(suppl):S178-S189. 41. Herman CP, Mack D. Restrained and unrestrained eating. J Personality. 1975;43:647-660. 42. Polivy J, Herman CP. Dieting and binging. A causal analysis. Am Psychol. 1985;40:193-201. 43. Ruderman AJ. Dietary restraint: A theoretical and empirical review. Psychol Bulletin. 1986;99:247-262. 44. Heatherton TF, Herman CP, Polivy J, King GA, McGree ST. The (mis)measurement of restraint: An analysis of conceptual and psychometric issues. J Abnorm Psychol. 1988;97:19-28. 45. Polivy J. Psychological consequences of food restriction. J Am Diet Assoc. 1996;96:589-592. 46. Lowe MR, Levine AS. Eating motives and the controversy over dieting: Eating less than needed versus less than wanted. Obes Res. 2005;13:797-806. 47. Wardle J. Compulsive eating and dietary restraint. Br J Clin Psychol. 1987;26(Pt 1):47-55. 48. Lowe MR, Kleifield EI. Cognitive restraint, weight suppression, and the regulation of eating. Appetite. 1988;10:159-168. 49. Smith CF, Geiselman PJ, Williamson DA, Champagne CM, Bray GA, Ryan DH. Association of dietary restraint and disinhibition with eating behavior, body mass, and hunger. Eat Weight Disord. 1998;3: 7-15. 50. Dritschel B, Cooper PJ, Charnock D. A problematic counter-regulation experiment: Implications for the link between dietary restraint and overeating. Int J Eat Disord. 1993;13:297-304. 51. Westerterp-Plantenga MS, Kempen KP, Saris WH. Determinants of weight maintenance in women after diet-induced weight reduction. Int J Obes Relat Metab Disord. 1998;22:1-6. 52. Vogels N, Westerterp-Plantenga MS. Successful long-term weight maintenance: A 2-year follow-up. Obesity. 2007;15:1258-1266. 53. Westenhoefer J, Broeckmann P, Munch AK, Pudel V. Cognitive control of eating behaviour and the disinhibition effect. Appetite. 1994; 23:27-41. 54. Teixeira PJ, Going SB, Houtkooper LB, Cussler EC, Metcalfe LL, Blew RM, Sardinha LB, Lohman TG. Pretreatment predictors of attrition and successful weight management in women. Int J Obes Relat Metab Disord. 2004;28:1124-1133. 55. McGuire MT, Wing RR, Klem ML, Lang W, Hill JO. What predicts weight regain in a group of successful weight losers? J Consult Clin Psychol. 1999;67:177-185. 56. Cuntz U, Leibbrand R, Ehrig C, Shaw R, Fichter MM. Predictors of post-treatment weight reduction after in-patient behavioral therapy. Int J Obes Relat Metab Disord. 2001;25(suppl 1):S99-S101. 57. Fabricatore AN. Behavior therapy and cognitive-behavioral therapy of obesity: Is there a difference? J Am Diet Assoc. 2007;107:92-99. 58. Grossi E, Dalle Grave R, Mannucci E, Molinari E, Compare A, Cuzzolaro M, Marchesini G. Complexity of attrition in the treatment of obesity: Clues from a structured telephone interview. Int J Obes (Lond). 2006;30:1132-1137. 59. Clinical guidelines on the identification, evaluation, and treatment of overweight and obesity in adults: Executive summary. Expert Panel on the Identification, Evaluation, and Treatment of Overweight in Adults. Am J Clin Nutr. 1998;68:899-917.